niobures commited on
Commit
2f6b10b
·
verified ·
1 Parent(s): 290ee21
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. .gitattributes +16 -0
  2. .gitignore +32 -0
  3. .pre-commit-config.yaml +40 -0
  4. .readthedocs.yaml +12 -0
  5. LICENSE +395 -0
  6. README.md +5 -0
  7. acoustic/abkhaz/cv/v2.0.0/README.md +110 -0
  8. acoustic/abkhaz/cv/v2.0.0/meta.json +16 -0
  9. acoustic/arabic.zip +3 -0
  10. acoustic/armenian/cv/v2.0.0/README.md +110 -0
  11. acoustic/armenian/cv/v2.0.0/meta.json +16 -0
  12. acoustic/bashkir/cv/v2.0.0/README.md +110 -0
  13. acoustic/bashkir/cv/v2.0.0/meta.json +16 -0
  14. acoustic/basque/cv/v2.0.0/README.md +110 -0
  15. acoustic/basque/cv/v2.0.0/meta.json +16 -0
  16. acoustic/belarusian/cv/v2.0.0/README.md +110 -0
  17. acoustic/belarusian/cv/v2.0.0/meta.json +16 -0
  18. acoustic/bulgarian.zip +3 -0
  19. acoustic/bulgarian/cv/v2.0.0/README.md +110 -0
  20. acoustic/bulgarian/cv/v2.0.0/meta.json +16 -0
  21. acoustic/bulgarian/mfa/v2.0.0/LICENSE +395 -0
  22. acoustic/bulgarian/mfa/v2.0.0/README.md +128 -0
  23. acoustic/bulgarian/mfa/v2.0.0/meta.json +16 -0
  24. acoustic/bulgarian/mfa/v2.0.0a/LICENSE +395 -0
  25. acoustic/bulgarian/mfa/v2.0.0a/README.md +128 -0
  26. acoustic/bulgarian/mfa/v2.0.0a/meta.json +16 -0
  27. acoustic/bulgarian/mfa/v3.0.0/LICENSE +395 -0
  28. acoustic/bulgarian/mfa/v3.0.0/README.md +127 -0
  29. acoustic/bulgarian/mfa/v3.0.0/meta.json +16 -0
  30. acoustic/chuvash/cv/v2.0.0/README.md +110 -0
  31. acoustic/chuvash/cv/v2.0.0/meta.json +16 -0
  32. acoustic/croatian.zip +3 -0
  33. acoustic/croatian/mfa/v2.0.0/LICENSE +395 -0
  34. acoustic/croatian/mfa/v2.0.0/README.md +128 -0
  35. acoustic/croatian/mfa/v2.0.0/meta.json +16 -0
  36. acoustic/croatian/mfa/v2.0.0a/LICENSE +395 -0
  37. acoustic/croatian/mfa/v2.0.0a/README.md +128 -0
  38. acoustic/croatian/mfa/v2.0.0a/meta.json +16 -0
  39. acoustic/czech.zip +3 -0
  40. acoustic/czech/cv/v2.0.0/README.md +110 -0
  41. acoustic/czech/cv/v2.0.0/meta.json +16 -0
  42. acoustic/czech/mfa/v2.0.0/LICENSE +395 -0
  43. acoustic/czech/mfa/v2.0.0/README.md +138 -0
  44. acoustic/czech/mfa/v2.0.0/meta.json +16 -0
  45. acoustic/czech/mfa/v2.0.0a/LICENSE +395 -0
  46. acoustic/czech/mfa/v2.0.0a/README.md +138 -0
  47. acoustic/czech/mfa/v2.0.0a/meta.json +16 -0
  48. acoustic/czech/mfa/v3.3.0/LICENSE +395 -0
  49. acoustic/czech/mfa/v3.3.0/README.md +137 -0
  50. acoustic/czech/mfa/v3.3.0/meta.json +16 -0
.gitattributes CHANGED
@@ -33,3 +33,19 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ dictionary/german_prosodylab.dict filter=lfs diff=lfs merge=lfs -text
37
+ dictionary/german/prosodylab/v2.0.0/german_prosodylab.dict filter=lfs diff=lfs merge=lfs -text
38
+ dictionary/japanese/mfa/japanese_mfa.dict filter=lfs diff=lfs merge=lfs -text
39
+ dictionary/russian/mfa/russian_mfa.dict filter=lfs diff=lfs merge=lfs -text
40
+ docs/source/_static/favicon.ico filter=lfs diff=lfs merge=lfs -text
41
+ docs/source/_static/fonts/GentiumPlus-Bold.woff filter=lfs diff=lfs merge=lfs -text
42
+ docs/source/_static/fonts/GentiumPlus-Bold.woff2 filter=lfs diff=lfs merge=lfs -text
43
+ docs/source/_static/fonts/GentiumPlus-BoldItalic.woff filter=lfs diff=lfs merge=lfs -text
44
+ docs/source/_static/fonts/GentiumPlus-BoldItalic.woff2 filter=lfs diff=lfs merge=lfs -text
45
+ docs/source/_static/fonts/GentiumPlus-Italic.woff filter=lfs diff=lfs merge=lfs -text
46
+ docs/source/_static/fonts/GentiumPlus-Italic.woff2 filter=lfs diff=lfs merge=lfs -text
47
+ docs/source/_static/fonts/GentiumPlus-Regular.woff filter=lfs diff=lfs merge=lfs -text
48
+ docs/source/_static/fonts/GentiumPlus-Regular.woff2 filter=lfs diff=lfs merge=lfs -text
49
+ manuals/how-i-used-montreal-forced-aligner-for-a-new-language-sinhalese.pdf filter=lfs diff=lfs merge=lfs -text
50
+ manuals/montreal-forced-alignment-hebrew.pdf filter=lfs diff=lfs merge=lfs -text
51
+ paper/Montreal[[:space:]]Forced[[:space:]]Aligner.[[:space:]]Trainable[[:space:]]text-speech[[:space:]]alignment[[:space:]]using[[:space:]]Kaldi.pdf filter=lfs diff=lfs merge=lfs -text
.gitignore ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+
2
+ .idea/
3
+
4
+ g2p_models_backup/
5
+
6
+ docs/build/
7
+
8
+ dictionary/temp/
9
+
10
+ training/
11
+ staging/
12
+ generated/
13
+
14
+ dictionary/filter_lists/
15
+
16
+ scripts/temp/
17
+ /scripts/token
18
+
19
+ *.db
20
+
21
+ *.pyc
22
+
23
+ *.pclprof
24
+ .Rproj.user
25
+
26
+ *.Rproj
27
+
28
+ scripts/corpus_prep/temp/
29
+
30
+ .Rhistory
31
+
32
+ .RData
.pre-commit-config.yaml ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ repos:
2
+ - repo: https://github.com/pre-commit/pre-commit-hooks
3
+ rev: v4.0.1 # Use the ref you want to point at
4
+ hooks:
5
+ - id: check-ast
6
+ - id: check-builtin-literals
7
+ - id: check-docstring-first
8
+ - id: check-merge-conflict
9
+ - id: check-yaml
10
+ - id: check-toml
11
+ - id: file-contents-sorter
12
+ files: ^(dictionary/.*.txt|dictionary/.*.dict|acoustic/index.txt|g2p/index.txt|ivector/index.txt)$
13
+ - id: end-of-file-fixer
14
+ - id: trailing-whitespace
15
+ - id: debug-statements
16
+ - id: mixed-line-ending
17
+ - id: check-added-large-files
18
+ args: ['--maxkb=3000']
19
+ - repo: https://github.com/psf/black
20
+ rev: 23.9.1
21
+ hooks:
22
+ - id: black
23
+ #- repo: https://github.com/pycqa/flake8
24
+ # rev: 7.0.0
25
+ # hooks:
26
+ # - id: flake8
27
+ # additional_dependencies:
28
+ # - pyproject-flake8
29
+ - repo: https://github.com/pre-commit/mirrors-isort
30
+ rev: v5.10.1
31
+ hooks:
32
+ - id: isort
33
+ additional_dependencies: [toml]
34
+ - repo: https://github.com/asottile/setup-cfg-fmt
35
+ rev: v2.2.0
36
+ hooks:
37
+ - id: setup-cfg-fmt
38
+ args:
39
+ - --min-py3-version
40
+ - "3.8"
.readthedocs.yaml ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ version: 2
2
+
3
+ build:
4
+ os: "ubuntu-20.04"
5
+ tools:
6
+ python: "mambaforge-4.10"
7
+
8
+ sphinx:
9
+ configuration: docs/source/conf.py
10
+
11
+ conda:
12
+ environment: rtd_environment.yml
LICENSE ADDED
@@ -0,0 +1,395 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ Attribution 4.0 International
2
+
3
+ =======================================================================
4
+
5
+ Creative Commons Corporation ("Creative Commons") is not a law firm and
6
+ does not provide legal services or legal advice. Distribution of
7
+ Creative Commons public licenses does not create a lawyer-client or
8
+ other relationship. Creative Commons makes its licenses and related
9
+ information available on an "as-is" basis. Creative Commons gives no
10
+ warranties regarding its licenses, any material licensed under their
11
+ terms and conditions, or any related information. Creative Commons
12
+ disclaims all liability for damages resulting from their use to the
13
+ fullest extent possible.
14
+
15
+ Using Creative Commons Public Licenses
16
+
17
+ Creative Commons public licenses provide a standard set of terms and
18
+ conditions that creators and other rights holders may use to share
19
+ original works of authorship and other material subject to copyright
20
+ and certain other rights specified in the public license below. The
21
+ following considerations are for informational purposes only, are not
22
+ exhaustive, and do not form part of our licenses.
23
+
24
+ Considerations for licensors: Our public licenses are
25
+ intended for use by those authorized to give the public
26
+ permission to use material in ways otherwise restricted by
27
+ copyright and certain other rights. Our licenses are
28
+ irrevocable. Licensors should read and understand the terms
29
+ and conditions of the license they choose before applying it.
30
+ Licensors should also secure all rights necessary before
31
+ applying our licenses so that the public can reuse the
32
+ material as expected. Licensors should clearly mark any
33
+ material not subject to the license. This includes other CC-
34
+ licensed material, or material used under an exception or
35
+ limitation to copyright. More considerations for licensors:
36
+ wiki.creativecommons.org/Considerations_for_licensors
37
+
38
+ Considerations for the public: By using one of our public
39
+ licenses, a licensor grants the public permission to use the
40
+ licensed material under specified terms and conditions. If
41
+ the licensor's permission is not necessary for any reason--for
42
+ example, because of any applicable exception or limitation to
43
+ copyright--then that use is not regulated by the license. Our
44
+ licenses grant only permissions under copyright and certain
45
+ other rights that a licensor has authority to grant. Use of
46
+ the licensed material may still be restricted for other
47
+ reasons, including because others have copyright or other
48
+ rights in the material. A licensor may make special requests,
49
+ such as asking that all changes be marked or described.
50
+ Although not required by our licenses, you are encouraged to
51
+ respect those requests where reasonable. More_considerations
52
+ for the public:
53
+ wiki.creativecommons.org/Considerations_for_licensees
54
+
55
+ =======================================================================
56
+
57
+ Creative Commons Attribution 4.0 International Public License
58
+
59
+ By exercising the Licensed Rights (defined below), You accept and agree
60
+ to be bound by the terms and conditions of this Creative Commons
61
+ Attribution 4.0 International Public License ("Public License"). To the
62
+ extent this Public License may be interpreted as a contract, You are
63
+ granted the Licensed Rights in consideration of Your acceptance of
64
+ these terms and conditions, and the Licensor grants You such rights in
65
+ consideration of benefits the Licensor receives from making the
66
+ Licensed Material available under these terms and conditions.
67
+
68
+
69
+ Section 1 -- Definitions.
70
+
71
+ a. Adapted Material means material subject to Copyright and Similar
72
+ Rights that is derived from or based upon the Licensed Material
73
+ and in which the Licensed Material is translated, altered,
74
+ arranged, transformed, or otherwise modified in a manner requiring
75
+ permission under the Copyright and Similar Rights held by the
76
+ Licensor. For purposes of this Public License, where the Licensed
77
+ Material is a musical work, performance, or sound recording,
78
+ Adapted Material is always produced where the Licensed Material is
79
+ synched in timed relation with a moving image.
80
+
81
+ b. Adapter's License means the license You apply to Your Copyright
82
+ and Similar Rights in Your contributions to Adapted Material in
83
+ accordance with the terms and conditions of this Public License.
84
+
85
+ c. Copyright and Similar Rights means copyright and/or similar rights
86
+ closely related to copyright including, without limitation,
87
+ performance, broadcast, sound recording, and Sui Generis Database
88
+ Rights, without regard to how the rights are labeled or
89
+ categorized. For purposes of this Public License, the rights
90
+ specified in Section 2(b)(1)-(2) are not Copyright and Similar
91
+ Rights.
92
+
93
+ d. Effective Technological Measures means those measures that, in the
94
+ absence of proper authority, may not be circumvented under laws
95
+ fulfilling obligations under Article 11 of the WIPO Copyright
96
+ Treaty adopted on December 20, 1996, and/or similar international
97
+ agreements.
98
+
99
+ e. Exceptions and Limitations means fair use, fair dealing, and/or
100
+ any other exception or limitation to Copyright and Similar Rights
101
+ that applies to Your use of the Licensed Material.
102
+
103
+ f. Licensed Material means the artistic or literary work, database,
104
+ or other material to which the Licensor applied this Public
105
+ License.
106
+
107
+ g. Licensed Rights means the rights granted to You subject to the
108
+ terms and conditions of this Public License, which are limited to
109
+ all Copyright and Similar Rights that apply to Your use of the
110
+ Licensed Material and that the Licensor has authority to license.
111
+
112
+ h. Licensor means the individual(s) or entity(ies) granting rights
113
+ under this Public License.
114
+
115
+ i. Share means to provide material to the public by any means or
116
+ process that requires permission under the Licensed Rights, such
117
+ as reproduction, public display, public performance, distribution,
118
+ dissemination, communication, or importation, and to make material
119
+ available to the public including in ways that members of the
120
+ public may access the material from a place and at a time
121
+ individually chosen by them.
122
+
123
+ j. Sui Generis Database Rights means rights other than copyright
124
+ resulting from Directive 96/9/EC of the European Parliament and of
125
+ the Council of 11 March 1996 on the legal protection of databases,
126
+ as amended and/or succeeded, as well as other essentially
127
+ equivalent rights anywhere in the world.
128
+
129
+ k. You means the individual or entity exercising the Licensed Rights
130
+ under this Public License. Your has a corresponding meaning.
131
+
132
+
133
+ Section 2 -- Scope.
134
+
135
+ a. License grant.
136
+
137
+ 1. Subject to the terms and conditions of this Public License,
138
+ the Licensor hereby grants You a worldwide, royalty-free,
139
+ non-sublicensable, non-exclusive, irrevocable license to
140
+ exercise the Licensed Rights in the Licensed Material to:
141
+
142
+ a. reproduce and Share the Licensed Material, in whole or
143
+ in part; and
144
+
145
+ b. produce, reproduce, and Share Adapted Material.
146
+
147
+ 2. Exceptions and Limitations. For the avoidance of doubt, where
148
+ Exceptions and Limitations apply to Your use, this Public
149
+ License does not apply, and You do not need to comply with
150
+ its terms and conditions.
151
+
152
+ 3. Term. The term of this Public License is specified in Section
153
+ 6(a).
154
+
155
+ 4. Media and formats; technical modifications allowed. The
156
+ Licensor authorizes You to exercise the Licensed Rights in
157
+ all media and formats whether now known or hereafter created,
158
+ and to make technical modifications necessary to do so. The
159
+ Licensor waives and/or agrees not to assert any right or
160
+ authority to forbid You from making technical modifications
161
+ necessary to exercise the Licensed Rights, including
162
+ technical modifications necessary to circumvent Effective
163
+ Technological Measures. For purposes of this Public License,
164
+ simply making modifications authorized by this Section 2(a)
165
+ (4) never produces Adapted Material.
166
+
167
+ 5. Downstream recipients.
168
+
169
+ a. Offer from the Licensor -- Licensed Material. Every
170
+ recipient of the Licensed Material automatically
171
+ receives an offer from the Licensor to exercise the
172
+ Licensed Rights under the terms and conditions of this
173
+ Public License.
174
+
175
+ b. No downstream restrictions. You may not offer or impose
176
+ any additional or different terms or conditions on, or
177
+ apply any Effective Technological Measures to, the
178
+ Licensed Material if doing so restricts exercise of the
179
+ Licensed Rights by any recipient of the Licensed
180
+ Material.
181
+
182
+ 6. No endorsement. Nothing in this Public License constitutes or
183
+ may be construed as permission to assert or imply that You
184
+ are, or that Your use of the Licensed Material is, connected
185
+ with, or sponsored, endorsed, or granted official status by,
186
+ the Licensor or others designated to receive attribution as
187
+ provided in Section 3(a)(1)(A)(i).
188
+
189
+ b. Other rights.
190
+
191
+ 1. Moral rights, such as the right of integrity, are not
192
+ licensed under this Public License, nor are publicity,
193
+ privacy, and/or other similar personality rights; however, to
194
+ the extent possible, the Licensor waives and/or agrees not to
195
+ assert any such rights held by the Licensor to the limited
196
+ extent necessary to allow You to exercise the Licensed
197
+ Rights, but not otherwise.
198
+
199
+ 2. Patent and trademark rights are not licensed under this
200
+ Public License.
201
+
202
+ 3. To the extent possible, the Licensor waives any right to
203
+ collect royalties from You for the exercise of the Licensed
204
+ Rights, whether directly or through a collecting society
205
+ under any voluntary or waivable statutory or compulsory
206
+ licensing scheme. In all other cases the Licensor expressly
207
+ reserves any right to collect such royalties.
208
+
209
+
210
+ Section 3 -- License Conditions.
211
+
212
+ Your exercise of the Licensed Rights is expressly made subject to the
213
+ following conditions.
214
+
215
+ a. Attribution.
216
+
217
+ 1. If You Share the Licensed Material (including in modified
218
+ form), You must:
219
+
220
+ a. retain the following if it is supplied by the Licensor
221
+ with the Licensed Material:
222
+
223
+ i. identification of the creator(s) of the Licensed
224
+ Material and any others designated to receive
225
+ attribution, in any reasonable manner requested by
226
+ the Licensor (including by pseudonym if
227
+ designated);
228
+
229
+ ii. a copyright notice;
230
+
231
+ iii. a notice that refers to this Public License;
232
+
233
+ iv. a notice that refers to the disclaimer of
234
+ warranties;
235
+
236
+ v. a URI or hyperlink to the Licensed Material to the
237
+ extent reasonably practicable;
238
+
239
+ b. indicate if You modified the Licensed Material and
240
+ retain an indication of any previous modifications; and
241
+
242
+ c. indicate the Licensed Material is licensed under this
243
+ Public License, and include the text of, or the URI or
244
+ hyperlink to, this Public License.
245
+
246
+ 2. You may satisfy the conditions in Section 3(a)(1) in any
247
+ reasonable manner based on the medium, means, and context in
248
+ which You Share the Licensed Material. For example, it may be
249
+ reasonable to satisfy the conditions by providing a URI or
250
+ hyperlink to a resource that includes the required
251
+ information.
252
+
253
+ 3. If requested by the Licensor, You must remove any of the
254
+ information required by Section 3(a)(1)(A) to the extent
255
+ reasonably practicable.
256
+
257
+ 4. If You Share Adapted Material You produce, the Adapter's
258
+ License You apply must not prevent recipients of the Adapted
259
+ Material from complying with this Public License.
260
+
261
+
262
+ Section 4 -- Sui Generis Database Rights.
263
+
264
+ Where the Licensed Rights include Sui Generis Database Rights that
265
+ apply to Your use of the Licensed Material:
266
+
267
+ a. for the avoidance of doubt, Section 2(a)(1) grants You the right
268
+ to extract, reuse, reproduce, and Share all or a substantial
269
+ portion of the contents of the database;
270
+
271
+ b. if You include all or a substantial portion of the database
272
+ contents in a database in which You have Sui Generis Database
273
+ Rights, then the database in which You have Sui Generis Database
274
+ Rights (but not its individual contents) is Adapted Material; and
275
+
276
+ c. You must comply with the conditions in Section 3(a) if You Share
277
+ all or a substantial portion of the contents of the database.
278
+
279
+ For the avoidance of doubt, this Section 4 supplements and does not
280
+ replace Your obligations under this Public License where the Licensed
281
+ Rights include other Copyright and Similar Rights.
282
+
283
+
284
+ Section 5 -- Disclaimer of Warranties and Limitation of Liability.
285
+
286
+ a. UNLESS OTHERWISE SEPARATELY UNDERTAKEN BY THE LICENSOR, TO THE
287
+ EXTENT POSSIBLE, THE LICENSOR OFFERS THE LICENSED MATERIAL AS-IS
288
+ AND AS-AVAILABLE, AND MAKES NO REPRESENTATIONS OR WARRANTIES OF
289
+ ANY KIND CONCERNING THE LICENSED MATERIAL, WHETHER EXPRESS,
290
+ IMPLIED, STATUTORY, OR OTHER. THIS INCLUDES, WITHOUT LIMITATION,
291
+ WARRANTIES OF TITLE, MERCHANTABILITY, FITNESS FOR A PARTICULAR
292
+ PURPOSE, NON-INFRINGEMENT, ABSENCE OF LATENT OR OTHER DEFECTS,
293
+ ACCURACY, OR THE PRESENCE OR ABSENCE OF ERRORS, WHETHER OR NOT
294
+ KNOWN OR DISCOVERABLE. WHERE DISCLAIMERS OF WARRANTIES ARE NOT
295
+ ALLOWED IN FULL OR IN PART, THIS DISCLAIMER MAY NOT APPLY TO YOU.
296
+
297
+ b. TO THE EXTENT POSSIBLE, IN NO EVENT WILL THE LICENSOR BE LIABLE
298
+ TO YOU ON ANY LEGAL THEORY (INCLUDING, WITHOUT LIMITATION,
299
+ NEGLIGENCE) OR OTHERWISE FOR ANY DIRECT, SPECIAL, INDIRECT,
300
+ INCIDENTAL, CONSEQUENTIAL, PUNITIVE, EXEMPLARY, OR OTHER LOSSES,
301
+ COSTS, EXPENSES, OR DAMAGES ARISING OUT OF THIS PUBLIC LICENSE OR
302
+ USE OF THE LICENSED MATERIAL, EVEN IF THE LICENSOR HAS BEEN
303
+ ADVISED OF THE POSSIBILITY OF SUCH LOSSES, COSTS, EXPENSES, OR
304
+ DAMAGES. WHERE A LIMITATION OF LIABILITY IS NOT ALLOWED IN FULL OR
305
+ IN PART, THIS LIMITATION MAY NOT APPLY TO YOU.
306
+
307
+ c. The disclaimer of warranties and limitation of liability provided
308
+ above shall be interpreted in a manner that, to the extent
309
+ possible, most closely approximates an absolute disclaimer and
310
+ waiver of all liability.
311
+
312
+
313
+ Section 6 -- Term and Termination.
314
+
315
+ a. This Public License applies for the term of the Copyright and
316
+ Similar Rights licensed here. However, if You fail to comply with
317
+ this Public License, then Your rights under this Public License
318
+ terminate automatically.
319
+
320
+ b. Where Your right to use the Licensed Material has terminated under
321
+ Section 6(a), it reinstates:
322
+
323
+ 1. automatically as of the date the violation is cured, provided
324
+ it is cured within 30 days of Your discovery of the
325
+ violation; or
326
+
327
+ 2. upon express reinstatement by the Licensor.
328
+
329
+ For the avoidance of doubt, this Section 6(b) does not affect any
330
+ right the Licensor may have to seek remedies for Your violations
331
+ of this Public License.
332
+
333
+ c. For the avoidance of doubt, the Licensor may also offer the
334
+ Licensed Material under separate terms or conditions or stop
335
+ distributing the Licensed Material at any time; however, doing so
336
+ will not terminate this Public License.
337
+
338
+ d. Sections 1, 5, 6, 7, and 8 survive termination of this Public
339
+ License.
340
+
341
+
342
+ Section 7 -- Other Terms and Conditions.
343
+
344
+ a. The Licensor shall not be bound by any additional or different
345
+ terms or conditions communicated by You unless expressly agreed.
346
+
347
+ b. Any arrangements, understandings, or agreements regarding the
348
+ Licensed Material not stated herein are separate from and
349
+ independent of the terms and conditions of this Public License.
350
+
351
+
352
+ Section 8 -- Interpretation.
353
+
354
+ a. For the avoidance of doubt, this Public License does not, and
355
+ shall not be interpreted to, reduce, limit, restrict, or impose
356
+ conditions on any use of the Licensed Material that could lawfully
357
+ be made without permission under this Public License.
358
+
359
+ b. To the extent possible, if any provision of this Public License is
360
+ deemed unenforceable, it shall be automatically reformed to the
361
+ minimum extent necessary to make it enforceable. If the provision
362
+ cannot be reformed, it shall be severed from this Public License
363
+ without affecting the enforceability of the remaining terms and
364
+ conditions.
365
+
366
+ c. No term or condition of this Public License will be waived and no
367
+ failure to comply consented to unless expressly agreed to by the
368
+ Licensor.
369
+
370
+ d. Nothing in this Public License constitutes or may be interpreted
371
+ as a limitation upon, or waiver of, any privileges and immunities
372
+ that apply to the Licensor or You, including from the legal
373
+ processes of any jurisdiction or authority.
374
+
375
+
376
+ =======================================================================
377
+
378
+ Creative Commons is not a party to its public
379
+ licenses. Notwithstanding, Creative Commons may elect to apply one of
380
+ its public licenses to material it publishes and in those instances
381
+ will be considered the “Licensor.” The text of the Creative Commons
382
+ public licenses is dedicated to the public domain under the CC0 Public
383
+ Domain Dedication. Except for the limited purpose of indicating that
384
+ material is shared under a Creative Commons public license or as
385
+ otherwise permitted by the Creative Commons policies published at
386
+ creativecommons.org/policies, Creative Commons does not authorize the
387
+ use of the trademark "Creative Commons" or any other trademark or logo
388
+ of Creative Commons without its prior written consent including,
389
+ without limitation, in connection with any unauthorized modifications
390
+ to any of its public licenses or any other arrangements,
391
+ understandings, or agreements concerning use of licensed material. For
392
+ the avoidance of doubt, this paragraph does not form part of the
393
+ public licenses.
394
+
395
+ Creative Commons may be contacted at creativecommons.org.
README.md ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ # Montreal Forced Aligner Models
2
+
3
+ Collection of pretrained models and pronunciation dictionaries for use with [Montreal Forced Aligner](https://montreal-forced-aligner.readthedocs.io/).
4
+
5
+ See also https://mfa-models.readthedocs.io/en/latest/ for full model details and how to use and cite them.
acoustic/abkhaz/cv/v2.0.0/README.md ADDED
@@ -0,0 +1,110 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+
2
+ # Abkhaz CV acoustic model v2.0.0
3
+
4
+ [Link to documentation on mfa-models](https://mfa-models.readthedocs.io/en/main/acoustic/abkhaz_cv.html)
5
+
6
+ Jump to section:
7
+
8
+ - [Model details](#model-details)
9
+ - [Installation](#installation)
10
+ - [Intended use](#intended-use)
11
+ - [Performance Factors](#performance-factors)
12
+ - [Training data](#training-data)
13
+ - [Ethical considerations](#ethical-considerations)
14
+ - [Troubleshooting issues](#troubleshooting-issues)
15
+
16
+ ## Model details
17
+
18
+ - **Maintainer:** [Vox Communis](https://osf.io/t957v/)
19
+ - **Language:** [Abkhaz](https://en.wikipedia.org/wiki/Abkhaz_language)
20
+ - **Dialect:** N/A
21
+ - **Phone set:** [XPF](https://github.com/CohenPr-XPF/XPF)
22
+ - **Model type:** `Acoustic model`
23
+ - **Features:** `MFCC`
24
+ - **Architecture:** `gmm-hmm`
25
+ - **Model version:** `v2.0.0`
26
+ - **Trained date:** `02-11-2022`
27
+ - **Compatible MFA version:** `v2.0.0`
28
+ - **License:** [CC-0](https://creativecommons.org/publicdomain/zero/1.0/)
29
+ - **Citation:**
30
+
31
+ ```bibtex
32
+ @misc{Ahn_Chodroff_2022,
33
+ author={Ahn, Emily and Chodroff, Eleanor},
34
+ title={VoxCommunis Corpus},
35
+ address={\url{https://osf.io/t957v}},
36
+ publisher={OSF},
37
+ year={2022},
38
+ month={Jan}
39
+ }
40
+ ```
41
+
42
+ - If you have comments or questions about this model, you can check [previous MFA model discussion posts](https://github.com/MontrealCorpusTools/mfa-models/discussions?discussions_q=Abkhaz+CV+acoustic+model+v2.0.0) or create [a new one](https://github.com/MontrealCorpusTools/mfa-models/discussions/new).
43
+
44
+ ## Installation
45
+
46
+ Install from the [MFA command line](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/models/index.html):
47
+
48
+ ```
49
+ mfa model download acoustic abkhaz_cv
50
+ ```
51
+
52
+ Or download from [the release page](https://github.com/MontrealCorpusTools/mfa-models/releases/tag/acoustic-abkhaz_cv-v2.0.0).
53
+
54
+ ## Intended use
55
+
56
+ This model is intended for forced alignment of [Abkhaz](https://en.wikipedia.org/wiki/Abkhaz_language) transcripts.
57
+
58
+ This model uses the [XPF](https://github.com/CohenPr-XPF/XPF) phone set for Abkhaz, and was trained with the pronunciation dictionaries above. Pronunciations can be added on top of the dictionary, as long as no additional phones are introduced.
59
+
60
+ ## Performance Factors
61
+
62
+ As forced alignment is a relatively well-constrained problem (given accurate transcripts), this model should be applicable to a range of recording conditions and speakers. However, please note that it was trained on read speech in low-noise environments, so as your data diverges from that, you may run into alignment issues or need to [increase the beam size of MFA](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/configuration/#configuring-specific-commands) or see other recommendations in the [troubleshooting section below](#troubleshooting-issues).
63
+
64
+ Please note as well that MFA does not use state-of-the-art ASR models for forced alignment. You may get better performance (especially on speech-to-text tasks) using other frameworks like [Coqui](https://coqui.ai/).
65
+
66
+ ## Ethical considerations
67
+
68
+ Deploying any Speech-to-Text model into any production setting has ethical implications. You should consider these implications before use.
69
+
70
+ ### Demographic Bias
71
+
72
+ You should assume every machine learning model has demographic bias unless proven otherwise. For STT models, it is often the case that transcription accuracy is better for men than it is for women. If you are using this model in production, you should acknowledge this as a potential issue.
73
+
74
+ ### Surveillance
75
+
76
+ Speech-to-Text technologies may be misused to invade the privacy of others by recording and mining information from private conversations. This kind of individual privacy is protected by law in many countries. You should not assume consent to record and analyze private speech.
77
+
78
+ ## Troubleshooting issues
79
+
80
+ Machine learning models (like this acoustic model) perform best on data that is similar to the data on which they were trained.
81
+
82
+ The primary sources of variability in forced alignment will be the applicability of the pronunciation dictionary and how similar the speech, demographics, and recording conditions are. If you encounter issues in alignment, there are couple of avenues to improve performance:
83
+
84
+ 1. [Increase the beam size of MFA](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/configuration/#configuring-specific-commands)
85
+
86
+ * MFA defaults to a narrow beam to ensure quick alignment and also as a way to detect potential issues in your dataset, but depending on your data, you might benefit from boosting the beam to 100 or higher.
87
+
88
+ 2. Add pronunciations to the pronunciation dictionary
89
+
90
+ * This model was trained a particular dialect/style, and so adding pronunciations more representative of the variety spoken in your dataset will help alignment.
91
+
92
+ 3. Check the quality of your data
93
+
94
+ * MFA includes a [validator utility](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/data_validation.html), which aims to detect issues in the dataset.
95
+ * Use MFA's [anchor utility](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/workflows/anchor.html) to visually inspect your data as MFA sees it and correct issues in transcription or OOV items.
96
+
97
+ 4. Adapt the model to your data
98
+
99
+ * MFA has an [adaptation command](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/workflows/adapt_acoustic_model.html) to adapt some of the model to your data based on an initial alignment, and then run another alignment with the adapted model.
100
+
101
+ ## Training data
102
+
103
+ This model was trained on the following corpora:
104
+
105
+
106
+
107
+ * [Common Voice Abkhaz](../../../../corpus/abkhaz/common_voice_abkhaz/7.0/README.md):
108
+ * **Hours:** `2.00`
109
+ * **Speakers:** `28`
110
+ * **Utterances:** `1,166`
acoustic/abkhaz/cv/v2.0.0/meta.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "name": "abkhaz_cv",
3
+ "language": "Abkhaz",
4
+ "dialect": "",
5
+ "phone_set": "CV",
6
+ "version": "2.0.0",
7
+ "maintainer": "[Vox Communis](https://osf.io/t957v/)",
8
+ "citation": "@misc{Ahn_Chodroff_2022,\n\tauthor={Ahn, Emily and Chodroff, Eleanor},\n\ttitle={VoxCommunis Corpus},\n\taddress={\\url{https://osf.io/t957v}},\n\tpublisher={OSF},\n\tyear={2022}, \n\tmonth={Jan}\n}",
9
+ "license": "CC-0",
10
+ "license_link": "[CC-0](https://creativecommons.org/publicdomain/zero/1.0/)",
11
+ "architecture": "gmm-hmm",
12
+ "features": "MFCC",
13
+ "evaluation": {},
14
+ "decode": {},
15
+ "train_date": "02-11-2022"
16
+ }
acoustic/arabic.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:399954e19f3ccdded9821b1b12ac371201a0f881bbd7c5bba8cb8cd73fbc99cb
3
+ size 14681790
acoustic/armenian/cv/v2.0.0/README.md ADDED
@@ -0,0 +1,110 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+
2
+ # Armenian CV acoustic model v2.0.0
3
+
4
+ [Link to documentation on mfa-models](https://mfa-models.readthedocs.io/en/main/acoustic/armenian_cv.html)
5
+
6
+ Jump to section:
7
+
8
+ - [Model details](#model-details)
9
+ - [Installation](#installation)
10
+ - [Intended use](#intended-use)
11
+ - [Performance Factors](#performance-factors)
12
+ - [Training data](#training-data)
13
+ - [Ethical considerations](#ethical-considerations)
14
+ - [Troubleshooting issues](#troubleshooting-issues)
15
+
16
+ ## Model details
17
+
18
+ - **Maintainer:** [Vox Communis](https://osf.io/t957v/)
19
+ - **Language:** [Armenian](https://en.wikipedia.org/wiki/Armenian_language)
20
+ - **Dialect:** N/A
21
+ - **Phone set:** [XPF](https://github.com/CohenPr-XPF/XPF)
22
+ - **Model type:** `Acoustic model`
23
+ - **Features:** `MFCC`
24
+ - **Architecture:** `gmm-hmm`
25
+ - **Model version:** `v2.0.0`
26
+ - **Trained date:** `02-11-2022`
27
+ - **Compatible MFA version:** `v2.0.0`
28
+ - **License:** [CC-0](https://creativecommons.org/publicdomain/zero/1.0/)
29
+ - **Citation:**
30
+
31
+ ```bibtex
32
+ @misc{Ahn_Chodroff_2022,
33
+ author={Ahn, Emily and Chodroff, Eleanor},
34
+ title={VoxCommunis Corpus},
35
+ address={\url{https://osf.io/t957v}},
36
+ publisher={OSF},
37
+ year={2022},
38
+ month={Jan}
39
+ }
40
+ ```
41
+
42
+ - If you have comments or questions about this model, you can check [previous MFA model discussion posts](https://github.com/MontrealCorpusTools/mfa-models/discussions?discussions_q=Armenian+CV+acoustic+model+v2.0.0) or create [a new one](https://github.com/MontrealCorpusTools/mfa-models/discussions/new).
43
+
44
+ ## Installation
45
+
46
+ Install from the [MFA command line](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/models/index.html):
47
+
48
+ ```
49
+ mfa model download acoustic armenian_cv
50
+ ```
51
+
52
+ Or download from [the release page](https://github.com/MontrealCorpusTools/mfa-models/releases/tag/acoustic-armenian_cv-v2.0.0).
53
+
54
+ ## Intended use
55
+
56
+ This model is intended for forced alignment of [Armenian](https://en.wikipedia.org/wiki/Armenian_language) transcripts.
57
+
58
+ This model uses the [XPF](https://github.com/CohenPr-XPF/XPF) phone set for Armenian, and was trained with the pronunciation dictionaries above. Pronunciations can be added on top of the dictionary, as long as no additional phones are introduced.
59
+
60
+ ## Performance Factors
61
+
62
+ As forced alignment is a relatively well-constrained problem (given accurate transcripts), this model should be applicable to a range of recording conditions and speakers. However, please note that it was trained on read speech in low-noise environments, so as your data diverges from that, you may run into alignment issues or need to [increase the beam size of MFA](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/configuration/#configuring-specific-commands) or see other recommendations in the [troubleshooting section below](#troubleshooting-issues).
63
+
64
+ Please note as well that MFA does not use state-of-the-art ASR models for forced alignment. You may get better performance (especially on speech-to-text tasks) using other frameworks like [Coqui](https://coqui.ai/).
65
+
66
+ ## Ethical considerations
67
+
68
+ Deploying any Speech-to-Text model into any production setting has ethical implications. You should consider these implications before use.
69
+
70
+ ### Demographic Bias
71
+
72
+ You should assume every machine learning model has demographic bias unless proven otherwise. For STT models, it is often the case that transcription accuracy is better for men than it is for women. If you are using this model in production, you should acknowledge this as a potential issue.
73
+
74
+ ### Surveillance
75
+
76
+ Speech-to-Text technologies may be misused to invade the privacy of others by recording and mining information from private conversations. This kind of individual privacy is protected by law in many countries. You should not assume consent to record and analyze private speech.
77
+
78
+ ## Troubleshooting issues
79
+
80
+ Machine learning models (like this acoustic model) perform best on data that is similar to the data on which they were trained.
81
+
82
+ The primary sources of variability in forced alignment will be the applicability of the pronunciation dictionary and how similar the speech, demographics, and recording conditions are. If you encounter issues in alignment, there are couple of avenues to improve performance:
83
+
84
+ 1. [Increase the beam size of MFA](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/configuration/#configuring-specific-commands)
85
+
86
+ * MFA defaults to a narrow beam to ensure quick alignment and also as a way to detect potential issues in your dataset, but depending on your data, you might benefit from boosting the beam to 100 or higher.
87
+
88
+ 2. Add pronunciations to the pronunciation dictionary
89
+
90
+ * This model was trained a particular dialect/style, and so adding pronunciations more representative of the variety spoken in your dataset will help alignment.
91
+
92
+ 3. Check the quality of your data
93
+
94
+ * MFA includes a [validator utility](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/data_validation.html), which aims to detect issues in the dataset.
95
+ * Use MFA's [anchor utility](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/workflows/anchor.html) to visually inspect your data as MFA sees it and correct issues in transcription or OOV items.
96
+
97
+ 4. Adapt the model to your data
98
+
99
+ * MFA has an [adaptation command](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/workflows/adapt_acoustic_model.html) to adapt some of the model to your data based on an initial alignment, and then run another alignment with the adapted model.
100
+
101
+ ## Training data
102
+
103
+ This model was trained on the following corpora:
104
+
105
+
106
+
107
+ * [Common Voice Armenian](../../../../corpus/armenian/common_voice_armenian/7.0/README.md):
108
+ * **Hours:** `1.00`
109
+ * **Speakers:** `22`
110
+ * **Utterances:** `767`
acoustic/armenian/cv/v2.0.0/meta.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "name": "armenian_cv",
3
+ "language": "Armenian",
4
+ "dialect": "",
5
+ "phone_set": "CV",
6
+ "version": "2.0.0",
7
+ "maintainer": "[Vox Communis](https://osf.io/t957v/)",
8
+ "citation": "@misc{Ahn_Chodroff_2022,\n\tauthor={Ahn, Emily and Chodroff, Eleanor},\n\ttitle={VoxCommunis Corpus},\n\taddress={\\url{https://osf.io/t957v}},\n\tpublisher={OSF},\n\tyear={2022}, \n\tmonth={Jan}\n}",
9
+ "license": "CC-0",
10
+ "license_link": "[CC-0](https://creativecommons.org/publicdomain/zero/1.0/)",
11
+ "architecture": "gmm-hmm",
12
+ "features": "MFCC",
13
+ "evaluation": {},
14
+ "decode": {},
15
+ "train_date": "02-11-2022"
16
+ }
acoustic/bashkir/cv/v2.0.0/README.md ADDED
@@ -0,0 +1,110 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+
2
+ # Bashkir CV acoustic model v2.0.0
3
+
4
+ [Link to documentation on mfa-models](https://mfa-models.readthedocs.io/en/main/acoustic/bashkir_cv.html)
5
+
6
+ Jump to section:
7
+
8
+ - [Model details](#model-details)
9
+ - [Installation](#installation)
10
+ - [Intended use](#intended-use)
11
+ - [Performance Factors](#performance-factors)
12
+ - [Training data](#training-data)
13
+ - [Ethical considerations](#ethical-considerations)
14
+ - [Troubleshooting issues](#troubleshooting-issues)
15
+
16
+ ## Model details
17
+
18
+ - **Maintainer:** [Vox Communis](https://osf.io/t957v/)
19
+ - **Language:** [Bashkir](https://en.wikipedia.org/wiki/Bashkir_language)
20
+ - **Dialect:** N/A
21
+ - **Phone set:** [XPF](https://github.com/CohenPr-XPF/XPF)
22
+ - **Model type:** `Acoustic model`
23
+ - **Features:** `MFCC`
24
+ - **Architecture:** `gmm-hmm`
25
+ - **Model version:** `v2.0.0`
26
+ - **Trained date:** `02-11-2022`
27
+ - **Compatible MFA version:** `v2.0.0`
28
+ - **License:** [CC-0](https://creativecommons.org/publicdomain/zero/1.0/)
29
+ - **Citation:**
30
+
31
+ ```bibtex
32
+ @misc{Ahn_Chodroff_2022,
33
+ author={Ahn, Emily and Chodroff, Eleanor},
34
+ title={VoxCommunis Corpus},
35
+ address={\url{https://osf.io/t957v}},
36
+ publisher={OSF},
37
+ year={2022},
38
+ month={Jan}
39
+ }
40
+ ```
41
+
42
+ - If you have comments or questions about this model, you can check [previous MFA model discussion posts](https://github.com/MontrealCorpusTools/mfa-models/discussions?discussions_q=Bashkir+CV+acoustic+model+v2.0.0) or create [a new one](https://github.com/MontrealCorpusTools/mfa-models/discussions/new).
43
+
44
+ ## Installation
45
+
46
+ Install from the [MFA command line](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/models/index.html):
47
+
48
+ ```
49
+ mfa model download acoustic bashkir_cv
50
+ ```
51
+
52
+ Or download from [the release page](https://github.com/MontrealCorpusTools/mfa-models/releases/tag/acoustic-bashkir_cv-v2.0.0).
53
+
54
+ ## Intended use
55
+
56
+ This model is intended for forced alignment of [Bashkir](https://en.wikipedia.org/wiki/Bashkir_language) transcripts.
57
+
58
+ This model uses the [XPF](https://github.com/CohenPr-XPF/XPF) phone set for Bashkir, and was trained with the pronunciation dictionaries above. Pronunciations can be added on top of the dictionary, as long as no additional phones are introduced.
59
+
60
+ ## Performance Factors
61
+
62
+ As forced alignment is a relatively well-constrained problem (given accurate transcripts), this model should be applicable to a range of recording conditions and speakers. However, please note that it was trained on read speech in low-noise environments, so as your data diverges from that, you may run into alignment issues or need to [increase the beam size of MFA](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/configuration/#configuring-specific-commands) or see other recommendations in the [troubleshooting section below](#troubleshooting-issues).
63
+
64
+ Please note as well that MFA does not use state-of-the-art ASR models for forced alignment. You may get better performance (especially on speech-to-text tasks) using other frameworks like [Coqui](https://coqui.ai/).
65
+
66
+ ## Ethical considerations
67
+
68
+ Deploying any Speech-to-Text model into any production setting has ethical implications. You should consider these implications before use.
69
+
70
+ ### Demographic Bias
71
+
72
+ You should assume every machine learning model has demographic bias unless proven otherwise. For STT models, it is often the case that transcription accuracy is better for men than it is for women. If you are using this model in production, you should acknowledge this as a potential issue.
73
+
74
+ ### Surveillance
75
+
76
+ Speech-to-Text technologies may be misused to invade the privacy of others by recording and mining information from private conversations. This kind of individual privacy is protected by law in many countries. You should not assume consent to record and analyze private speech.
77
+
78
+ ## Troubleshooting issues
79
+
80
+ Machine learning models (like this acoustic model) perform best on data that is similar to the data on which they were trained.
81
+
82
+ The primary sources of variability in forced alignment will be the applicability of the pronunciation dictionary and how similar the speech, demographics, and recording conditions are. If you encounter issues in alignment, there are couple of avenues to improve performance:
83
+
84
+ 1. [Increase the beam size of MFA](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/configuration/#configuring-specific-commands)
85
+
86
+ * MFA defaults to a narrow beam to ensure quick alignment and also as a way to detect potential issues in your dataset, but depending on your data, you might benefit from boosting the beam to 100 or higher.
87
+
88
+ 2. Add pronunciations to the pronunciation dictionary
89
+
90
+ * This model was trained a particular dialect/style, and so adding pronunciations more representative of the variety spoken in your dataset will help alignment.
91
+
92
+ 3. Check the quality of your data
93
+
94
+ * MFA includes a [validator utility](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/data_validation.html), which aims to detect issues in the dataset.
95
+ * Use MFA's [anchor utility](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/workflows/anchor.html) to visually inspect your data as MFA sees it and correct issues in transcription or OOV items.
96
+
97
+ 4. Adapt the model to your data
98
+
99
+ * MFA has an [adaptation command](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/workflows/adapt_acoustic_model.html) to adapt some of the model to your data based on an initial alignment, and then run another alignment with the adapted model.
100
+
101
+ ## Training data
102
+
103
+ This model was trained on the following corpora:
104
+
105
+
106
+
107
+ * [Common Voice Bashkir](../../../../corpus/bashkir/common_voice_bashkir/7.0/README.md):
108
+ * **Hours:** `247.00`
109
+ * **Speakers:** `835`
110
+ * **Utterances:** `200,869`
acoustic/bashkir/cv/v2.0.0/meta.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "name": "bashkir_cv",
3
+ "language": "Bashkir",
4
+ "dialect": "",
5
+ "phone_set": "CV",
6
+ "version": "2.0.0",
7
+ "maintainer": "[Vox Communis](https://osf.io/t957v/)",
8
+ "citation": "@misc{Ahn_Chodroff_2022,\n\tauthor={Ahn, Emily and Chodroff, Eleanor},\n\ttitle={VoxCommunis Corpus},\n\taddress={\\url{https://osf.io/t957v}},\n\tpublisher={OSF},\n\tyear={2022}, \n\tmonth={Jan}\n}",
9
+ "license": "CC-0",
10
+ "license_link": "[CC-0](https://creativecommons.org/publicdomain/zero/1.0/)",
11
+ "architecture": "gmm-hmm",
12
+ "features": "MFCC",
13
+ "evaluation": {},
14
+ "decode": {},
15
+ "train_date": "02-11-2022"
16
+ }
acoustic/basque/cv/v2.0.0/README.md ADDED
@@ -0,0 +1,110 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+
2
+ # Basque CV acoustic model v2.0.0
3
+
4
+ [Link to documentation on mfa-models](https://mfa-models.readthedocs.io/en/main/acoustic/basque_cv.html)
5
+
6
+ Jump to section:
7
+
8
+ - [Model details](#model-details)
9
+ - [Installation](#installation)
10
+ - [Intended use](#intended-use)
11
+ - [Performance Factors](#performance-factors)
12
+ - [Training data](#training-data)
13
+ - [Ethical considerations](#ethical-considerations)
14
+ - [Troubleshooting issues](#troubleshooting-issues)
15
+
16
+ ## Model details
17
+
18
+ - **Maintainer:** [Vox Communis](https://osf.io/t957v/)
19
+ - **Language:** [Basque](https://en.wikipedia.org/wiki/Basque_language)
20
+ - **Dialect:** N/A
21
+ - **Phone set:** [XPF](https://github.com/CohenPr-XPF/XPF)
22
+ - **Model type:** `Acoustic model`
23
+ - **Features:** `MFCC`
24
+ - **Architecture:** `gmm-hmm`
25
+ - **Model version:** `v2.0.0`
26
+ - **Trained date:** `02-11-2022`
27
+ - **Compatible MFA version:** `v2.0.0`
28
+ - **License:** [CC-0](https://creativecommons.org/publicdomain/zero/1.0/)
29
+ - **Citation:**
30
+
31
+ ```bibtex
32
+ @misc{Ahn_Chodroff_2022,
33
+ author={Ahn, Emily and Chodroff, Eleanor},
34
+ title={VoxCommunis Corpus},
35
+ address={\url{https://osf.io/t957v}},
36
+ publisher={OSF},
37
+ year={2022},
38
+ month={Jan}
39
+ }
40
+ ```
41
+
42
+ - If you have comments or questions about this model, you can check [previous MFA model discussion posts](https://github.com/MontrealCorpusTools/mfa-models/discussions?discussions_q=Basque+CV+acoustic+model+v2.0.0) or create [a new one](https://github.com/MontrealCorpusTools/mfa-models/discussions/new).
43
+
44
+ ## Installation
45
+
46
+ Install from the [MFA command line](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/models/index.html):
47
+
48
+ ```
49
+ mfa model download acoustic basque_cv
50
+ ```
51
+
52
+ Or download from [the release page](https://github.com/MontrealCorpusTools/mfa-models/releases/tag/acoustic-basque_cv-v2.0.0).
53
+
54
+ ## Intended use
55
+
56
+ This model is intended for forced alignment of [Basque](https://en.wikipedia.org/wiki/Basque_language) transcripts.
57
+
58
+ This model uses the [XPF](https://github.com/CohenPr-XPF/XPF) phone set for Basque, and was trained with the pronunciation dictionaries above. Pronunciations can be added on top of the dictionary, as long as no additional phones are introduced.
59
+
60
+ ## Performance Factors
61
+
62
+ As forced alignment is a relatively well-constrained problem (given accurate transcripts), this model should be applicable to a range of recording conditions and speakers. However, please note that it was trained on read speech in low-noise environments, so as your data diverges from that, you may run into alignment issues or need to [increase the beam size of MFA](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/configuration/#configuring-specific-commands) or see other recommendations in the [troubleshooting section below](#troubleshooting-issues).
63
+
64
+ Please note as well that MFA does not use state-of-the-art ASR models for forced alignment. You may get better performance (especially on speech-to-text tasks) using other frameworks like [Coqui](https://coqui.ai/).
65
+
66
+ ## Ethical considerations
67
+
68
+ Deploying any Speech-to-Text model into any production setting has ethical implications. You should consider these implications before use.
69
+
70
+ ### Demographic Bias
71
+
72
+ You should assume every machine learning model has demographic bias unless proven otherwise. For STT models, it is often the case that transcription accuracy is better for men than it is for women. If you are using this model in production, you should acknowledge this as a potential issue.
73
+
74
+ ### Surveillance
75
+
76
+ Speech-to-Text technologies may be misused to invade the privacy of others by recording and mining information from private conversations. This kind of individual privacy is protected by law in many countries. You should not assume consent to record and analyze private speech.
77
+
78
+ ## Troubleshooting issues
79
+
80
+ Machine learning models (like this acoustic model) perform best on data that is similar to the data on which they were trained.
81
+
82
+ The primary sources of variability in forced alignment will be the applicability of the pronunciation dictionary and how similar the speech, demographics, and recording conditions are. If you encounter issues in alignment, there are couple of avenues to improve performance:
83
+
84
+ 1. [Increase the beam size of MFA](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/configuration/#configuring-specific-commands)
85
+
86
+ * MFA defaults to a narrow beam to ensure quick alignment and also as a way to detect potential issues in your dataset, but depending on your data, you might benefit from boosting the beam to 100 or higher.
87
+
88
+ 2. Add pronunciations to the pronunciation dictionary
89
+
90
+ * This model was trained a particular dialect/style, and so adding pronunciations more representative of the variety spoken in your dataset will help alignment.
91
+
92
+ 3. Check the quality of your data
93
+
94
+ * MFA includes a [validator utility](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/data_validation.html), which aims to detect issues in the dataset.
95
+ * Use MFA's [anchor utility](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/workflows/anchor.html) to visually inspect your data as MFA sees it and correct issues in transcription or OOV items.
96
+
97
+ 4. Adapt the model to your data
98
+
99
+ * MFA has an [adaptation command](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/workflows/adapt_acoustic_model.html) to adapt some of the model to your data based on an initial alignment, and then run another alignment with the adapted model.
100
+
101
+ ## Training data
102
+
103
+ This model was trained on the following corpora:
104
+
105
+
106
+
107
+ * [Common Voice Basque](../../../../corpus/basque/common_voice_basque/7.0/README.md):
108
+ * **Hours:** `91.00`
109
+ * **Speakers:** `842`
110
+ * **Utterances:** `63,916`
acoustic/basque/cv/v2.0.0/meta.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "name": "basque_cv",
3
+ "language": "Basque",
4
+ "dialect": "",
5
+ "phone_set": "CV",
6
+ "version": "2.0.0",
7
+ "maintainer": "[Vox Communis](https://osf.io/t957v/)",
8
+ "citation": "@misc{Ahn_Chodroff_2022,\n\tauthor={Ahn, Emily and Chodroff, Eleanor},\n\ttitle={VoxCommunis Corpus},\n\taddress={\\url{https://osf.io/t957v}},\n\tpublisher={OSF},\n\tyear={2022}, \n\tmonth={Jan}\n}",
9
+ "license": "CC-0",
10
+ "license_link": "[CC-0](https://creativecommons.org/publicdomain/zero/1.0/)",
11
+ "architecture": "gmm-hmm",
12
+ "features": "MFCC",
13
+ "evaluation": {},
14
+ "decode": {},
15
+ "train_date": "02-11-2022"
16
+ }
acoustic/belarusian/cv/v2.0.0/README.md ADDED
@@ -0,0 +1,110 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+
2
+ # Belarusian CV acoustic model v2.0.0
3
+
4
+ [Link to documentation on mfa-models](https://mfa-models.readthedocs.io/en/main/acoustic/belarusian_cv.html)
5
+
6
+ Jump to section:
7
+
8
+ - [Model details](#model-details)
9
+ - [Installation](#installation)
10
+ - [Intended use](#intended-use)
11
+ - [Performance Factors](#performance-factors)
12
+ - [Training data](#training-data)
13
+ - [Ethical considerations](#ethical-considerations)
14
+ - [Troubleshooting issues](#troubleshooting-issues)
15
+
16
+ ## Model details
17
+
18
+ - **Maintainer:** [Vox Communis](https://osf.io/t957v/)
19
+ - **Language:** [Belarusian](https://en.wikipedia.org/wiki/Belarusian_language)
20
+ - **Dialect:** N/A
21
+ - **Phone set:** [XPF](https://github.com/CohenPr-XPF/XPF)
22
+ - **Model type:** `Acoustic model`
23
+ - **Features:** `MFCC`
24
+ - **Architecture:** `gmm-hmm`
25
+ - **Model version:** `v2.0.0`
26
+ - **Trained date:** `02-11-2022`
27
+ - **Compatible MFA version:** `v2.0.0`
28
+ - **License:** [CC-0](https://creativecommons.org/publicdomain/zero/1.0/)
29
+ - **Citation:**
30
+
31
+ ```bibtex
32
+ @misc{Ahn_Chodroff_2022,
33
+ author={Ahn, Emily and Chodroff, Eleanor},
34
+ title={VoxCommunis Corpus},
35
+ address={\url{https://osf.io/t957v}},
36
+ publisher={OSF},
37
+ year={2022},
38
+ month={Jan}
39
+ }
40
+ ```
41
+
42
+ - If you have comments or questions about this model, you can check [previous MFA model discussion posts](https://github.com/MontrealCorpusTools/mfa-models/discussions?discussions_q=Belarusian+CV+acoustic+model+v2.0.0) or create [a new one](https://github.com/MontrealCorpusTools/mfa-models/discussions/new).
43
+
44
+ ## Installation
45
+
46
+ Install from the [MFA command line](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/models/index.html):
47
+
48
+ ```
49
+ mfa model download acoustic belarusian_cv
50
+ ```
51
+
52
+ Or download from [the release page](https://github.com/MontrealCorpusTools/mfa-models/releases/tag/acoustic-belarusian_cv-v2.0.0).
53
+
54
+ ## Intended use
55
+
56
+ This model is intended for forced alignment of [Belarusian](https://en.wikipedia.org/wiki/Belarusian_language) transcripts.
57
+
58
+ This model uses the [XPF](https://github.com/CohenPr-XPF/XPF) phone set for Belarusian, and was trained with the pronunciation dictionaries above. Pronunciations can be added on top of the dictionary, as long as no additional phones are introduced.
59
+
60
+ ## Performance Factors
61
+
62
+ As forced alignment is a relatively well-constrained problem (given accurate transcripts), this model should be applicable to a range of recording conditions and speakers. However, please note that it was trained on read speech in low-noise environments, so as your data diverges from that, you may run into alignment issues or need to [increase the beam size of MFA](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/configuration/#configuring-specific-commands) or see other recommendations in the [troubleshooting section below](#troubleshooting-issues).
63
+
64
+ Please note as well that MFA does not use state-of-the-art ASR models for forced alignment. You may get better performance (especially on speech-to-text tasks) using other frameworks like [Coqui](https://coqui.ai/).
65
+
66
+ ## Ethical considerations
67
+
68
+ Deploying any Speech-to-Text model into any production setting has ethical implications. You should consider these implications before use.
69
+
70
+ ### Demographic Bias
71
+
72
+ You should assume every machine learning model has demographic bias unless proven otherwise. For STT models, it is often the case that transcription accuracy is better for men than it is for women. If you are using this model in production, you should acknowledge this as a potential issue.
73
+
74
+ ### Surveillance
75
+
76
+ Speech-to-Text technologies may be misused to invade the privacy of others by recording and mining information from private conversations. This kind of individual privacy is protected by law in many countries. You should not assume consent to record and analyze private speech.
77
+
78
+ ## Troubleshooting issues
79
+
80
+ Machine learning models (like this acoustic model) perform best on data that is similar to the data on which they were trained.
81
+
82
+ The primary sources of variability in forced alignment will be the applicability of the pronunciation dictionary and how similar the speech, demographics, and recording conditions are. If you encounter issues in alignment, there are couple of avenues to improve performance:
83
+
84
+ 1. [Increase the beam size of MFA](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/configuration/#configuring-specific-commands)
85
+
86
+ * MFA defaults to a narrow beam to ensure quick alignment and also as a way to detect potential issues in your dataset, but depending on your data, you might benefit from boosting the beam to 100 or higher.
87
+
88
+ 2. Add pronunciations to the pronunciation dictionary
89
+
90
+ * This model was trained a particular dialect/style, and so adding pronunciations more representative of the variety spoken in your dataset will help alignment.
91
+
92
+ 3. Check the quality of your data
93
+
94
+ * MFA includes a [validator utility](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/data_validation.html), which aims to detect issues in the dataset.
95
+ * Use MFA's [anchor utility](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/workflows/anchor.html) to visually inspect your data as MFA sees it and correct issues in transcription or OOV items.
96
+
97
+ 4. Adapt the model to your data
98
+
99
+ * MFA has an [adaptation command](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/workflows/adapt_acoustic_model.html) to adapt some of the model to your data based on an initial alignment, and then run another alignment with the adapted model.
100
+
101
+ ## Training data
102
+
103
+ This model was trained on the following corpora:
104
+
105
+
106
+
107
+ * [Common Voice Belarusian](../../../../corpus/belarusian/common_voice_belarusian/7.0/README.md):
108
+ * **Hours:** `91.00`
109
+ * **Speakers:** `3,620`
110
+ * **Utterances:** `182,840`
acoustic/belarusian/cv/v2.0.0/meta.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "name": "belarusian_cv",
3
+ "language": "Belarusian",
4
+ "dialect": "",
5
+ "phone_set": "CV",
6
+ "version": "2.0.0",
7
+ "maintainer": "[Vox Communis](https://osf.io/t957v/)",
8
+ "citation": "@misc{Ahn_Chodroff_2022,\n\tauthor={Ahn, Emily and Chodroff, Eleanor},\n\ttitle={VoxCommunis Corpus},\n\taddress={\\url{https://osf.io/t957v}},\n\tpublisher={OSF},\n\tyear={2022}, \n\tmonth={Jan}\n}",
9
+ "license": "CC-0",
10
+ "license_link": "[CC-0](https://creativecommons.org/publicdomain/zero/1.0/)",
11
+ "architecture": "gmm-hmm",
12
+ "features": "MFCC",
13
+ "evaluation": {},
14
+ "decode": {},
15
+ "train_date": "02-11-2022"
16
+ }
acoustic/bulgarian.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d08c109d1bc790542533bc4d8eb4422656bce9e75832fa292c6a718df0481a60
3
+ size 14665607
acoustic/bulgarian/cv/v2.0.0/README.md ADDED
@@ -0,0 +1,110 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+
2
+ # Bulgarian CV acoustic model v2.0.0
3
+
4
+ [Link to documentation on mfa-models](https://mfa-models.readthedocs.io/en/main/acoustic/bulgarian_cv.html)
5
+
6
+ Jump to section:
7
+
8
+ - [Model details](#model-details)
9
+ - [Installation](#installation)
10
+ - [Intended use](#intended-use)
11
+ - [Performance Factors](#performance-factors)
12
+ - [Training data](#training-data)
13
+ - [Ethical considerations](#ethical-considerations)
14
+ - [Troubleshooting issues](#troubleshooting-issues)
15
+
16
+ ## Model details
17
+
18
+ - **Maintainer:** [Vox Communis](https://osf.io/t957v/)
19
+ - **Language:** [Bulgarian](https://en.wikipedia.org/wiki/Bulgarian_language)
20
+ - **Dialect:** N/A
21
+ - **Phone set:** [XPF](https://github.com/CohenPr-XPF/XPF)
22
+ - **Model type:** `Acoustic model`
23
+ - **Features:** `MFCC`
24
+ - **Architecture:** `gmm-hmm`
25
+ - **Model version:** `v2.0.0`
26
+ - **Trained date:** `02-11-2022`
27
+ - **Compatible MFA version:** `v2.0.0`
28
+ - **License:** [CC-0](https://creativecommons.org/publicdomain/zero/1.0/)
29
+ - **Citation:**
30
+
31
+ ```bibtex
32
+ @misc{Ahn_Chodroff_2022,
33
+ author={Ahn, Emily and Chodroff, Eleanor},
34
+ title={VoxCommunis Corpus},
35
+ address={\url{https://osf.io/t957v}},
36
+ publisher={OSF},
37
+ year={2022},
38
+ month={Jan}
39
+ }
40
+ ```
41
+
42
+ - If you have comments or questions about this model, you can check [previous MFA model discussion posts](https://github.com/MontrealCorpusTools/mfa-models/discussions?discussions_q=Bulgarian+CV+acoustic+model+v2.0.0) or create [a new one](https://github.com/MontrealCorpusTools/mfa-models/discussions/new).
43
+
44
+ ## Installation
45
+
46
+ Install from the [MFA command line](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/models/index.html):
47
+
48
+ ```
49
+ mfa model download acoustic bulgarian_cv
50
+ ```
51
+
52
+ Or download from [the release page](https://github.com/MontrealCorpusTools/mfa-models/releases/tag/acoustic-bulgarian_cv-v2.0.0).
53
+
54
+ ## Intended use
55
+
56
+ This model is intended for forced alignment of [Bulgarian](https://en.wikipedia.org/wiki/Bulgarian_language) transcripts.
57
+
58
+ This model uses the [XPF](https://github.com/CohenPr-XPF/XPF) phone set for Bulgarian, and was trained with the pronunciation dictionaries above. Pronunciations can be added on top of the dictionary, as long as no additional phones are introduced.
59
+
60
+ ## Performance Factors
61
+
62
+ As forced alignment is a relatively well-constrained problem (given accurate transcripts), this model should be applicable to a range of recording conditions and speakers. However, please note that it was trained on read speech in low-noise environments, so as your data diverges from that, you may run into alignment issues or need to [increase the beam size of MFA](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/configuration/#configuring-specific-commands) or see other recommendations in the [troubleshooting section below](#troubleshooting-issues).
63
+
64
+ Please note as well that MFA does not use state-of-the-art ASR models for forced alignment. You may get better performance (especially on speech-to-text tasks) using other frameworks like [Coqui](https://coqui.ai/).
65
+
66
+ ## Ethical considerations
67
+
68
+ Deploying any Speech-to-Text model into any production setting has ethical implications. You should consider these implications before use.
69
+
70
+ ### Demographic Bias
71
+
72
+ You should assume every machine learning model has demographic bias unless proven otherwise. For STT models, it is often the case that transcription accuracy is better for men than it is for women. If you are using this model in production, you should acknowledge this as a potential issue.
73
+
74
+ ### Surveillance
75
+
76
+ Speech-to-Text technologies may be misused to invade the privacy of others by recording and mining information from private conversations. This kind of individual privacy is protected by law in many countries. You should not assume consent to record and analyze private speech.
77
+
78
+ ## Troubleshooting issues
79
+
80
+ Machine learning models (like this acoustic model) perform best on data that is similar to the data on which they were trained.
81
+
82
+ The primary sources of variability in forced alignment will be the applicability of the pronunciation dictionary and how similar the speech, demographics, and recording conditions are. If you encounter issues in alignment, there are couple of avenues to improve performance:
83
+
84
+ 1. [Increase the beam size of MFA](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/configuration/#configuring-specific-commands)
85
+
86
+ * MFA defaults to a narrow beam to ensure quick alignment and also as a way to detect potential issues in your dataset, but depending on your data, you might benefit from boosting the beam to 100 or higher.
87
+
88
+ 2. Add pronunciations to the pronunciation dictionary
89
+
90
+ * This model was trained a particular dialect/style, and so adding pronunciations more representative of the variety spoken in your dataset will help alignment.
91
+
92
+ 3. Check the quality of your data
93
+
94
+ * MFA includes a [validator utility](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/data_validation.html), which aims to detect issues in the dataset.
95
+ * Use MFA's [anchor utility](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/workflows/anchor.html) to visually inspect your data as MFA sees it and correct issues in transcription or OOV items.
96
+
97
+ 4. Adapt the model to your data
98
+
99
+ * MFA has an [adaptation command](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/workflows/adapt_acoustic_model.html) to adapt some of the model to your data based on an initial alignment, and then run another alignment with the adapted model.
100
+
101
+ ## Training data
102
+
103
+ This model was trained on the following corpora:
104
+
105
+
106
+
107
+ * [Common Voice Bulgarian](../../../../corpus/bulgarian/common_voice_bulgarian/7.0/README.md):
108
+ * **Hours:** `5.00`
109
+ * **Speakers:** `35`
110
+ * **Utterances:** `3,459`
acoustic/bulgarian/cv/v2.0.0/meta.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "name": "bulgarian_cv",
3
+ "language": "Bulgarian",
4
+ "dialect": "",
5
+ "phone_set": "CV",
6
+ "version": "2.0.0",
7
+ "maintainer": "[Vox Communis](https://osf.io/t957v/)",
8
+ "citation": "@misc{Ahn_Chodroff_2022,\n\tauthor={Ahn, Emily and Chodroff, Eleanor},\n\ttitle={VoxCommunis Corpus},\n\taddress={\\url{https://osf.io/t957v}},\n\tpublisher={OSF},\n\tyear={2022}, \n\tmonth={Jan}\n}",
9
+ "license": "CC-0",
10
+ "license_link": "[CC-0](https://creativecommons.org/publicdomain/zero/1.0/)",
11
+ "architecture": "gmm-hmm",
12
+ "features": "MFCC",
13
+ "evaluation": {},
14
+ "decode": {},
15
+ "train_date": "02-11-2022"
16
+ }
acoustic/bulgarian/mfa/v2.0.0/LICENSE ADDED
@@ -0,0 +1,395 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ Attribution 4.0 International
2
+
3
+ =======================================================================
4
+
5
+ Creative Commons Corporation ("Creative Commons") is not a law firm and
6
+ does not provide legal services or legal advice. Distribution of
7
+ Creative Commons public licenses does not create a lawyer-client or
8
+ other relationship. Creative Commons makes its licenses and related
9
+ information available on an "as-is" basis. Creative Commons gives no
10
+ warranties regarding its licenses, any material licensed under their
11
+ terms and conditions, or any related information. Creative Commons
12
+ disclaims all liability for damages resulting from their use to the
13
+ fullest extent possible.
14
+
15
+ Using Creative Commons Public Licenses
16
+
17
+ Creative Commons public licenses provide a standard set of terms and
18
+ conditions that creators and other rights holders may use to share
19
+ original works of authorship and other material subject to copyright
20
+ and certain other rights specified in the public license below. The
21
+ following considerations are for informational purposes only, are not
22
+ exhaustive, and do not form part of our licenses.
23
+
24
+ Considerations for licensors: Our public licenses are
25
+ intended for use by those authorized to give the public
26
+ permission to use material in ways otherwise restricted by
27
+ copyright and certain other rights. Our licenses are
28
+ irrevocable. Licensors should read and understand the terms
29
+ and conditions of the license they choose before applying it.
30
+ Licensors should also secure all rights necessary before
31
+ applying our licenses so that the public can reuse the
32
+ material as expected. Licensors should clearly mark any
33
+ material not subject to the license. This includes other CC-
34
+ licensed material, or material used under an exception or
35
+ limitation to copyright. More considerations for licensors:
36
+ wiki.creativecommons.org/Considerations_for_licensors
37
+
38
+ Considerations for the public: By using one of our public
39
+ licenses, a licensor grants the public permission to use the
40
+ licensed material under specified terms and conditions. If
41
+ the licensor's permission is not necessary for any reason--for
42
+ example, because of any applicable exception or limitation to
43
+ copyright--then that use is not regulated by the license. Our
44
+ licenses grant only permissions under copyright and certain
45
+ other rights that a licensor has authority to grant. Use of
46
+ the licensed material may still be restricted for other
47
+ reasons, including because others have copyright or other
48
+ rights in the material. A licensor may make special requests,
49
+ such as asking that all changes be marked or described.
50
+ Although not required by our licenses, you are encouraged to
51
+ respect those requests where reasonable. More_considerations
52
+ for the public:
53
+ wiki.creativecommons.org/Considerations_for_licensees
54
+
55
+ =======================================================================
56
+
57
+ Creative Commons Attribution 4.0 International Public License
58
+
59
+ By exercising the Licensed Rights (defined below), You accept and agree
60
+ to be bound by the terms and conditions of this Creative Commons
61
+ Attribution 4.0 International Public License ("Public License"). To the
62
+ extent this Public License may be interpreted as a contract, You are
63
+ granted the Licensed Rights in consideration of Your acceptance of
64
+ these terms and conditions, and the Licensor grants You such rights in
65
+ consideration of benefits the Licensor receives from making the
66
+ Licensed Material available under these terms and conditions.
67
+
68
+
69
+ Section 1 -- Definitions.
70
+
71
+ a. Adapted Material means material subject to Copyright and Similar
72
+ Rights that is derived from or based upon the Licensed Material
73
+ and in which the Licensed Material is translated, altered,
74
+ arranged, transformed, or otherwise modified in a manner requiring
75
+ permission under the Copyright and Similar Rights held by the
76
+ Licensor. For purposes of this Public License, where the Licensed
77
+ Material is a musical work, performance, or sound recording,
78
+ Adapted Material is always produced where the Licensed Material is
79
+ synched in timed relation with a moving image.
80
+
81
+ b. Adapter's License means the license You apply to Your Copyright
82
+ and Similar Rights in Your contributions to Adapted Material in
83
+ accordance with the terms and conditions of this Public License.
84
+
85
+ c. Copyright and Similar Rights means copyright and/or similar rights
86
+ closely related to copyright including, without limitation,
87
+ performance, broadcast, sound recording, and Sui Generis Database
88
+ Rights, without regard to how the rights are labeled or
89
+ categorized. For purposes of this Public License, the rights
90
+ specified in Section 2(b)(1)-(2) are not Copyright and Similar
91
+ Rights.
92
+
93
+ d. Effective Technological Measures means those measures that, in the
94
+ absence of proper authority, may not be circumvented under laws
95
+ fulfilling obligations under Article 11 of the WIPO Copyright
96
+ Treaty adopted on December 20, 1996, and/or similar international
97
+ agreements.
98
+
99
+ e. Exceptions and Limitations means fair use, fair dealing, and/or
100
+ any other exception or limitation to Copyright and Similar Rights
101
+ that applies to Your use of the Licensed Material.
102
+
103
+ f. Licensed Material means the artistic or literary work, database,
104
+ or other material to which the Licensor applied this Public
105
+ License.
106
+
107
+ g. Licensed Rights means the rights granted to You subject to the
108
+ terms and conditions of this Public License, which are limited to
109
+ all Copyright and Similar Rights that apply to Your use of the
110
+ Licensed Material and that the Licensor has authority to license.
111
+
112
+ h. Licensor means the individual(s) or entity(ies) granting rights
113
+ under this Public License.
114
+
115
+ i. Share means to provide material to the public by any means or
116
+ process that requires permission under the Licensed Rights, such
117
+ as reproduction, public display, public performance, distribution,
118
+ dissemination, communication, or importation, and to make material
119
+ available to the public including in ways that members of the
120
+ public may access the material from a place and at a time
121
+ individually chosen by them.
122
+
123
+ j. Sui Generis Database Rights means rights other than copyright
124
+ resulting from Directive 96/9/EC of the European Parliament and of
125
+ the Council of 11 March 1996 on the legal protection of databases,
126
+ as amended and/or succeeded, as well as other essentially
127
+ equivalent rights anywhere in the world.
128
+
129
+ k. You means the individual or entity exercising the Licensed Rights
130
+ under this Public License. Your has a corresponding meaning.
131
+
132
+
133
+ Section 2 -- Scope.
134
+
135
+ a. License grant.
136
+
137
+ 1. Subject to the terms and conditions of this Public License,
138
+ the Licensor hereby grants You a worldwide, royalty-free,
139
+ non-sublicensable, non-exclusive, irrevocable license to
140
+ exercise the Licensed Rights in the Licensed Material to:
141
+
142
+ a. reproduce and Share the Licensed Material, in whole or
143
+ in part; and
144
+
145
+ b. produce, reproduce, and Share Adapted Material.
146
+
147
+ 2. Exceptions and Limitations. For the avoidance of doubt, where
148
+ Exceptions and Limitations apply to Your use, this Public
149
+ License does not apply, and You do not need to comply with
150
+ its terms and conditions.
151
+
152
+ 3. Term. The term of this Public License is specified in Section
153
+ 6(a).
154
+
155
+ 4. Media and formats; technical modifications allowed. The
156
+ Licensor authorizes You to exercise the Licensed Rights in
157
+ all media and formats whether now known or hereafter created,
158
+ and to make technical modifications necessary to do so. The
159
+ Licensor waives and/or agrees not to assert any right or
160
+ authority to forbid You from making technical modifications
161
+ necessary to exercise the Licensed Rights, including
162
+ technical modifications necessary to circumvent Effective
163
+ Technological Measures. For purposes of this Public License,
164
+ simply making modifications authorized by this Section 2(a)
165
+ (4) never produces Adapted Material.
166
+
167
+ 5. Downstream recipients.
168
+
169
+ a. Offer from the Licensor -- Licensed Material. Every
170
+ recipient of the Licensed Material automatically
171
+ receives an offer from the Licensor to exercise the
172
+ Licensed Rights under the terms and conditions of this
173
+ Public License.
174
+
175
+ b. No downstream restrictions. You may not offer or impose
176
+ any additional or different terms or conditions on, or
177
+ apply any Effective Technological Measures to, the
178
+ Licensed Material if doing so restricts exercise of the
179
+ Licensed Rights by any recipient of the Licensed
180
+ Material.
181
+
182
+ 6. No endorsement. Nothing in this Public License constitutes or
183
+ may be construed as permission to assert or imply that You
184
+ are, or that Your use of the Licensed Material is, connected
185
+ with, or sponsored, endorsed, or granted official status by,
186
+ the Licensor or others designated to receive attribution as
187
+ provided in Section 3(a)(1)(A)(i).
188
+
189
+ b. Other rights.
190
+
191
+ 1. Moral rights, such as the right of integrity, are not
192
+ licensed under this Public License, nor are publicity,
193
+ privacy, and/or other similar personality rights; however, to
194
+ the extent possible, the Licensor waives and/or agrees not to
195
+ assert any such rights held by the Licensor to the limited
196
+ extent necessary to allow You to exercise the Licensed
197
+ Rights, but not otherwise.
198
+
199
+ 2. Patent and trademark rights are not licensed under this
200
+ Public License.
201
+
202
+ 3. To the extent possible, the Licensor waives any right to
203
+ collect royalties from You for the exercise of the Licensed
204
+ Rights, whether directly or through a collecting society
205
+ under any voluntary or waivable statutory or compulsory
206
+ licensing scheme. In all other cases the Licensor expressly
207
+ reserves any right to collect such royalties.
208
+
209
+
210
+ Section 3 -- License Conditions.
211
+
212
+ Your exercise of the Licensed Rights is expressly made subject to the
213
+ following conditions.
214
+
215
+ a. Attribution.
216
+
217
+ 1. If You Share the Licensed Material (including in modified
218
+ form), You must:
219
+
220
+ a. retain the following if it is supplied by the Licensor
221
+ with the Licensed Material:
222
+
223
+ i. identification of the creator(s) of the Licensed
224
+ Material and any others designated to receive
225
+ attribution, in any reasonable manner requested by
226
+ the Licensor (including by pseudonym if
227
+ designated);
228
+
229
+ ii. a copyright notice;
230
+
231
+ iii. a notice that refers to this Public License;
232
+
233
+ iv. a notice that refers to the disclaimer of
234
+ warranties;
235
+
236
+ v. a URI or hyperlink to the Licensed Material to the
237
+ extent reasonably practicable;
238
+
239
+ b. indicate if You modified the Licensed Material and
240
+ retain an indication of any previous modifications; and
241
+
242
+ c. indicate the Licensed Material is licensed under this
243
+ Public License, and include the text of, or the URI or
244
+ hyperlink to, this Public License.
245
+
246
+ 2. You may satisfy the conditions in Section 3(a)(1) in any
247
+ reasonable manner based on the medium, means, and context in
248
+ which You Share the Licensed Material. For example, it may be
249
+ reasonable to satisfy the conditions by providing a URI or
250
+ hyperlink to a resource that includes the required
251
+ information.
252
+
253
+ 3. If requested by the Licensor, You must remove any of the
254
+ information required by Section 3(a)(1)(A) to the extent
255
+ reasonably practicable.
256
+
257
+ 4. If You Share Adapted Material You produce, the Adapter's
258
+ License You apply must not prevent recipients of the Adapted
259
+ Material from complying with this Public License.
260
+
261
+
262
+ Section 4 -- Sui Generis Database Rights.
263
+
264
+ Where the Licensed Rights include Sui Generis Database Rights that
265
+ apply to Your use of the Licensed Material:
266
+
267
+ a. for the avoidance of doubt, Section 2(a)(1) grants You the right
268
+ to extract, reuse, reproduce, and Share all or a substantial
269
+ portion of the contents of the database;
270
+
271
+ b. if You include all or a substantial portion of the database
272
+ contents in a database in which You have Sui Generis Database
273
+ Rights, then the database in which You have Sui Generis Database
274
+ Rights (but not its individual contents) is Adapted Material; and
275
+
276
+ c. You must comply with the conditions in Section 3(a) if You Share
277
+ all or a substantial portion of the contents of the database.
278
+
279
+ For the avoidance of doubt, this Section 4 supplements and does not
280
+ replace Your obligations under this Public License where the Licensed
281
+ Rights include other Copyright and Similar Rights.
282
+
283
+
284
+ Section 5 -- Disclaimer of Warranties and Limitation of Liability.
285
+
286
+ a. UNLESS OTHERWISE SEPARATELY UNDERTAKEN BY THE LICENSOR, TO THE
287
+ EXTENT POSSIBLE, THE LICENSOR OFFERS THE LICENSED MATERIAL AS-IS
288
+ AND AS-AVAILABLE, AND MAKES NO REPRESENTATIONS OR WARRANTIES OF
289
+ ANY KIND CONCERNING THE LICENSED MATERIAL, WHETHER EXPRESS,
290
+ IMPLIED, STATUTORY, OR OTHER. THIS INCLUDES, WITHOUT LIMITATION,
291
+ WARRANTIES OF TITLE, MERCHANTABILITY, FITNESS FOR A PARTICULAR
292
+ PURPOSE, NON-INFRINGEMENT, ABSENCE OF LATENT OR OTHER DEFECTS,
293
+ ACCURACY, OR THE PRESENCE OR ABSENCE OF ERRORS, WHETHER OR NOT
294
+ KNOWN OR DISCOVERABLE. WHERE DISCLAIMERS OF WARRANTIES ARE NOT
295
+ ALLOWED IN FULL OR IN PART, THIS DISCLAIMER MAY NOT APPLY TO YOU.
296
+
297
+ b. TO THE EXTENT POSSIBLE, IN NO EVENT WILL THE LICENSOR BE LIABLE
298
+ TO YOU ON ANY LEGAL THEORY (INCLUDING, WITHOUT LIMITATION,
299
+ NEGLIGENCE) OR OTHERWISE FOR ANY DIRECT, SPECIAL, INDIRECT,
300
+ INCIDENTAL, CONSEQUENTIAL, PUNITIVE, EXEMPLARY, OR OTHER LOSSES,
301
+ COSTS, EXPENSES, OR DAMAGES ARISING OUT OF THIS PUBLIC LICENSE OR
302
+ USE OF THE LICENSED MATERIAL, EVEN IF THE LICENSOR HAS BEEN
303
+ ADVISED OF THE POSSIBILITY OF SUCH LOSSES, COSTS, EXPENSES, OR
304
+ DAMAGES. WHERE A LIMITATION OF LIABILITY IS NOT ALLOWED IN FULL OR
305
+ IN PART, THIS LIMITATION MAY NOT APPLY TO YOU.
306
+
307
+ c. The disclaimer of warranties and limitation of liability provided
308
+ above shall be interpreted in a manner that, to the extent
309
+ possible, most closely approximates an absolute disclaimer and
310
+ waiver of all liability.
311
+
312
+
313
+ Section 6 -- Term and Termination.
314
+
315
+ a. This Public License applies for the term of the Copyright and
316
+ Similar Rights licensed here. However, if You fail to comply with
317
+ this Public License, then Your rights under this Public License
318
+ terminate automatically.
319
+
320
+ b. Where Your right to use the Licensed Material has terminated under
321
+ Section 6(a), it reinstates:
322
+
323
+ 1. automatically as of the date the violation is cured, provided
324
+ it is cured within 30 days of Your discovery of the
325
+ violation; or
326
+
327
+ 2. upon express reinstatement by the Licensor.
328
+
329
+ For the avoidance of doubt, this Section 6(b) does not affect any
330
+ right the Licensor may have to seek remedies for Your violations
331
+ of this Public License.
332
+
333
+ c. For the avoidance of doubt, the Licensor may also offer the
334
+ Licensed Material under separate terms or conditions or stop
335
+ distributing the Licensed Material at any time; however, doing so
336
+ will not terminate this Public License.
337
+
338
+ d. Sections 1, 5, 6, 7, and 8 survive termination of this Public
339
+ License.
340
+
341
+
342
+ Section 7 -- Other Terms and Conditions.
343
+
344
+ a. The Licensor shall not be bound by any additional or different
345
+ terms or conditions communicated by You unless expressly agreed.
346
+
347
+ b. Any arrangements, understandings, or agreements regarding the
348
+ Licensed Material not stated herein are separate from and
349
+ independent of the terms and conditions of this Public License.
350
+
351
+
352
+ Section 8 -- Interpretation.
353
+
354
+ a. For the avoidance of doubt, this Public License does not, and
355
+ shall not be interpreted to, reduce, limit, restrict, or impose
356
+ conditions on any use of the Licensed Material that could lawfully
357
+ be made without permission under this Public License.
358
+
359
+ b. To the extent possible, if any provision of this Public License is
360
+ deemed unenforceable, it shall be automatically reformed to the
361
+ minimum extent necessary to make it enforceable. If the provision
362
+ cannot be reformed, it shall be severed from this Public License
363
+ without affecting the enforceability of the remaining terms and
364
+ conditions.
365
+
366
+ c. No term or condition of this Public License will be waived and no
367
+ failure to comply consented to unless expressly agreed to by the
368
+ Licensor.
369
+
370
+ d. Nothing in this Public License constitutes or may be interpreted
371
+ as a limitation upon, or waiver of, any privileges and immunities
372
+ that apply to the Licensor or You, including from the legal
373
+ processes of any jurisdiction or authority.
374
+
375
+
376
+ =======================================================================
377
+
378
+ Creative Commons is not a party to its public
379
+ licenses. Notwithstanding, Creative Commons may elect to apply one of
380
+ its public licenses to material it publishes and in those instances
381
+ will be considered the “Licensor.” The text of the Creative Commons
382
+ public licenses is dedicated to the public domain under the CC0 Public
383
+ Domain Dedication. Except for the limited purpose of indicating that
384
+ material is shared under a Creative Commons public license or as
385
+ otherwise permitted by the Creative Commons policies published at
386
+ creativecommons.org/policies, Creative Commons does not authorize the
387
+ use of the trademark "Creative Commons" or any other trademark or logo
388
+ of Creative Commons without its prior written consent including,
389
+ without limitation, in connection with any unauthorized modifications
390
+ to any of its public licenses or any other arrangements,
391
+ understandings, or agreements concerning use of licensed material. For
392
+ the avoidance of doubt, this paragraph does not form part of the
393
+ public licenses.
394
+
395
+ Creative Commons may be contacted at creativecommons.org.
acoustic/bulgarian/mfa/v2.0.0/README.md ADDED
@@ -0,0 +1,128 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+
2
+ # Bulgarian MFA acoustic model v2.0.0
3
+
4
+ [Link to documentation on mfa-models](https://mfa-models.readthedocs.io/en/main/acoustic/bulgarian_mfa.html)
5
+
6
+ Jump to section:
7
+
8
+ - [Model details](#model-details)
9
+ - [Installation](#installation)
10
+ - [Intended use](#intended-use)
11
+ - [Performance Factors](#performance-factors)
12
+ - [Metrics](#metrics)
13
+ - [Ethical considerations](#ethical-considerations)
14
+ - [Troubleshooting issues](#troubleshooting-issues)
15
+ - [Training data](#training-data)
16
+ - [Evaluation data](#evaluation-data)
17
+
18
+ ## Model details
19
+
20
+ - **Maintainer:** [Montreal Forced Aligner](https://montreal-forced-aligner.readthedocs.io/)
21
+ - **Language:** [Bulgarian](https://en.wikipedia.org/wiki/Bulgarian_language)
22
+ - **Dialect:** N/A
23
+ - **Phone set:** [MFA](https://mfa-models.readthedocs.io/en/refactor/mfa_phone_set.html#bulgarian)
24
+ - **Model type:** `Acoustic`
25
+ - **Features:** `MFCC + pitch`
26
+ - **Architecture:** `gmm-hmm`
27
+ - **Model version:** `v2.0.0`
28
+ - **Trained date:** `2022-03-21`
29
+ - **Compatible MFA version:** `v2.0.0`
30
+ - **License:** [CC BY 4.0](https://github.com/MontrealCorpusTools/mfa-models/tree/main/acoustic/bulgarian/mfa/v2.0.0/LICENSE)
31
+ - **Citation:**
32
+
33
+ ```bibtex
34
+ @techreport{mfa_bulgarian_mfa_acoustic_2022,
35
+ author={McAuliffe, Michael and Sonderegger, Morgan},
36
+ title={Bulgarian MFA acoustic model v2.0.0},
37
+ address={\url{https://mfa-models.readthedocs.io/acoustic/Bulgarian/Bulgarian MFA acoustic model v2_0_0.html}},
38
+ year={2022},
39
+ month={Mar},
40
+ }
41
+ ```
42
+
43
+ - If you have comments or questions about this model, you can check [previous MFA model discussion posts](https://github.com/MontrealCorpusTools/mfa-models/discussions?discussions_q=Bulgarian+MFA+acoustic+model+v2.0.0) or create [a new one](https://github.com/MontrealCorpusTools/mfa-models/discussions/new).
44
+
45
+ ## Installation
46
+
47
+ Install from the [MFA command line](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/models/index.html):
48
+
49
+ ```
50
+ mfa model download acoustic bulgarian_mfa
51
+ ```
52
+
53
+ Or download from [the release page](https://github.com/MontrealCorpusTools/mfa-models/releases/tag/acoustic-bulgarian_mfa-v2.0.0).
54
+
55
+ ## Intended use
56
+
57
+ This model is intended for forced alignment of [Bulgarian](https://en.wikipedia.org/wiki/Bulgarian_language) transcripts.
58
+
59
+ This model uses the [MFA](https://mfa-models.readthedocs.io/en/refactor/mfa_phone_set.html#bulgarian) phone set for Bulgarian, and was trained with the pronunciation dictionaries above. Pronunciations can be added on top of the dictionary, as long as no additional phones are introduced.
60
+
61
+ ## Performance Factors
62
+
63
+ As forced alignment is a relatively well-constrained problem (given accurate transcripts), this model should be applicable to a range of recording conditions and speakers. However, please note that it was trained on read speech in low-noise environments, so as your data diverges from that, you may run into alignment issues or need to [increase the beam size of MFA](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/configuration/#configuring-specific-commands) or see other recommendations in the [troubleshooting section below](#troubleshooting-issues).
64
+
65
+ Please note as well that MFA does not use state-of-the-art ASR models for forced alignment. You may get better performance (especially on speech-to-text tasks) using other frameworks like [Coqui](https://coqui.ai/).
66
+
67
+ ## Metrics
68
+
69
+ Acoustic models are typically generated as one component of a larger ASR system where the metric is word error rate (WER). For forced alignment, there is typically not the same sort of gold standard measure for most languages.
70
+
71
+ As a rough approximation of the acoustic model quality, we evaluated it against the corpus it was trained on alongside a language model trained from the same data. Key caveat here is that this is not a typical WER measure on held out data, so it should not be taken as a hard measure of how well an acoustic model will generalize to your data, but rather is more of a sanity check that the training data quality was sufficiently high.
72
+
73
+ Using the pronunciation dictionary above and a language model trained on the GlobalPhone Bulgarian corpus (see the [GlobalPhone benchmark analysis page](https://montreal-forced-aligner.readthedocs.io/en/latest/benchmarks/globalphone_transcriptions.html) for more details):
74
+
75
+ - **WER:** `2.5%`
76
+ - **CER:** `0.8%`
77
+
78
+ ## Ethical considerations
79
+
80
+ Deploying any Speech-to-Text model into any production setting has ethical implications. You should consider these implications before use.
81
+
82
+ ### Demographic Bias
83
+
84
+ You should assume every machine learning model has demographic bias unless proven otherwise. For STT models, it is often the case that transcription accuracy is better for men than it is for women. If you are using this model in production, you should acknowledge this as a potential issue.
85
+
86
+ ### Surveillance
87
+
88
+ Speech-to-Text technologies may be misused to invade the privacy of others by recording and mining information from private conversations. This kind of individual privacy is protected by law in many countries. You should not assume consent to record and analyze private speech.
89
+
90
+
91
+ ## Troubleshooting issues
92
+
93
+ Machine learning models (like this acoustic model) perform best on data that is similar to the data on which they were trained.
94
+
95
+ The primary sources of variability in forced alignment will be the applicability of the pronunciation dictionary and how similar the speech, demographics, and recording conditions are. If you encounter issues in alignment, there are couple of avenues to improve performance:
96
+
97
+ 1. [Increase the beam size of MFA](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/configuration/#configuring-specific-commands)
98
+
99
+ * MFA defaults to a narrow beam to ensure quick alignment and also as a way to detect potential issues in your dataset, but depending on your data, you might benefit from boosting the beam to 100 or higher.
100
+
101
+ 2. Add pronunciations to the pronunciation dictionary
102
+
103
+ * This model was trained a particular dialect/style, and so adding pronunciations more representative of the variety spoken in your dataset will help alignment.
104
+
105
+ 3. Check the quality of your data
106
+
107
+ * MFA includes a [validator utility](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/data_validation.html), which aims to detect issues in the dataset.
108
+ * Use MFA's [anchor utility](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/workflows/anchor.html) to visually inspect your data as MFA sees it and correct issues in transcription or OOV items.
109
+
110
+ 4. Adapt the model to your data
111
+
112
+ * MFA has an [adaptation command](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/workflows/adapt_acoustic_model.html) to adapt some of the model to your data based on an initial alignment, and then run another alignment with the adapted model.
113
+
114
+ ## Training data
115
+
116
+ This model was trained on the following corpora:
117
+
118
+
119
+
120
+ * [Common Voice Bulgarian](../../../../corpus/bulgarian/common_voice_bulgarian/8.0/README.md):
121
+ * **Hours:** `7.98`
122
+ * **Speakers:** `53`
123
+ * **Utterances:** `5,387`
124
+
125
+ * [GlobalPhone Bulgarian](../../../../corpus/bulgarian/globalphone_bulgarian/3.1/README.md):
126
+ * **Hours:** `7.98`
127
+ * **Speakers:** `53`
128
+ * **Utterances:** `5,387`
acoustic/bulgarian/mfa/v2.0.0/meta.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "name": "bulgarian_mfa",
3
+ "language": "Bulgarian",
4
+ "dialect": "",
5
+ "phone_set": "MFA",
6
+ "version": "2.0.0",
7
+ "maintainer": "[Montreal Forced Aligner](https://montreal-forced-aligner.readthedocs.io/)",
8
+ "citation": "@techreport{mfa_bulgarian_mfa_acoustic_2022,\n\tauthor={McAuliffe, Michael and Sonderegger, Morgan},\n\ttitle={Bulgarian MFA acoustic model v2.0.0},\n\taddress={\\url{https://mfa-models.readthedocs.io/acoustic/Bulgarian/Bulgarian MFA acoustic model v2_0_0.html}},\n\tyear={2022},\n\tmonth={Mar},\n}",
9
+ "license": "CC BY 4.0",
10
+ "license_link": "[CC BY 4.0](https://github.com/MontrealCorpusTools/mfa-models/tree/main/acoustic/bulgarian/mfa/v2.0.0/LICENSE)",
11
+ "architecture": "gmm-hmm",
12
+ "features": "MFCC + pitch",
13
+ "evaluation": {},
14
+ "decode": {},
15
+ "train_date": "2022-03-21"
16
+ }
acoustic/bulgarian/mfa/v2.0.0a/LICENSE ADDED
@@ -0,0 +1,395 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ Attribution 4.0 International
2
+
3
+ =======================================================================
4
+
5
+ Creative Commons Corporation ("Creative Commons") is not a law firm and
6
+ does not provide legal services or legal advice. Distribution of
7
+ Creative Commons public licenses does not create a lawyer-client or
8
+ other relationship. Creative Commons makes its licenses and related
9
+ information available on an "as-is" basis. Creative Commons gives no
10
+ warranties regarding its licenses, any material licensed under their
11
+ terms and conditions, or any related information. Creative Commons
12
+ disclaims all liability for damages resulting from their use to the
13
+ fullest extent possible.
14
+
15
+ Using Creative Commons Public Licenses
16
+
17
+ Creative Commons public licenses provide a standard set of terms and
18
+ conditions that creators and other rights holders may use to share
19
+ original works of authorship and other material subject to copyright
20
+ and certain other rights specified in the public license below. The
21
+ following considerations are for informational purposes only, are not
22
+ exhaustive, and do not form part of our licenses.
23
+
24
+ Considerations for licensors: Our public licenses are
25
+ intended for use by those authorized to give the public
26
+ permission to use material in ways otherwise restricted by
27
+ copyright and certain other rights. Our licenses are
28
+ irrevocable. Licensors should read and understand the terms
29
+ and conditions of the license they choose before applying it.
30
+ Licensors should also secure all rights necessary before
31
+ applying our licenses so that the public can reuse the
32
+ material as expected. Licensors should clearly mark any
33
+ material not subject to the license. This includes other CC-
34
+ licensed material, or material used under an exception or
35
+ limitation to copyright. More considerations for licensors:
36
+ wiki.creativecommons.org/Considerations_for_licensors
37
+
38
+ Considerations for the public: By using one of our public
39
+ licenses, a licensor grants the public permission to use the
40
+ licensed material under specified terms and conditions. If
41
+ the licensor's permission is not necessary for any reason--for
42
+ example, because of any applicable exception or limitation to
43
+ copyright--then that use is not regulated by the license. Our
44
+ licenses grant only permissions under copyright and certain
45
+ other rights that a licensor has authority to grant. Use of
46
+ the licensed material may still be restricted for other
47
+ reasons, including because others have copyright or other
48
+ rights in the material. A licensor may make special requests,
49
+ such as asking that all changes be marked or described.
50
+ Although not required by our licenses, you are encouraged to
51
+ respect those requests where reasonable. More_considerations
52
+ for the public:
53
+ wiki.creativecommons.org/Considerations_for_licensees
54
+
55
+ =======================================================================
56
+
57
+ Creative Commons Attribution 4.0 International Public License
58
+
59
+ By exercising the Licensed Rights (defined below), You accept and agree
60
+ to be bound by the terms and conditions of this Creative Commons
61
+ Attribution 4.0 International Public License ("Public License"). To the
62
+ extent this Public License may be interpreted as a contract, You are
63
+ granted the Licensed Rights in consideration of Your acceptance of
64
+ these terms and conditions, and the Licensor grants You such rights in
65
+ consideration of benefits the Licensor receives from making the
66
+ Licensed Material available under these terms and conditions.
67
+
68
+
69
+ Section 1 -- Definitions.
70
+
71
+ a. Adapted Material means material subject to Copyright and Similar
72
+ Rights that is derived from or based upon the Licensed Material
73
+ and in which the Licensed Material is translated, altered,
74
+ arranged, transformed, or otherwise modified in a manner requiring
75
+ permission under the Copyright and Similar Rights held by the
76
+ Licensor. For purposes of this Public License, where the Licensed
77
+ Material is a musical work, performance, or sound recording,
78
+ Adapted Material is always produced where the Licensed Material is
79
+ synched in timed relation with a moving image.
80
+
81
+ b. Adapter's License means the license You apply to Your Copyright
82
+ and Similar Rights in Your contributions to Adapted Material in
83
+ accordance with the terms and conditions of this Public License.
84
+
85
+ c. Copyright and Similar Rights means copyright and/or similar rights
86
+ closely related to copyright including, without limitation,
87
+ performance, broadcast, sound recording, and Sui Generis Database
88
+ Rights, without regard to how the rights are labeled or
89
+ categorized. For purposes of this Public License, the rights
90
+ specified in Section 2(b)(1)-(2) are not Copyright and Similar
91
+ Rights.
92
+
93
+ d. Effective Technological Measures means those measures that, in the
94
+ absence of proper authority, may not be circumvented under laws
95
+ fulfilling obligations under Article 11 of the WIPO Copyright
96
+ Treaty adopted on December 20, 1996, and/or similar international
97
+ agreements.
98
+
99
+ e. Exceptions and Limitations means fair use, fair dealing, and/or
100
+ any other exception or limitation to Copyright and Similar Rights
101
+ that applies to Your use of the Licensed Material.
102
+
103
+ f. Licensed Material means the artistic or literary work, database,
104
+ or other material to which the Licensor applied this Public
105
+ License.
106
+
107
+ g. Licensed Rights means the rights granted to You subject to the
108
+ terms and conditions of this Public License, which are limited to
109
+ all Copyright and Similar Rights that apply to Your use of the
110
+ Licensed Material and that the Licensor has authority to license.
111
+
112
+ h. Licensor means the individual(s) or entity(ies) granting rights
113
+ under this Public License.
114
+
115
+ i. Share means to provide material to the public by any means or
116
+ process that requires permission under the Licensed Rights, such
117
+ as reproduction, public display, public performance, distribution,
118
+ dissemination, communication, or importation, and to make material
119
+ available to the public including in ways that members of the
120
+ public may access the material from a place and at a time
121
+ individually chosen by them.
122
+
123
+ j. Sui Generis Database Rights means rights other than copyright
124
+ resulting from Directive 96/9/EC of the European Parliament and of
125
+ the Council of 11 March 1996 on the legal protection of databases,
126
+ as amended and/or succeeded, as well as other essentially
127
+ equivalent rights anywhere in the world.
128
+
129
+ k. You means the individual or entity exercising the Licensed Rights
130
+ under this Public License. Your has a corresponding meaning.
131
+
132
+
133
+ Section 2 -- Scope.
134
+
135
+ a. License grant.
136
+
137
+ 1. Subject to the terms and conditions of this Public License,
138
+ the Licensor hereby grants You a worldwide, royalty-free,
139
+ non-sublicensable, non-exclusive, irrevocable license to
140
+ exercise the Licensed Rights in the Licensed Material to:
141
+
142
+ a. reproduce and Share the Licensed Material, in whole or
143
+ in part; and
144
+
145
+ b. produce, reproduce, and Share Adapted Material.
146
+
147
+ 2. Exceptions and Limitations. For the avoidance of doubt, where
148
+ Exceptions and Limitations apply to Your use, this Public
149
+ License does not apply, and You do not need to comply with
150
+ its terms and conditions.
151
+
152
+ 3. Term. The term of this Public License is specified in Section
153
+ 6(a).
154
+
155
+ 4. Media and formats; technical modifications allowed. The
156
+ Licensor authorizes You to exercise the Licensed Rights in
157
+ all media and formats whether now known or hereafter created,
158
+ and to make technical modifications necessary to do so. The
159
+ Licensor waives and/or agrees not to assert any right or
160
+ authority to forbid You from making technical modifications
161
+ necessary to exercise the Licensed Rights, including
162
+ technical modifications necessary to circumvent Effective
163
+ Technological Measures. For purposes of this Public License,
164
+ simply making modifications authorized by this Section 2(a)
165
+ (4) never produces Adapted Material.
166
+
167
+ 5. Downstream recipients.
168
+
169
+ a. Offer from the Licensor -- Licensed Material. Every
170
+ recipient of the Licensed Material automatically
171
+ receives an offer from the Licensor to exercise the
172
+ Licensed Rights under the terms and conditions of this
173
+ Public License.
174
+
175
+ b. No downstream restrictions. You may not offer or impose
176
+ any additional or different terms or conditions on, or
177
+ apply any Effective Technological Measures to, the
178
+ Licensed Material if doing so restricts exercise of the
179
+ Licensed Rights by any recipient of the Licensed
180
+ Material.
181
+
182
+ 6. No endorsement. Nothing in this Public License constitutes or
183
+ may be construed as permission to assert or imply that You
184
+ are, or that Your use of the Licensed Material is, connected
185
+ with, or sponsored, endorsed, or granted official status by,
186
+ the Licensor or others designated to receive attribution as
187
+ provided in Section 3(a)(1)(A)(i).
188
+
189
+ b. Other rights.
190
+
191
+ 1. Moral rights, such as the right of integrity, are not
192
+ licensed under this Public License, nor are publicity,
193
+ privacy, and/or other similar personality rights; however, to
194
+ the extent possible, the Licensor waives and/or agrees not to
195
+ assert any such rights held by the Licensor to the limited
196
+ extent necessary to allow You to exercise the Licensed
197
+ Rights, but not otherwise.
198
+
199
+ 2. Patent and trademark rights are not licensed under this
200
+ Public License.
201
+
202
+ 3. To the extent possible, the Licensor waives any right to
203
+ collect royalties from You for the exercise of the Licensed
204
+ Rights, whether directly or through a collecting society
205
+ under any voluntary or waivable statutory or compulsory
206
+ licensing scheme. In all other cases the Licensor expressly
207
+ reserves any right to collect such royalties.
208
+
209
+
210
+ Section 3 -- License Conditions.
211
+
212
+ Your exercise of the Licensed Rights is expressly made subject to the
213
+ following conditions.
214
+
215
+ a. Attribution.
216
+
217
+ 1. If You Share the Licensed Material (including in modified
218
+ form), You must:
219
+
220
+ a. retain the following if it is supplied by the Licensor
221
+ with the Licensed Material:
222
+
223
+ i. identification of the creator(s) of the Licensed
224
+ Material and any others designated to receive
225
+ attribution, in any reasonable manner requested by
226
+ the Licensor (including by pseudonym if
227
+ designated);
228
+
229
+ ii. a copyright notice;
230
+
231
+ iii. a notice that refers to this Public License;
232
+
233
+ iv. a notice that refers to the disclaimer of
234
+ warranties;
235
+
236
+ v. a URI or hyperlink to the Licensed Material to the
237
+ extent reasonably practicable;
238
+
239
+ b. indicate if You modified the Licensed Material and
240
+ retain an indication of any previous modifications; and
241
+
242
+ c. indicate the Licensed Material is licensed under this
243
+ Public License, and include the text of, or the URI or
244
+ hyperlink to, this Public License.
245
+
246
+ 2. You may satisfy the conditions in Section 3(a)(1) in any
247
+ reasonable manner based on the medium, means, and context in
248
+ which You Share the Licensed Material. For example, it may be
249
+ reasonable to satisfy the conditions by providing a URI or
250
+ hyperlink to a resource that includes the required
251
+ information.
252
+
253
+ 3. If requested by the Licensor, You must remove any of the
254
+ information required by Section 3(a)(1)(A) to the extent
255
+ reasonably practicable.
256
+
257
+ 4. If You Share Adapted Material You produce, the Adapter's
258
+ License You apply must not prevent recipients of the Adapted
259
+ Material from complying with this Public License.
260
+
261
+
262
+ Section 4 -- Sui Generis Database Rights.
263
+
264
+ Where the Licensed Rights include Sui Generis Database Rights that
265
+ apply to Your use of the Licensed Material:
266
+
267
+ a. for the avoidance of doubt, Section 2(a)(1) grants You the right
268
+ to extract, reuse, reproduce, and Share all or a substantial
269
+ portion of the contents of the database;
270
+
271
+ b. if You include all or a substantial portion of the database
272
+ contents in a database in which You have Sui Generis Database
273
+ Rights, then the database in which You have Sui Generis Database
274
+ Rights (but not its individual contents) is Adapted Material; and
275
+
276
+ c. You must comply with the conditions in Section 3(a) if You Share
277
+ all or a substantial portion of the contents of the database.
278
+
279
+ For the avoidance of doubt, this Section 4 supplements and does not
280
+ replace Your obligations under this Public License where the Licensed
281
+ Rights include other Copyright and Similar Rights.
282
+
283
+
284
+ Section 5 -- Disclaimer of Warranties and Limitation of Liability.
285
+
286
+ a. UNLESS OTHERWISE SEPARATELY UNDERTAKEN BY THE LICENSOR, TO THE
287
+ EXTENT POSSIBLE, THE LICENSOR OFFERS THE LICENSED MATERIAL AS-IS
288
+ AND AS-AVAILABLE, AND MAKES NO REPRESENTATIONS OR WARRANTIES OF
289
+ ANY KIND CONCERNING THE LICENSED MATERIAL, WHETHER EXPRESS,
290
+ IMPLIED, STATUTORY, OR OTHER. THIS INCLUDES, WITHOUT LIMITATION,
291
+ WARRANTIES OF TITLE, MERCHANTABILITY, FITNESS FOR A PARTICULAR
292
+ PURPOSE, NON-INFRINGEMENT, ABSENCE OF LATENT OR OTHER DEFECTS,
293
+ ACCURACY, OR THE PRESENCE OR ABSENCE OF ERRORS, WHETHER OR NOT
294
+ KNOWN OR DISCOVERABLE. WHERE DISCLAIMERS OF WARRANTIES ARE NOT
295
+ ALLOWED IN FULL OR IN PART, THIS DISCLAIMER MAY NOT APPLY TO YOU.
296
+
297
+ b. TO THE EXTENT POSSIBLE, IN NO EVENT WILL THE LICENSOR BE LIABLE
298
+ TO YOU ON ANY LEGAL THEORY (INCLUDING, WITHOUT LIMITATION,
299
+ NEGLIGENCE) OR OTHERWISE FOR ANY DIRECT, SPECIAL, INDIRECT,
300
+ INCIDENTAL, CONSEQUENTIAL, PUNITIVE, EXEMPLARY, OR OTHER LOSSES,
301
+ COSTS, EXPENSES, OR DAMAGES ARISING OUT OF THIS PUBLIC LICENSE OR
302
+ USE OF THE LICENSED MATERIAL, EVEN IF THE LICENSOR HAS BEEN
303
+ ADVISED OF THE POSSIBILITY OF SUCH LOSSES, COSTS, EXPENSES, OR
304
+ DAMAGES. WHERE A LIMITATION OF LIABILITY IS NOT ALLOWED IN FULL OR
305
+ IN PART, THIS LIMITATION MAY NOT APPLY TO YOU.
306
+
307
+ c. The disclaimer of warranties and limitation of liability provided
308
+ above shall be interpreted in a manner that, to the extent
309
+ possible, most closely approximates an absolute disclaimer and
310
+ waiver of all liability.
311
+
312
+
313
+ Section 6 -- Term and Termination.
314
+
315
+ a. This Public License applies for the term of the Copyright and
316
+ Similar Rights licensed here. However, if You fail to comply with
317
+ this Public License, then Your rights under this Public License
318
+ terminate automatically.
319
+
320
+ b. Where Your right to use the Licensed Material has terminated under
321
+ Section 6(a), it reinstates:
322
+
323
+ 1. automatically as of the date the violation is cured, provided
324
+ it is cured within 30 days of Your discovery of the
325
+ violation; or
326
+
327
+ 2. upon express reinstatement by the Licensor.
328
+
329
+ For the avoidance of doubt, this Section 6(b) does not affect any
330
+ right the Licensor may have to seek remedies for Your violations
331
+ of this Public License.
332
+
333
+ c. For the avoidance of doubt, the Licensor may also offer the
334
+ Licensed Material under separate terms or conditions or stop
335
+ distributing the Licensed Material at any time; however, doing so
336
+ will not terminate this Public License.
337
+
338
+ d. Sections 1, 5, 6, 7, and 8 survive termination of this Public
339
+ License.
340
+
341
+
342
+ Section 7 -- Other Terms and Conditions.
343
+
344
+ a. The Licensor shall not be bound by any additional or different
345
+ terms or conditions communicated by You unless expressly agreed.
346
+
347
+ b. Any arrangements, understandings, or agreements regarding the
348
+ Licensed Material not stated herein are separate from and
349
+ independent of the terms and conditions of this Public License.
350
+
351
+
352
+ Section 8 -- Interpretation.
353
+
354
+ a. For the avoidance of doubt, this Public License does not, and
355
+ shall not be interpreted to, reduce, limit, restrict, or impose
356
+ conditions on any use of the Licensed Material that could lawfully
357
+ be made without permission under this Public License.
358
+
359
+ b. To the extent possible, if any provision of this Public License is
360
+ deemed unenforceable, it shall be automatically reformed to the
361
+ minimum extent necessary to make it enforceable. If the provision
362
+ cannot be reformed, it shall be severed from this Public License
363
+ without affecting the enforceability of the remaining terms and
364
+ conditions.
365
+
366
+ c. No term or condition of this Public License will be waived and no
367
+ failure to comply consented to unless expressly agreed to by the
368
+ Licensor.
369
+
370
+ d. Nothing in this Public License constitutes or may be interpreted
371
+ as a limitation upon, or waiver of, any privileges and immunities
372
+ that apply to the Licensor or You, including from the legal
373
+ processes of any jurisdiction or authority.
374
+
375
+
376
+ =======================================================================
377
+
378
+ Creative Commons is not a party to its public
379
+ licenses. Notwithstanding, Creative Commons may elect to apply one of
380
+ its public licenses to material it publishes and in those instances
381
+ will be considered the “Licensor.” The text of the Creative Commons
382
+ public licenses is dedicated to the public domain under the CC0 Public
383
+ Domain Dedication. Except for the limited purpose of indicating that
384
+ material is shared under a Creative Commons public license or as
385
+ otherwise permitted by the Creative Commons policies published at
386
+ creativecommons.org/policies, Creative Commons does not authorize the
387
+ use of the trademark "Creative Commons" or any other trademark or logo
388
+ of Creative Commons without its prior written consent including,
389
+ without limitation, in connection with any unauthorized modifications
390
+ to any of its public licenses or any other arrangements,
391
+ understandings, or agreements concerning use of licensed material. For
392
+ the avoidance of doubt, this paragraph does not form part of the
393
+ public licenses.
394
+
395
+ Creative Commons may be contacted at creativecommons.org.
acoustic/bulgarian/mfa/v2.0.0a/README.md ADDED
@@ -0,0 +1,128 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+
2
+ # Bulgarian MFA acoustic model v2.0.0a
3
+
4
+ [Link to documentation on mfa-models](https://mfa-models.readthedocs.io/en/main/acoustic/bulgarian_mfa.html)
5
+
6
+ Jump to section:
7
+
8
+ - [Model details](#model-details)
9
+ - [Installation](#installation)
10
+ - [Intended use](#intended-use)
11
+ - [Performance Factors](#performance-factors)
12
+ - [Metrics](#metrics)
13
+ - [Ethical considerations](#ethical-considerations)
14
+ - [Troubleshooting issues](#troubleshooting-issues)
15
+ - [Training data](#training-data)
16
+ - [Evaluation data](#evaluation-data)
17
+
18
+ ## Model details
19
+
20
+ - **Maintainer:** [Montreal Forced Aligner](https://montreal-forced-aligner.readthedocs.io/)
21
+ - **Language:** [Bulgarian](https://en.wikipedia.org/wiki/Bulgarian_language)
22
+ - **Dialect:** N/A
23
+ - **Phone set:** [MFA](https://mfa-models.readthedocs.io/en/refactor/mfa_phone_set.html#bulgarian)
24
+ - **Model type:** `Acoustic`
25
+ - **Features:** `MFCC + pitch`
26
+ - **Architecture:** `gmm-hmm`
27
+ - **Model version:** `v2.0.0a`
28
+ - **Trained date:** `2022-05-29`
29
+ - **Compatible MFA version:** `v2.0.0`
30
+ - **License:** [CC BY 4.0](https://github.com/MontrealCorpusTools/mfa-models/tree/main/acoustic/bulgarian/mfa/v2.0.0a/LICENSE)
31
+ - **Citation:**
32
+
33
+ ```bibtex
34
+ @techreport{mfa_bulgarian_mfa_acoustic_2022,
35
+ author={McAuliffe, Michael and Sonderegger, Morgan},
36
+ title={Bulgarian MFA acoustic model v2.0.0a},
37
+ address={\url{https://mfa-models.readthedocs.io/acoustic/Bulgarian/Bulgarian MFA acoustic model v2_0_0a.html}},
38
+ year={2022},
39
+ month={May},
40
+ }
41
+ ```
42
+
43
+ - If you have comments or questions about this model, you can check [previous MFA model discussion posts](https://github.com/MontrealCorpusTools/mfa-models/discussions?discussions_q=Bulgarian+MFA+acoustic+model+v2.0.0a) or create [a new one](https://github.com/MontrealCorpusTools/mfa-models/discussions/new).
44
+
45
+ ## Installation
46
+
47
+ Install from the [MFA command line](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/models/index.html):
48
+
49
+ ```
50
+ mfa model download acoustic bulgarian_mfa
51
+ ```
52
+
53
+ Or download from [the release page](https://github.com/MontrealCorpusTools/mfa-models/releases/tag/acoustic-bulgarian_mfa-v2.0.0a).
54
+
55
+ ## Intended use
56
+
57
+ This model is intended for forced alignment of [Bulgarian](https://en.wikipedia.org/wiki/Bulgarian_language) transcripts.
58
+
59
+ This model uses the [MFA](https://mfa-models.readthedocs.io/en/refactor/mfa_phone_set.html#bulgarian) phone set for Bulgarian, and was trained with the pronunciation dictionaries above. Pronunciations can be added on top of the dictionary, as long as no additional phones are introduced.
60
+
61
+ ## Performance Factors
62
+
63
+ As forced alignment is a relatively well-constrained problem (given accurate transcripts), this model should be applicable to a range of recording conditions and speakers. However, please note that it was trained on read speech in low-noise environments, so as your data diverges from that, you may run into alignment issues or need to [increase the beam size of MFA](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/configuration/#configuring-specific-commands) or see other recommendations in the [troubleshooting section below](#troubleshooting-issues).
64
+
65
+ Please note as well that MFA does not use state-of-the-art ASR models for forced alignment. You may get better performance (especially on speech-to-text tasks) using other frameworks like [Coqui](https://coqui.ai/).
66
+
67
+ ## Metrics
68
+
69
+ Acoustic models are typically generated as one component of a larger ASR system where the metric is word error rate (WER). For forced alignment, there is typically not the same sort of gold standard measure for most languages.
70
+
71
+ As a rough approximation of the acoustic model quality, we evaluated it against the corpus it was trained on alongside a language model trained from the same data. Key caveat here is that this is not a typical WER measure on held out data, so it should not be taken as a hard measure of how well an acoustic model will generalize to your data, but rather is more of a sanity check that the training data quality was sufficiently high.
72
+
73
+ Using the pronunciation dictionaries and language models above:
74
+
75
+ - **WER:** `0%`
76
+ - **CER:** `0%`
77
+
78
+ ## Ethical considerations
79
+
80
+ Deploying any Speech-to-Text model into any production setting has ethical implications. You should consider these implications before use.
81
+
82
+ ### Demographic Bias
83
+
84
+ You should assume every machine learning model has demographic bias unless proven otherwise. For STT models, it is often the case that transcription accuracy is better for men than it is for women. If you are using this model in production, you should acknowledge this as a potential issue.
85
+
86
+ ### Surveillance
87
+
88
+ Speech-to-Text technologies may be misused to invade the privacy of others by recording and mining information from private conversations. This kind of individual privacy is protected by law in many countries. You should not assume consent to record and analyze private speech.
89
+
90
+
91
+ ## Troubleshooting issues
92
+
93
+ Machine learning models (like this acoustic model) perform best on data that is similar to the data on which they were trained.
94
+
95
+ The primary sources of variability in forced alignment will be the applicability of the pronunciation dictionary and how similar the speech, demographics, and recording conditions are. If you encounter issues in alignment, there are couple of avenues to improve performance:
96
+
97
+ 1. [Increase the beam size of MFA](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/configuration/#configuring-specific-commands)
98
+
99
+ * MFA defaults to a narrow beam to ensure quick alignment and also as a way to detect potential issues in your dataset, but depending on your data, you might benefit from boosting the beam to 100 or higher.
100
+
101
+ 2. Add pronunciations to the pronunciation dictionary
102
+
103
+ * This model was trained a particular dialect/style, and so adding pronunciations more representative of the variety spoken in your dataset will help alignment.
104
+
105
+ 3. Check the quality of your data
106
+
107
+ * MFA includes a [validator utility](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/data_validation.html), which aims to detect issues in the dataset.
108
+ * Use MFA's [anchor utility](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/workflows/anchor.html) to visually inspect your data as MFA sees it and correct issues in transcription or OOV items.
109
+
110
+ 4. Adapt the model to your data
111
+
112
+ * MFA has an [adaptation command](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/workflows/adapt_acoustic_model.html) to adapt some of the model to your data based on an initial alignment, and then run another alignment with the adapted model.
113
+
114
+ ## Training data
115
+
116
+ This model was trained on the following corpora:
117
+
118
+
119
+
120
+ * [Common Voice Bulgarian](../../../../corpus/bulgarian/common_voice_bulgarian/9.0/README.md):
121
+ * **Hours:** `12.06`
122
+ * **Speakers:** `62`
123
+ * **Utterances:** `7,921`
124
+
125
+ * [GlobalPhone Bulgarian](../../../../corpus/bulgarian/globalphone_bulgarian/3.1/README.md):
126
+ * **Hours:** `7.98`
127
+ * **Speakers:** `53`
128
+ * **Utterances:** `5,387`
acoustic/bulgarian/mfa/v2.0.0a/meta.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "name": "bulgarian_mfa",
3
+ "language": "Bulgarian",
4
+ "dialect": "",
5
+ "phone_set": "MFA",
6
+ "version": "2.0.0a",
7
+ "maintainer": "[Montreal Forced Aligner](https://montreal-forced-aligner.readthedocs.io/)",
8
+ "citation": "@techreport{mfa_bulgarian_mfa_acoustic_2022,\n\tauthor={McAuliffe, Michael and Sonderegger, Morgan},\n\ttitle={Bulgarian MFA acoustic model v2.0.0a},\n\taddress={\\url{https://mfa-models.readthedocs.io/acoustic/Bulgarian/Bulgarian MFA acoustic model v2_0_0a.html}},\n\tyear={2022},\n\tmonth={May},\n}",
9
+ "license": "CC BY 4.0",
10
+ "license_link": "[CC BY 4.0](https://github.com/MontrealCorpusTools/mfa-models/tree/main/acoustic/bulgarian/mfa/v2.0.0a/LICENSE)",
11
+ "architecture": "gmm-hmm",
12
+ "features": "MFCC + pitch",
13
+ "evaluation": {},
14
+ "decode": {},
15
+ "train_date": "2022-05-29"
16
+ }
acoustic/bulgarian/mfa/v3.0.0/LICENSE ADDED
@@ -0,0 +1,395 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ Attribution 4.0 International
2
+
3
+ =======================================================================
4
+
5
+ Creative Commons Corporation ("Creative Commons") is not a law firm and
6
+ does not provide legal services or legal advice. Distribution of
7
+ Creative Commons public licenses does not create a lawyer-client or
8
+ other relationship. Creative Commons makes its licenses and related
9
+ information available on an "as-is" basis. Creative Commons gives no
10
+ warranties regarding its licenses, any material licensed under their
11
+ terms and conditions, or any related information. Creative Commons
12
+ disclaims all liability for damages resulting from their use to the
13
+ fullest extent possible.
14
+
15
+ Using Creative Commons Public Licenses
16
+
17
+ Creative Commons public licenses provide a standard set of terms and
18
+ conditions that creators and other rights holders may use to share
19
+ original works of authorship and other material subject to copyright
20
+ and certain other rights specified in the public license below. The
21
+ following considerations are for informational purposes only, are not
22
+ exhaustive, and do not form part of our licenses.
23
+
24
+ Considerations for licensors: Our public licenses are
25
+ intended for use by those authorized to give the public
26
+ permission to use material in ways otherwise restricted by
27
+ copyright and certain other rights. Our licenses are
28
+ irrevocable. Licensors should read and understand the terms
29
+ and conditions of the license they choose before applying it.
30
+ Licensors should also secure all rights necessary before
31
+ applying our licenses so that the public can reuse the
32
+ material as expected. Licensors should clearly mark any
33
+ material not subject to the license. This includes other CC-
34
+ licensed material, or material used under an exception or
35
+ limitation to copyright. More considerations for licensors:
36
+ wiki.creativecommons.org/Considerations_for_licensors
37
+
38
+ Considerations for the public: By using one of our public
39
+ licenses, a licensor grants the public permission to use the
40
+ licensed material under specified terms and conditions. If
41
+ the licensor's permission is not necessary for any reason--for
42
+ example, because of any applicable exception or limitation to
43
+ copyright--then that use is not regulated by the license. Our
44
+ licenses grant only permissions under copyright and certain
45
+ other rights that a licensor has authority to grant. Use of
46
+ the licensed material may still be restricted for other
47
+ reasons, including because others have copyright or other
48
+ rights in the material. A licensor may make special requests,
49
+ such as asking that all changes be marked or described.
50
+ Although not required by our licenses, you are encouraged to
51
+ respect those requests where reasonable. More_considerations
52
+ for the public:
53
+ wiki.creativecommons.org/Considerations_for_licensees
54
+
55
+ =======================================================================
56
+
57
+ Creative Commons Attribution 4.0 International Public License
58
+
59
+ By exercising the Licensed Rights (defined below), You accept and agree
60
+ to be bound by the terms and conditions of this Creative Commons
61
+ Attribution 4.0 International Public License ("Public License"). To the
62
+ extent this Public License may be interpreted as a contract, You are
63
+ granted the Licensed Rights in consideration of Your acceptance of
64
+ these terms and conditions, and the Licensor grants You such rights in
65
+ consideration of benefits the Licensor receives from making the
66
+ Licensed Material available under these terms and conditions.
67
+
68
+
69
+ Section 1 -- Definitions.
70
+
71
+ a. Adapted Material means material subject to Copyright and Similar
72
+ Rights that is derived from or based upon the Licensed Material
73
+ and in which the Licensed Material is translated, altered,
74
+ arranged, transformed, or otherwise modified in a manner requiring
75
+ permission under the Copyright and Similar Rights held by the
76
+ Licensor. For purposes of this Public License, where the Licensed
77
+ Material is a musical work, performance, or sound recording,
78
+ Adapted Material is always produced where the Licensed Material is
79
+ synched in timed relation with a moving image.
80
+
81
+ b. Adapter's License means the license You apply to Your Copyright
82
+ and Similar Rights in Your contributions to Adapted Material in
83
+ accordance with the terms and conditions of this Public License.
84
+
85
+ c. Copyright and Similar Rights means copyright and/or similar rights
86
+ closely related to copyright including, without limitation,
87
+ performance, broadcast, sound recording, and Sui Generis Database
88
+ Rights, without regard to how the rights are labeled or
89
+ categorized. For purposes of this Public License, the rights
90
+ specified in Section 2(b)(1)-(2) are not Copyright and Similar
91
+ Rights.
92
+
93
+ d. Effective Technological Measures means those measures that, in the
94
+ absence of proper authority, may not be circumvented under laws
95
+ fulfilling obligations under Article 11 of the WIPO Copyright
96
+ Treaty adopted on December 20, 1996, and/or similar international
97
+ agreements.
98
+
99
+ e. Exceptions and Limitations means fair use, fair dealing, and/or
100
+ any other exception or limitation to Copyright and Similar Rights
101
+ that applies to Your use of the Licensed Material.
102
+
103
+ f. Licensed Material means the artistic or literary work, database,
104
+ or other material to which the Licensor applied this Public
105
+ License.
106
+
107
+ g. Licensed Rights means the rights granted to You subject to the
108
+ terms and conditions of this Public License, which are limited to
109
+ all Copyright and Similar Rights that apply to Your use of the
110
+ Licensed Material and that the Licensor has authority to license.
111
+
112
+ h. Licensor means the individual(s) or entity(ies) granting rights
113
+ under this Public License.
114
+
115
+ i. Share means to provide material to the public by any means or
116
+ process that requires permission under the Licensed Rights, such
117
+ as reproduction, public display, public performance, distribution,
118
+ dissemination, communication, or importation, and to make material
119
+ available to the public including in ways that members of the
120
+ public may access the material from a place and at a time
121
+ individually chosen by them.
122
+
123
+ j. Sui Generis Database Rights means rights other than copyright
124
+ resulting from Directive 96/9/EC of the European Parliament and of
125
+ the Council of 11 March 1996 on the legal protection of databases,
126
+ as amended and/or succeeded, as well as other essentially
127
+ equivalent rights anywhere in the world.
128
+
129
+ k. You means the individual or entity exercising the Licensed Rights
130
+ under this Public License. Your has a corresponding meaning.
131
+
132
+
133
+ Section 2 -- Scope.
134
+
135
+ a. License grant.
136
+
137
+ 1. Subject to the terms and conditions of this Public License,
138
+ the Licensor hereby grants You a worldwide, royalty-free,
139
+ non-sublicensable, non-exclusive, irrevocable license to
140
+ exercise the Licensed Rights in the Licensed Material to:
141
+
142
+ a. reproduce and Share the Licensed Material, in whole or
143
+ in part; and
144
+
145
+ b. produce, reproduce, and Share Adapted Material.
146
+
147
+ 2. Exceptions and Limitations. For the avoidance of doubt, where
148
+ Exceptions and Limitations apply to Your use, this Public
149
+ License does not apply, and You do not need to comply with
150
+ its terms and conditions.
151
+
152
+ 3. Term. The term of this Public License is specified in Section
153
+ 6(a).
154
+
155
+ 4. Media and formats; technical modifications allowed. The
156
+ Licensor authorizes You to exercise the Licensed Rights in
157
+ all media and formats whether now known or hereafter created,
158
+ and to make technical modifications necessary to do so. The
159
+ Licensor waives and/or agrees not to assert any right or
160
+ authority to forbid You from making technical modifications
161
+ necessary to exercise the Licensed Rights, including
162
+ technical modifications necessary to circumvent Effective
163
+ Technological Measures. For purposes of this Public License,
164
+ simply making modifications authorized by this Section 2(a)
165
+ (4) never produces Adapted Material.
166
+
167
+ 5. Downstream recipients.
168
+
169
+ a. Offer from the Licensor -- Licensed Material. Every
170
+ recipient of the Licensed Material automatically
171
+ receives an offer from the Licensor to exercise the
172
+ Licensed Rights under the terms and conditions of this
173
+ Public License.
174
+
175
+ b. No downstream restrictions. You may not offer or impose
176
+ any additional or different terms or conditions on, or
177
+ apply any Effective Technological Measures to, the
178
+ Licensed Material if doing so restricts exercise of the
179
+ Licensed Rights by any recipient of the Licensed
180
+ Material.
181
+
182
+ 6. No endorsement. Nothing in this Public License constitutes or
183
+ may be construed as permission to assert or imply that You
184
+ are, or that Your use of the Licensed Material is, connected
185
+ with, or sponsored, endorsed, or granted official status by,
186
+ the Licensor or others designated to receive attribution as
187
+ provided in Section 3(a)(1)(A)(i).
188
+
189
+ b. Other rights.
190
+
191
+ 1. Moral rights, such as the right of integrity, are not
192
+ licensed under this Public License, nor are publicity,
193
+ privacy, and/or other similar personality rights; however, to
194
+ the extent possible, the Licensor waives and/or agrees not to
195
+ assert any such rights held by the Licensor to the limited
196
+ extent necessary to allow You to exercise the Licensed
197
+ Rights, but not otherwise.
198
+
199
+ 2. Patent and trademark rights are not licensed under this
200
+ Public License.
201
+
202
+ 3. To the extent possible, the Licensor waives any right to
203
+ collect royalties from You for the exercise of the Licensed
204
+ Rights, whether directly or through a collecting society
205
+ under any voluntary or waivable statutory or compulsory
206
+ licensing scheme. In all other cases the Licensor expressly
207
+ reserves any right to collect such royalties.
208
+
209
+
210
+ Section 3 -- License Conditions.
211
+
212
+ Your exercise of the Licensed Rights is expressly made subject to the
213
+ following conditions.
214
+
215
+ a. Attribution.
216
+
217
+ 1. If You Share the Licensed Material (including in modified
218
+ form), You must:
219
+
220
+ a. retain the following if it is supplied by the Licensor
221
+ with the Licensed Material:
222
+
223
+ i. identification of the creator(s) of the Licensed
224
+ Material and any others designated to receive
225
+ attribution, in any reasonable manner requested by
226
+ the Licensor (including by pseudonym if
227
+ designated);
228
+
229
+ ii. a copyright notice;
230
+
231
+ iii. a notice that refers to this Public License;
232
+
233
+ iv. a notice that refers to the disclaimer of
234
+ warranties;
235
+
236
+ v. a URI or hyperlink to the Licensed Material to the
237
+ extent reasonably practicable;
238
+
239
+ b. indicate if You modified the Licensed Material and
240
+ retain an indication of any previous modifications; and
241
+
242
+ c. indicate the Licensed Material is licensed under this
243
+ Public License, and include the text of, or the URI or
244
+ hyperlink to, this Public License.
245
+
246
+ 2. You may satisfy the conditions in Section 3(a)(1) in any
247
+ reasonable manner based on the medium, means, and context in
248
+ which You Share the Licensed Material. For example, it may be
249
+ reasonable to satisfy the conditions by providing a URI or
250
+ hyperlink to a resource that includes the required
251
+ information.
252
+
253
+ 3. If requested by the Licensor, You must remove any of the
254
+ information required by Section 3(a)(1)(A) to the extent
255
+ reasonably practicable.
256
+
257
+ 4. If You Share Adapted Material You produce, the Adapter's
258
+ License You apply must not prevent recipients of the Adapted
259
+ Material from complying with this Public License.
260
+
261
+
262
+ Section 4 -- Sui Generis Database Rights.
263
+
264
+ Where the Licensed Rights include Sui Generis Database Rights that
265
+ apply to Your use of the Licensed Material:
266
+
267
+ a. for the avoidance of doubt, Section 2(a)(1) grants You the right
268
+ to extract, reuse, reproduce, and Share all or a substantial
269
+ portion of the contents of the database;
270
+
271
+ b. if You include all or a substantial portion of the database
272
+ contents in a database in which You have Sui Generis Database
273
+ Rights, then the database in which You have Sui Generis Database
274
+ Rights (but not its individual contents) is Adapted Material; and
275
+
276
+ c. You must comply with the conditions in Section 3(a) if You Share
277
+ all or a substantial portion of the contents of the database.
278
+
279
+ For the avoidance of doubt, this Section 4 supplements and does not
280
+ replace Your obligations under this Public License where the Licensed
281
+ Rights include other Copyright and Similar Rights.
282
+
283
+
284
+ Section 5 -- Disclaimer of Warranties and Limitation of Liability.
285
+
286
+ a. UNLESS OTHERWISE SEPARATELY UNDERTAKEN BY THE LICENSOR, TO THE
287
+ EXTENT POSSIBLE, THE LICENSOR OFFERS THE LICENSED MATERIAL AS-IS
288
+ AND AS-AVAILABLE, AND MAKES NO REPRESENTATIONS OR WARRANTIES OF
289
+ ANY KIND CONCERNING THE LICENSED MATERIAL, WHETHER EXPRESS,
290
+ IMPLIED, STATUTORY, OR OTHER. THIS INCLUDES, WITHOUT LIMITATION,
291
+ WARRANTIES OF TITLE, MERCHANTABILITY, FITNESS FOR A PARTICULAR
292
+ PURPOSE, NON-INFRINGEMENT, ABSENCE OF LATENT OR OTHER DEFECTS,
293
+ ACCURACY, OR THE PRESENCE OR ABSENCE OF ERRORS, WHETHER OR NOT
294
+ KNOWN OR DISCOVERABLE. WHERE DISCLAIMERS OF WARRANTIES ARE NOT
295
+ ALLOWED IN FULL OR IN PART, THIS DISCLAIMER MAY NOT APPLY TO YOU.
296
+
297
+ b. TO THE EXTENT POSSIBLE, IN NO EVENT WILL THE LICENSOR BE LIABLE
298
+ TO YOU ON ANY LEGAL THEORY (INCLUDING, WITHOUT LIMITATION,
299
+ NEGLIGENCE) OR OTHERWISE FOR ANY DIRECT, SPECIAL, INDIRECT,
300
+ INCIDENTAL, CONSEQUENTIAL, PUNITIVE, EXEMPLARY, OR OTHER LOSSES,
301
+ COSTS, EXPENSES, OR DAMAGES ARISING OUT OF THIS PUBLIC LICENSE OR
302
+ USE OF THE LICENSED MATERIAL, EVEN IF THE LICENSOR HAS BEEN
303
+ ADVISED OF THE POSSIBILITY OF SUCH LOSSES, COSTS, EXPENSES, OR
304
+ DAMAGES. WHERE A LIMITATION OF LIABILITY IS NOT ALLOWED IN FULL OR
305
+ IN PART, THIS LIMITATION MAY NOT APPLY TO YOU.
306
+
307
+ c. The disclaimer of warranties and limitation of liability provided
308
+ above shall be interpreted in a manner that, to the extent
309
+ possible, most closely approximates an absolute disclaimer and
310
+ waiver of all liability.
311
+
312
+
313
+ Section 6 -- Term and Termination.
314
+
315
+ a. This Public License applies for the term of the Copyright and
316
+ Similar Rights licensed here. However, if You fail to comply with
317
+ this Public License, then Your rights under this Public License
318
+ terminate automatically.
319
+
320
+ b. Where Your right to use the Licensed Material has terminated under
321
+ Section 6(a), it reinstates:
322
+
323
+ 1. automatically as of the date the violation is cured, provided
324
+ it is cured within 30 days of Your discovery of the
325
+ violation; or
326
+
327
+ 2. upon express reinstatement by the Licensor.
328
+
329
+ For the avoidance of doubt, this Section 6(b) does not affect any
330
+ right the Licensor may have to seek remedies for Your violations
331
+ of this Public License.
332
+
333
+ c. For the avoidance of doubt, the Licensor may also offer the
334
+ Licensed Material under separate terms or conditions or stop
335
+ distributing the Licensed Material at any time; however, doing so
336
+ will not terminate this Public License.
337
+
338
+ d. Sections 1, 5, 6, 7, and 8 survive termination of this Public
339
+ License.
340
+
341
+
342
+ Section 7 -- Other Terms and Conditions.
343
+
344
+ a. The Licensor shall not be bound by any additional or different
345
+ terms or conditions communicated by You unless expressly agreed.
346
+
347
+ b. Any arrangements, understandings, or agreements regarding the
348
+ Licensed Material not stated herein are separate from and
349
+ independent of the terms and conditions of this Public License.
350
+
351
+
352
+ Section 8 -- Interpretation.
353
+
354
+ a. For the avoidance of doubt, this Public License does not, and
355
+ shall not be interpreted to, reduce, limit, restrict, or impose
356
+ conditions on any use of the Licensed Material that could lawfully
357
+ be made without permission under this Public License.
358
+
359
+ b. To the extent possible, if any provision of this Public License is
360
+ deemed unenforceable, it shall be automatically reformed to the
361
+ minimum extent necessary to make it enforceable. If the provision
362
+ cannot be reformed, it shall be severed from this Public License
363
+ without affecting the enforceability of the remaining terms and
364
+ conditions.
365
+
366
+ c. No term or condition of this Public License will be waived and no
367
+ failure to comply consented to unless expressly agreed to by the
368
+ Licensor.
369
+
370
+ d. Nothing in this Public License constitutes or may be interpreted
371
+ as a limitation upon, or waiver of, any privileges and immunities
372
+ that apply to the Licensor or You, including from the legal
373
+ processes of any jurisdiction or authority.
374
+
375
+
376
+ =======================================================================
377
+
378
+ Creative Commons is not a party to its public
379
+ licenses. Notwithstanding, Creative Commons may elect to apply one of
380
+ its public licenses to material it publishes and in those instances
381
+ will be considered the “Licensor.” The text of the Creative Commons
382
+ public licenses is dedicated to the public domain under the CC0 Public
383
+ Domain Dedication. Except for the limited purpose of indicating that
384
+ material is shared under a Creative Commons public license or as
385
+ otherwise permitted by the Creative Commons policies published at
386
+ creativecommons.org/policies, Creative Commons does not authorize the
387
+ use of the trademark "Creative Commons" or any other trademark or logo
388
+ of Creative Commons without its prior written consent including,
389
+ without limitation, in connection with any unauthorized modifications
390
+ to any of its public licenses or any other arrangements,
391
+ understandings, or agreements concerning use of licensed material. For
392
+ the avoidance of doubt, this paragraph does not form part of the
393
+ public licenses.
394
+
395
+ Creative Commons may be contacted at creativecommons.org.
acoustic/bulgarian/mfa/v3.0.0/README.md ADDED
@@ -0,0 +1,127 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Bulgarian MFA acoustic model v3.0.0
2
+
3
+ [Link to documentation on mfa-models](https://mfa-models.readthedocs.io/en/main/acoustic/bulgarian_mfa.html)
4
+
5
+ Jump to section:
6
+
7
+ - [Model details](#model-details)
8
+ - [Installation](#installation)
9
+ - [Intended use](#intended-use)
10
+ - [Performance Factors](#performance-factors)
11
+ - [Metrics](#metrics)
12
+ - [Ethical considerations](#ethical-considerations)
13
+ - [Troubleshooting issues](#troubleshooting-issues)
14
+ - [Training data](#training-data)
15
+ - [Evaluation data](#evaluation-data)
16
+
17
+ ## Model details
18
+
19
+ - **Maintainer:** [Montreal Forced Aligner](https://montreal-forced-aligner.readthedocs.io/)
20
+ - **Language:** [Bulgarian](https://en.wikipedia.org/wiki/Bulgarian_language)
21
+ - **Dialect:** N/A
22
+ - **Phone set:** [MFA](https://mfa-models.readthedocs.io/en/refactor/mfa_phone_set.html#bulgarian)
23
+ - **Model type:** `Acoustic`
24
+ - **Features:** `MFCC`
25
+ - **Architecture:** `gmm-hmm`
26
+ - **Model version:** `v3.0.0`
27
+ - **Trained date:** `2024-02-26`
28
+ - **Compatible MFA version:** `v3.0.0`
29
+ - **License:** [CC BY 4.0](https://github.com/MontrealCorpusTools/mfa-models/tree/main/acoustic/bulgarian/mfa/v3.0.0/LICENSE)
30
+ - **Citation:**
31
+
32
+ ```bibtex
33
+ @techreport{mfa_bulgarian_mfa_acoustic_2024,
34
+ author={McAuliffe, Michael and Sonderegger, Morgan},
35
+ title={Bulgarian MFA acoustic model v3.0.0},
36
+ address={\url{https://mfa-models.readthedocs.io/acoustic/Bulgarian/Bulgarian MFA acoustic model v3_0_0.html}},
37
+ year={2024},
38
+ month={Feb},
39
+ }
40
+ ```
41
+
42
+ - If you have comments or questions about this model, you can check [previous MFA model discussion posts](https://github.com/MontrealCorpusTools/mfa-models/discussions?discussions_q=Bulgarian+MFA+acoustic+model+v3.0.0) or create [a new one](https://github.com/MontrealCorpusTools/mfa-models/discussions/new).
43
+
44
+ ## Installation
45
+
46
+ Install from the [MFA command line](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/models/index.html):
47
+
48
+ ```
49
+ mfa model download acoustic bulgarian_mfa
50
+ ```
51
+
52
+ Or download from [the release page](https://github.com/MontrealCorpusTools/mfa-models/releases/tag/acoustic-bulgarian_mfa-v3.0.0).
53
+
54
+ ## Intended use
55
+
56
+ This model is intended for forced alignment of [Bulgarian](https://en.wikipedia.org/wiki/Bulgarian_language) transcripts.
57
+
58
+ This model uses the [MFA](https://mfa-models.readthedocs.io/en/refactor/mfa_phone_set.html#bulgarian) phone set for Bulgarian, and was trained with the pronunciation dictionaries above. Pronunciations can be added on top of the dictionary, as long as no additional phones are introduced.
59
+
60
+ ## Performance Factors
61
+
62
+ As forced alignment is a relatively well-constrained problem (given accurate transcripts), this model should be applicable to a range of recording conditions and speakers. However, please note that it was trained on read speech in low-noise environments, so as your data diverges from that, you may run into alignment issues or need to [increase the beam size of MFA](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/configuration/#configuring-specific-commands) or see other recommendations in the [troubleshooting section below](#troubleshooting-issues).
63
+
64
+ Please note as well that MFA does not use state-of-the-art ASR models for forced alignment. You may get better performance (especially on speech-to-text tasks) using other frameworks like [Coqui](https://coqui.ai/).
65
+
66
+ ## Metrics
67
+
68
+ Acoustic models are typically generated as one component of a larger ASR system where the metric is word error rate (WER). For forced alignment, there is typically not the same sort of gold standard measure for most languages.
69
+
70
+ As a rough approximation of the acoustic model quality, we evaluated it against the corpus it was trained on alongside a language model trained from the same data. Key caveat here is that this is not a typical WER measure on held out data, so it should not be taken as a hard measure of how well an acoustic model will generalize to your data, but rather is more of a sanity check that the training data quality was sufficiently high.
71
+
72
+ Using the pronunciation dictionaries and language models above:
73
+
74
+ - **WER:** `0%`
75
+ - **CER:** `0%`
76
+
77
+ ## Ethical considerations
78
+
79
+ Deploying any Speech-to-Text model into any production setting has ethical implications. You should consider these implications before use.
80
+
81
+ ### Demographic Bias
82
+
83
+ You should assume every machine learning model has demographic bias unless proven otherwise. For STT models, it is often the case that transcription accuracy is better for men than it is for women. If you are using this model in production, you should acknowledge this as a potential issue.
84
+
85
+ ### Surveillance
86
+
87
+ Speech-to-Text technologies may be misused to invade the privacy of others by recording and mining information from private conversations. This kind of individual privacy is protected by law in many countries. You should not assume consent to record and analyze private speech.
88
+
89
+
90
+ ## Troubleshooting issues
91
+
92
+ Machine learning models (like this acoustic model) perform best on data that is similar to the data on which they were trained.
93
+
94
+ The primary sources of variability in forced alignment will be the applicability of the pronunciation dictionary and how similar the speech, demographics, and recording conditions are. If you encounter issues in alignment, there are couple of avenues to improve performance:
95
+
96
+ 1. [Increase the beam size of MFA](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/configuration/#configuring-specific-commands)
97
+
98
+ * MFA defaults to a narrow beam to ensure quick alignment and also as a way to detect potential issues in your dataset, but depending on your data, you might benefit from boosting the beam to 100 or higher.
99
+
100
+ 2. Add pronunciations to the pronunciation dictionary
101
+
102
+ * This model was trained a particular dialect/style, and so adding pronunciations more representative of the variety spoken in your dataset will help alignment.
103
+
104
+ 3. Check the quality of your data
105
+
106
+ * MFA includes a [validator utility](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/data_validation.html), which aims to detect issues in the dataset.
107
+ * Use MFA's [anchor utility](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/workflows/anchor.html) to visually inspect your data as MFA sees it and correct issues in transcription or OOV items.
108
+
109
+ 4. Adapt the model to your data
110
+
111
+ * MFA has an [adaptation command](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/workflows/adapt_acoustic_model.html) to adapt some of the model to your data based on an initial alignment, and then run another alignment with the adapted model.
112
+
113
+ ## Training data
114
+
115
+ This model was trained on the following corpora:
116
+
117
+
118
+
119
+ * [Common Voice Bulgarian](../../../../corpus/bulgarian/common_voice_bulgarian/16.1/README.md):
120
+ * **Hours:** `13.72`
121
+ * **Speakers:** `95`
122
+ * **Utterances:** `9,126`
123
+
124
+ * [GlobalPhone Bulgarian](../../../../corpus/bulgarian/globalphone_bulgarian/3.1/README.md):
125
+ * **Hours:** `7.98`
126
+ * **Speakers:** `53`
127
+ * **Utterances:** `5,387`
acoustic/bulgarian/mfa/v3.0.0/meta.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "name": "bulgarian_mfa",
3
+ "language": "Bulgarian",
4
+ "dialect": "",
5
+ "phone_set": "MFA",
6
+ "version": "3.0.0",
7
+ "maintainer": "[Montreal Forced Aligner](https://montreal-forced-aligner.readthedocs.io/)",
8
+ "citation": "@techreport{mfa_bulgarian_mfa_acoustic_2024,\n\tauthor={McAuliffe, Michael and Sonderegger, Morgan},\n\ttitle={Bulgarian MFA acoustic model v3.0.0},\n\taddress={\\url{https://mfa-models.readthedocs.io/acoustic/Bulgarian/Bulgarian MFA acoustic model v3_0_0.html}},\n\tyear={2024},\n\tmonth={Feb},\n}",
9
+ "license": "CC BY 4.0",
10
+ "license_link": "[CC BY 4.0](https://github.com/MontrealCorpusTools/mfa-models/tree/main/acoustic/bulgarian/mfa/v3.0.0/LICENSE)",
11
+ "architecture": "gmm-hmm",
12
+ "features": "MFCC",
13
+ "evaluation": {},
14
+ "decode": {},
15
+ "train_date": "2024-02-26"
16
+ }
acoustic/chuvash/cv/v2.0.0/README.md ADDED
@@ -0,0 +1,110 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+
2
+ # Chuvash CV acoustic model v2.0.0
3
+
4
+ [Link to documentation on mfa-models](https://mfa-models.readthedocs.io/en/main/acoustic/chuvash_cv.html)
5
+
6
+ Jump to section:
7
+
8
+ - [Model details](#model-details)
9
+ - [Installation](#installation)
10
+ - [Intended use](#intended-use)
11
+ - [Performance Factors](#performance-factors)
12
+ - [Training data](#training-data)
13
+ - [Ethical considerations](#ethical-considerations)
14
+ - [Troubleshooting issues](#troubleshooting-issues)
15
+
16
+ ## Model details
17
+
18
+ - **Maintainer:** [Vox Communis](https://osf.io/t957v/)
19
+ - **Language:** [Chuvash](https://en.wikipedia.org/wiki/Chuvash_language)
20
+ - **Dialect:** N/A
21
+ - **Phone set:** [XPF](https://github.com/CohenPr-XPF/XPF)
22
+ - **Model type:** `Acoustic model`
23
+ - **Features:** `MFCC`
24
+ - **Architecture:** `gmm-hmm`
25
+ - **Model version:** `v2.0.0`
26
+ - **Trained date:** `02-11-2022`
27
+ - **Compatible MFA version:** `v2.0.0`
28
+ - **License:** [CC-0](https://creativecommons.org/publicdomain/zero/1.0/)
29
+ - **Citation:**
30
+
31
+ ```bibtex
32
+ @misc{Ahn_Chodroff_2022,
33
+ author={Ahn, Emily and Chodroff, Eleanor},
34
+ title={VoxCommunis Corpus},
35
+ address={\url{https://osf.io/t957v}},
36
+ publisher={OSF},
37
+ year={2022},
38
+ month={Jan}
39
+ }
40
+ ```
41
+
42
+ - If you have comments or questions about this model, you can check [previous MFA model discussion posts](https://github.com/MontrealCorpusTools/mfa-models/discussions?discussions_q=Chuvash+CV+acoustic+model+v2.0.0) or create [a new one](https://github.com/MontrealCorpusTools/mfa-models/discussions/new).
43
+
44
+ ## Installation
45
+
46
+ Install from the [MFA command line](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/models/index.html):
47
+
48
+ ```
49
+ mfa model download acoustic chuvash_cv
50
+ ```
51
+
52
+ Or download from [the release page](https://github.com/MontrealCorpusTools/mfa-models/releases/tag/acoustic-chuvash_cv-v2.0.0).
53
+
54
+ ## Intended use
55
+
56
+ This model is intended for forced alignment of [Chuvash](https://en.wikipedia.org/wiki/Chuvash_language) transcripts.
57
+
58
+ This model uses the [XPF](https://github.com/CohenPr-XPF/XPF) phone set for Chuvash, and was trained with the pronunciation dictionaries above. Pronunciations can be added on top of the dictionary, as long as no additional phones are introduced.
59
+
60
+ ## Performance Factors
61
+
62
+ As forced alignment is a relatively well-constrained problem (given accurate transcripts), this model should be applicable to a range of recording conditions and speakers. However, please note that it was trained on read speech in low-noise environments, so as your data diverges from that, you may run into alignment issues or need to [increase the beam size of MFA](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/configuration/#configuring-specific-commands) or see other recommendations in the [troubleshooting section below](#troubleshooting-issues).
63
+
64
+ Please note as well that MFA does not use state-of-the-art ASR models for forced alignment. You may get better performance (especially on speech-to-text tasks) using other frameworks like [Coqui](https://coqui.ai/).
65
+
66
+ ## Ethical considerations
67
+
68
+ Deploying any Speech-to-Text model into any production setting has ethical implications. You should consider these implications before use.
69
+
70
+ ### Demographic Bias
71
+
72
+ You should assume every machine learning model has demographic bias unless proven otherwise. For STT models, it is often the case that transcription accuracy is better for men than it is for women. If you are using this model in production, you should acknowledge this as a potential issue.
73
+
74
+ ### Surveillance
75
+
76
+ Speech-to-Text technologies may be misused to invade the privacy of others by recording and mining information from private conversations. This kind of individual privacy is protected by law in many countries. You should not assume consent to record and analyze private speech.
77
+
78
+ ## Troubleshooting issues
79
+
80
+ Machine learning models (like this acoustic model) perform best on data that is similar to the data on which they were trained.
81
+
82
+ The primary sources of variability in forced alignment will be the applicability of the pronunciation dictionary and how similar the speech, demographics, and recording conditions are. If you encounter issues in alignment, there are couple of avenues to improve performance:
83
+
84
+ 1. [Increase the beam size of MFA](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/configuration/#configuring-specific-commands)
85
+
86
+ * MFA defaults to a narrow beam to ensure quick alignment and also as a way to detect potential issues in your dataset, but depending on your data, you might benefit from boosting the beam to 100 or higher.
87
+
88
+ 2. Add pronunciations to the pronunciation dictionary
89
+
90
+ * This model was trained a particular dialect/style, and so adding pronunciations more representative of the variety spoken in your dataset will help alignment.
91
+
92
+ 3. Check the quality of your data
93
+
94
+ * MFA includes a [validator utility](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/data_validation.html), which aims to detect issues in the dataset.
95
+ * Use MFA's [anchor utility](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/workflows/anchor.html) to visually inspect your data as MFA sees it and correct issues in transcription or OOV items.
96
+
97
+ 4. Adapt the model to your data
98
+
99
+ * MFA has an [adaptation command](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/workflows/adapt_acoustic_model.html) to adapt some of the model to your data based on an initial alignment, and then run another alignment with the adapted model.
100
+
101
+ ## Training data
102
+
103
+ This model was trained on the following corpora:
104
+
105
+
106
+
107
+ * [Common Voice Chuvash](../../../../corpus/chuvash/common_voice_chuvash/7.0/README.md):
108
+ * **Hours:** `5.00`
109
+ * **Speakers:** `82`
110
+ * **Utterances:** `3,748`
acoustic/chuvash/cv/v2.0.0/meta.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "name": "chuvash_cv",
3
+ "language": "Chuvash",
4
+ "dialect": "",
5
+ "phone_set": "CV",
6
+ "version": "2.0.0",
7
+ "maintainer": "[Vox Communis](https://osf.io/t957v/)",
8
+ "citation": "@misc{Ahn_Chodroff_2022,\n\tauthor={Ahn, Emily and Chodroff, Eleanor},\n\ttitle={VoxCommunis Corpus},\n\taddress={\\url{https://osf.io/t957v}},\n\tpublisher={OSF},\n\tyear={2022}, \n\tmonth={Jan}\n}",
9
+ "license": "CC-0",
10
+ "license_link": "[CC-0](https://creativecommons.org/publicdomain/zero/1.0/)",
11
+ "architecture": "gmm-hmm",
12
+ "features": "MFCC",
13
+ "evaluation": {},
14
+ "decode": {},
15
+ "train_date": "02-11-2022"
16
+ }
acoustic/croatian.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:103cf14062d937cab272b5db959918e0f911b2a3f66286edbdcda4bb11aa15d9
3
+ size 14668100
acoustic/croatian/mfa/v2.0.0/LICENSE ADDED
@@ -0,0 +1,395 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ Attribution 4.0 International
2
+
3
+ =======================================================================
4
+
5
+ Creative Commons Corporation ("Creative Commons") is not a law firm and
6
+ does not provide legal services or legal advice. Distribution of
7
+ Creative Commons public licenses does not create a lawyer-client or
8
+ other relationship. Creative Commons makes its licenses and related
9
+ information available on an "as-is" basis. Creative Commons gives no
10
+ warranties regarding its licenses, any material licensed under their
11
+ terms and conditions, or any related information. Creative Commons
12
+ disclaims all liability for damages resulting from their use to the
13
+ fullest extent possible.
14
+
15
+ Using Creative Commons Public Licenses
16
+
17
+ Creative Commons public licenses provide a standard set of terms and
18
+ conditions that creators and other rights holders may use to share
19
+ original works of authorship and other material subject to copyright
20
+ and certain other rights specified in the public license below. The
21
+ following considerations are for informational purposes only, are not
22
+ exhaustive, and do not form part of our licenses.
23
+
24
+ Considerations for licensors: Our public licenses are
25
+ intended for use by those authorized to give the public
26
+ permission to use material in ways otherwise restricted by
27
+ copyright and certain other rights. Our licenses are
28
+ irrevocable. Licensors should read and understand the terms
29
+ and conditions of the license they choose before applying it.
30
+ Licensors should also secure all rights necessary before
31
+ applying our licenses so that the public can reuse the
32
+ material as expected. Licensors should clearly mark any
33
+ material not subject to the license. This includes other CC-
34
+ licensed material, or material used under an exception or
35
+ limitation to copyright. More considerations for licensors:
36
+ wiki.creativecommons.org/Considerations_for_licensors
37
+
38
+ Considerations for the public: By using one of our public
39
+ licenses, a licensor grants the public permission to use the
40
+ licensed material under specified terms and conditions. If
41
+ the licensor's permission is not necessary for any reason--for
42
+ example, because of any applicable exception or limitation to
43
+ copyright--then that use is not regulated by the license. Our
44
+ licenses grant only permissions under copyright and certain
45
+ other rights that a licensor has authority to grant. Use of
46
+ the licensed material may still be restricted for other
47
+ reasons, including because others have copyright or other
48
+ rights in the material. A licensor may make special requests,
49
+ such as asking that all changes be marked or described.
50
+ Although not required by our licenses, you are encouraged to
51
+ respect those requests where reasonable. More_considerations
52
+ for the public:
53
+ wiki.creativecommons.org/Considerations_for_licensees
54
+
55
+ =======================================================================
56
+
57
+ Creative Commons Attribution 4.0 International Public License
58
+
59
+ By exercising the Licensed Rights (defined below), You accept and agree
60
+ to be bound by the terms and conditions of this Creative Commons
61
+ Attribution 4.0 International Public License ("Public License"). To the
62
+ extent this Public License may be interpreted as a contract, You are
63
+ granted the Licensed Rights in consideration of Your acceptance of
64
+ these terms and conditions, and the Licensor grants You such rights in
65
+ consideration of benefits the Licensor receives from making the
66
+ Licensed Material available under these terms and conditions.
67
+
68
+
69
+ Section 1 -- Definitions.
70
+
71
+ a. Adapted Material means material subject to Copyright and Similar
72
+ Rights that is derived from or based upon the Licensed Material
73
+ and in which the Licensed Material is translated, altered,
74
+ arranged, transformed, or otherwise modified in a manner requiring
75
+ permission under the Copyright and Similar Rights held by the
76
+ Licensor. For purposes of this Public License, where the Licensed
77
+ Material is a musical work, performance, or sound recording,
78
+ Adapted Material is always produced where the Licensed Material is
79
+ synched in timed relation with a moving image.
80
+
81
+ b. Adapter's License means the license You apply to Your Copyright
82
+ and Similar Rights in Your contributions to Adapted Material in
83
+ accordance with the terms and conditions of this Public License.
84
+
85
+ c. Copyright and Similar Rights means copyright and/or similar rights
86
+ closely related to copyright including, without limitation,
87
+ performance, broadcast, sound recording, and Sui Generis Database
88
+ Rights, without regard to how the rights are labeled or
89
+ categorized. For purposes of this Public License, the rights
90
+ specified in Section 2(b)(1)-(2) are not Copyright and Similar
91
+ Rights.
92
+
93
+ d. Effective Technological Measures means those measures that, in the
94
+ absence of proper authority, may not be circumvented under laws
95
+ fulfilling obligations under Article 11 of the WIPO Copyright
96
+ Treaty adopted on December 20, 1996, and/or similar international
97
+ agreements.
98
+
99
+ e. Exceptions and Limitations means fair use, fair dealing, and/or
100
+ any other exception or limitation to Copyright and Similar Rights
101
+ that applies to Your use of the Licensed Material.
102
+
103
+ f. Licensed Material means the artistic or literary work, database,
104
+ or other material to which the Licensor applied this Public
105
+ License.
106
+
107
+ g. Licensed Rights means the rights granted to You subject to the
108
+ terms and conditions of this Public License, which are limited to
109
+ all Copyright and Similar Rights that apply to Your use of the
110
+ Licensed Material and that the Licensor has authority to license.
111
+
112
+ h. Licensor means the individual(s) or entity(ies) granting rights
113
+ under this Public License.
114
+
115
+ i. Share means to provide material to the public by any means or
116
+ process that requires permission under the Licensed Rights, such
117
+ as reproduction, public display, public performance, distribution,
118
+ dissemination, communication, or importation, and to make material
119
+ available to the public including in ways that members of the
120
+ public may access the material from a place and at a time
121
+ individually chosen by them.
122
+
123
+ j. Sui Generis Database Rights means rights other than copyright
124
+ resulting from Directive 96/9/EC of the European Parliament and of
125
+ the Council of 11 March 1996 on the legal protection of databases,
126
+ as amended and/or succeeded, as well as other essentially
127
+ equivalent rights anywhere in the world.
128
+
129
+ k. You means the individual or entity exercising the Licensed Rights
130
+ under this Public License. Your has a corresponding meaning.
131
+
132
+
133
+ Section 2 -- Scope.
134
+
135
+ a. License grant.
136
+
137
+ 1. Subject to the terms and conditions of this Public License,
138
+ the Licensor hereby grants You a worldwide, royalty-free,
139
+ non-sublicensable, non-exclusive, irrevocable license to
140
+ exercise the Licensed Rights in the Licensed Material to:
141
+
142
+ a. reproduce and Share the Licensed Material, in whole or
143
+ in part; and
144
+
145
+ b. produce, reproduce, and Share Adapted Material.
146
+
147
+ 2. Exceptions and Limitations. For the avoidance of doubt, where
148
+ Exceptions and Limitations apply to Your use, this Public
149
+ License does not apply, and You do not need to comply with
150
+ its terms and conditions.
151
+
152
+ 3. Term. The term of this Public License is specified in Section
153
+ 6(a).
154
+
155
+ 4. Media and formats; technical modifications allowed. The
156
+ Licensor authorizes You to exercise the Licensed Rights in
157
+ all media and formats whether now known or hereafter created,
158
+ and to make technical modifications necessary to do so. The
159
+ Licensor waives and/or agrees not to assert any right or
160
+ authority to forbid You from making technical modifications
161
+ necessary to exercise the Licensed Rights, including
162
+ technical modifications necessary to circumvent Effective
163
+ Technological Measures. For purposes of this Public License,
164
+ simply making modifications authorized by this Section 2(a)
165
+ (4) never produces Adapted Material.
166
+
167
+ 5. Downstream recipients.
168
+
169
+ a. Offer from the Licensor -- Licensed Material. Every
170
+ recipient of the Licensed Material automatically
171
+ receives an offer from the Licensor to exercise the
172
+ Licensed Rights under the terms and conditions of this
173
+ Public License.
174
+
175
+ b. No downstream restrictions. You may not offer or impose
176
+ any additional or different terms or conditions on, or
177
+ apply any Effective Technological Measures to, the
178
+ Licensed Material if doing so restricts exercise of the
179
+ Licensed Rights by any recipient of the Licensed
180
+ Material.
181
+
182
+ 6. No endorsement. Nothing in this Public License constitutes or
183
+ may be construed as permission to assert or imply that You
184
+ are, or that Your use of the Licensed Material is, connected
185
+ with, or sponsored, endorsed, or granted official status by,
186
+ the Licensor or others designated to receive attribution as
187
+ provided in Section 3(a)(1)(A)(i).
188
+
189
+ b. Other rights.
190
+
191
+ 1. Moral rights, such as the right of integrity, are not
192
+ licensed under this Public License, nor are publicity,
193
+ privacy, and/or other similar personality rights; however, to
194
+ the extent possible, the Licensor waives and/or agrees not to
195
+ assert any such rights held by the Licensor to the limited
196
+ extent necessary to allow You to exercise the Licensed
197
+ Rights, but not otherwise.
198
+
199
+ 2. Patent and trademark rights are not licensed under this
200
+ Public License.
201
+
202
+ 3. To the extent possible, the Licensor waives any right to
203
+ collect royalties from You for the exercise of the Licensed
204
+ Rights, whether directly or through a collecting society
205
+ under any voluntary or waivable statutory or compulsory
206
+ licensing scheme. In all other cases the Licensor expressly
207
+ reserves any right to collect such royalties.
208
+
209
+
210
+ Section 3 -- License Conditions.
211
+
212
+ Your exercise of the Licensed Rights is expressly made subject to the
213
+ following conditions.
214
+
215
+ a. Attribution.
216
+
217
+ 1. If You Share the Licensed Material (including in modified
218
+ form), You must:
219
+
220
+ a. retain the following if it is supplied by the Licensor
221
+ with the Licensed Material:
222
+
223
+ i. identification of the creator(s) of the Licensed
224
+ Material and any others designated to receive
225
+ attribution, in any reasonable manner requested by
226
+ the Licensor (including by pseudonym if
227
+ designated);
228
+
229
+ ii. a copyright notice;
230
+
231
+ iii. a notice that refers to this Public License;
232
+
233
+ iv. a notice that refers to the disclaimer of
234
+ warranties;
235
+
236
+ v. a URI or hyperlink to the Licensed Material to the
237
+ extent reasonably practicable;
238
+
239
+ b. indicate if You modified the Licensed Material and
240
+ retain an indication of any previous modifications; and
241
+
242
+ c. indicate the Licensed Material is licensed under this
243
+ Public License, and include the text of, or the URI or
244
+ hyperlink to, this Public License.
245
+
246
+ 2. You may satisfy the conditions in Section 3(a)(1) in any
247
+ reasonable manner based on the medium, means, and context in
248
+ which You Share the Licensed Material. For example, it may be
249
+ reasonable to satisfy the conditions by providing a URI or
250
+ hyperlink to a resource that includes the required
251
+ information.
252
+
253
+ 3. If requested by the Licensor, You must remove any of the
254
+ information required by Section 3(a)(1)(A) to the extent
255
+ reasonably practicable.
256
+
257
+ 4. If You Share Adapted Material You produce, the Adapter's
258
+ License You apply must not prevent recipients of the Adapted
259
+ Material from complying with this Public License.
260
+
261
+
262
+ Section 4 -- Sui Generis Database Rights.
263
+
264
+ Where the Licensed Rights include Sui Generis Database Rights that
265
+ apply to Your use of the Licensed Material:
266
+
267
+ a. for the avoidance of doubt, Section 2(a)(1) grants You the right
268
+ to extract, reuse, reproduce, and Share all or a substantial
269
+ portion of the contents of the database;
270
+
271
+ b. if You include all or a substantial portion of the database
272
+ contents in a database in which You have Sui Generis Database
273
+ Rights, then the database in which You have Sui Generis Database
274
+ Rights (but not its individual contents) is Adapted Material; and
275
+
276
+ c. You must comply with the conditions in Section 3(a) if You Share
277
+ all or a substantial portion of the contents of the database.
278
+
279
+ For the avoidance of doubt, this Section 4 supplements and does not
280
+ replace Your obligations under this Public License where the Licensed
281
+ Rights include other Copyright and Similar Rights.
282
+
283
+
284
+ Section 5 -- Disclaimer of Warranties and Limitation of Liability.
285
+
286
+ a. UNLESS OTHERWISE SEPARATELY UNDERTAKEN BY THE LICENSOR, TO THE
287
+ EXTENT POSSIBLE, THE LICENSOR OFFERS THE LICENSED MATERIAL AS-IS
288
+ AND AS-AVAILABLE, AND MAKES NO REPRESENTATIONS OR WARRANTIES OF
289
+ ANY KIND CONCERNING THE LICENSED MATERIAL, WHETHER EXPRESS,
290
+ IMPLIED, STATUTORY, OR OTHER. THIS INCLUDES, WITHOUT LIMITATION,
291
+ WARRANTIES OF TITLE, MERCHANTABILITY, FITNESS FOR A PARTICULAR
292
+ PURPOSE, NON-INFRINGEMENT, ABSENCE OF LATENT OR OTHER DEFECTS,
293
+ ACCURACY, OR THE PRESENCE OR ABSENCE OF ERRORS, WHETHER OR NOT
294
+ KNOWN OR DISCOVERABLE. WHERE DISCLAIMERS OF WARRANTIES ARE NOT
295
+ ALLOWED IN FULL OR IN PART, THIS DISCLAIMER MAY NOT APPLY TO YOU.
296
+
297
+ b. TO THE EXTENT POSSIBLE, IN NO EVENT WILL THE LICENSOR BE LIABLE
298
+ TO YOU ON ANY LEGAL THEORY (INCLUDING, WITHOUT LIMITATION,
299
+ NEGLIGENCE) OR OTHERWISE FOR ANY DIRECT, SPECIAL, INDIRECT,
300
+ INCIDENTAL, CONSEQUENTIAL, PUNITIVE, EXEMPLARY, OR OTHER LOSSES,
301
+ COSTS, EXPENSES, OR DAMAGES ARISING OUT OF THIS PUBLIC LICENSE OR
302
+ USE OF THE LICENSED MATERIAL, EVEN IF THE LICENSOR HAS BEEN
303
+ ADVISED OF THE POSSIBILITY OF SUCH LOSSES, COSTS, EXPENSES, OR
304
+ DAMAGES. WHERE A LIMITATION OF LIABILITY IS NOT ALLOWED IN FULL OR
305
+ IN PART, THIS LIMITATION MAY NOT APPLY TO YOU.
306
+
307
+ c. The disclaimer of warranties and limitation of liability provided
308
+ above shall be interpreted in a manner that, to the extent
309
+ possible, most closely approximates an absolute disclaimer and
310
+ waiver of all liability.
311
+
312
+
313
+ Section 6 -- Term and Termination.
314
+
315
+ a. This Public License applies for the term of the Copyright and
316
+ Similar Rights licensed here. However, if You fail to comply with
317
+ this Public License, then Your rights under this Public License
318
+ terminate automatically.
319
+
320
+ b. Where Your right to use the Licensed Material has terminated under
321
+ Section 6(a), it reinstates:
322
+
323
+ 1. automatically as of the date the violation is cured, provided
324
+ it is cured within 30 days of Your discovery of the
325
+ violation; or
326
+
327
+ 2. upon express reinstatement by the Licensor.
328
+
329
+ For the avoidance of doubt, this Section 6(b) does not affect any
330
+ right the Licensor may have to seek remedies for Your violations
331
+ of this Public License.
332
+
333
+ c. For the avoidance of doubt, the Licensor may also offer the
334
+ Licensed Material under separate terms or conditions or stop
335
+ distributing the Licensed Material at any time; however, doing so
336
+ will not terminate this Public License.
337
+
338
+ d. Sections 1, 5, 6, 7, and 8 survive termination of this Public
339
+ License.
340
+
341
+
342
+ Section 7 -- Other Terms and Conditions.
343
+
344
+ a. The Licensor shall not be bound by any additional or different
345
+ terms or conditions communicated by You unless expressly agreed.
346
+
347
+ b. Any arrangements, understandings, or agreements regarding the
348
+ Licensed Material not stated herein are separate from and
349
+ independent of the terms and conditions of this Public License.
350
+
351
+
352
+ Section 8 -- Interpretation.
353
+
354
+ a. For the avoidance of doubt, this Public License does not, and
355
+ shall not be interpreted to, reduce, limit, restrict, or impose
356
+ conditions on any use of the Licensed Material that could lawfully
357
+ be made without permission under this Public License.
358
+
359
+ b. To the extent possible, if any provision of this Public License is
360
+ deemed unenforceable, it shall be automatically reformed to the
361
+ minimum extent necessary to make it enforceable. If the provision
362
+ cannot be reformed, it shall be severed from this Public License
363
+ without affecting the enforceability of the remaining terms and
364
+ conditions.
365
+
366
+ c. No term or condition of this Public License will be waived and no
367
+ failure to comply consented to unless expressly agreed to by the
368
+ Licensor.
369
+
370
+ d. Nothing in this Public License constitutes or may be interpreted
371
+ as a limitation upon, or waiver of, any privileges and immunities
372
+ that apply to the Licensor or You, including from the legal
373
+ processes of any jurisdiction or authority.
374
+
375
+
376
+ =======================================================================
377
+
378
+ Creative Commons is not a party to its public
379
+ licenses. Notwithstanding, Creative Commons may elect to apply one of
380
+ its public licenses to material it publishes and in those instances
381
+ will be considered the “Licensor.” The text of the Creative Commons
382
+ public licenses is dedicated to the public domain under the CC0 Public
383
+ Domain Dedication. Except for the limited purpose of indicating that
384
+ material is shared under a Creative Commons public license or as
385
+ otherwise permitted by the Creative Commons policies published at
386
+ creativecommons.org/policies, Creative Commons does not authorize the
387
+ use of the trademark "Creative Commons" or any other trademark or logo
388
+ of Creative Commons without its prior written consent including,
389
+ without limitation, in connection with any unauthorized modifications
390
+ to any of its public licenses or any other arrangements,
391
+ understandings, or agreements concerning use of licensed material. For
392
+ the avoidance of doubt, this paragraph does not form part of the
393
+ public licenses.
394
+
395
+ Creative Commons may be contacted at creativecommons.org.
acoustic/croatian/mfa/v2.0.0/README.md ADDED
@@ -0,0 +1,128 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+
2
+ # Croatian MFA acoustic model v2.0.0
3
+
4
+ [Link to documentation on mfa-models](https://mfa-models.readthedocs.io/en/main/acoustic/croatian_mfa.html)
5
+
6
+ Jump to section:
7
+
8
+ - [Model details](#model-details)
9
+ - [Installation](#installation)
10
+ - [Intended use](#intended-use)
11
+ - [Performance Factors](#performance-factors)
12
+ - [Metrics](#metrics)
13
+ - [Ethical considerations](#ethical-considerations)
14
+ - [Troubleshooting issues](#troubleshooting-issues)
15
+ - [Training data](#training-data)
16
+ - [Evaluation data](#evaluation-data)
17
+
18
+ ## Model details
19
+
20
+ - **Maintainer:** [Montreal Forced Aligner](https://montreal-forced-aligner.readthedocs.io/)
21
+ - **Language:** [Serbo-Croatian](https://en.wikipedia.org/wiki/Serbo-Croatian)
22
+ - **Dialect:** N/A
23
+ - **Phone set:** [MFA](https://mfa-models.readthedocs.io/en/refactor/mfa_phone_set.html#croatian)
24
+ - **Model type:** `Acoustic`
25
+ - **Features:** `MFCC + pitch`
26
+ - **Architecture:** `gmm-hmm`
27
+ - **Model version:** `v2.0.0`
28
+ - **Trained date:** `2022-03-21`
29
+ - **Compatible MFA version:** `v2.0.0`
30
+ - **License:** [CC BY 4.0](https://github.com/MontrealCorpusTools/mfa-models/tree/main/acoustic/croatian/mfa/v2.0.0/LICENSE)
31
+ - **Citation:**
32
+
33
+ ```bibtex
34
+ @techreport{mfa_croatian_mfa_acoustic_2022,
35
+ author={McAuliffe, Michael and Sonderegger, Morgan},
36
+ title={Croatian MFA acoustic model v2.0.0},
37
+ address={\url{https://mfa-models.readthedocs.io/acoustic/Croatian/Croatian MFA acoustic model v2_0_0.html}},
38
+ year={2022},
39
+ month={Mar},
40
+ }
41
+ ```
42
+
43
+ - If you have comments or questions about this model, you can check [previous MFA model discussion posts](https://github.com/MontrealCorpusTools/mfa-models/discussions?discussions_q=Croatian+MFA+acoustic+model+v2.0.0) or create [a new one](https://github.com/MontrealCorpusTools/mfa-models/discussions/new).
44
+
45
+ ## Installation
46
+
47
+ Install from the [MFA command line](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/models/index.html):
48
+
49
+ ```
50
+ mfa model download acoustic croatian_mfa
51
+ ```
52
+
53
+ Or download from [the release page](https://github.com/MontrealCorpusTools/mfa-models/releases/tag/acoustic-croatian_mfa-v2.0.0).
54
+
55
+ ## Intended use
56
+
57
+ This model is intended for forced alignment of [Serbo-Croatian](https://en.wikipedia.org/wiki/Serbo-Croatian) transcripts.
58
+
59
+ This model uses the [MFA](https://mfa-models.readthedocs.io/en/refactor/mfa_phone_set.html#croatian) phone set for Croatian, and was trained with the pronunciation dictionaries above. Pronunciations can be added on top of the dictionary, as long as no additional phones are introduced.
60
+
61
+ ## Performance Factors
62
+
63
+ As forced alignment is a relatively well-constrained problem (given accurate transcripts), this model should be applicable to a range of recording conditions and speakers. However, please note that it was trained on read speech in low-noise environments, so as your data diverges from that, you may run into alignment issues or need to [increase the beam size of MFA](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/configuration/#configuring-specific-commands) or see other recommendations in the [troubleshooting section below](#troubleshooting-issues).
64
+
65
+ Please note as well that MFA does not use state-of-the-art ASR models for forced alignment. You may get better performance (especially on speech-to-text tasks) using other frameworks like [Coqui](https://coqui.ai/).
66
+
67
+ ## Metrics
68
+
69
+ Acoustic models are typically generated as one component of a larger ASR system where the metric is word error rate (WER). For forced alignment, there is typically not the same sort of gold standard measure for most languages.
70
+
71
+ As a rough approximation of the acoustic model quality, we evaluated it against the corpus it was trained on alongside a language model trained from the same data. Key caveat here is that this is not a typical WER measure on held out data, so it should not be taken as a hard measure of how well an acoustic model will generalize to your data, but rather is more of a sanity check that the training data quality was sufficiently high.
72
+
73
+ Using the pronunciation dictionary above and a language model trained on the GlobalPhone Croatian corpus (see the [GlobalPhone benchmark analysis page](https://montreal-forced-aligner.readthedocs.io/en/latest/benchmarks/globalphone_transcriptions.html) for more details):
74
+
75
+ - **WER:** `6.5%`
76
+ - **CER:** `3.6%`
77
+
78
+ ## Ethical considerations
79
+
80
+ Deploying any Speech-to-Text model into any production setting has ethical implications. You should consider these implications before use.
81
+
82
+ ### Demographic Bias
83
+
84
+ You should assume every machine learning model has demographic bias unless proven otherwise. For STT models, it is often the case that transcription accuracy is better for men than it is for women. If you are using this model in production, you should acknowledge this as a potential issue.
85
+
86
+ ### Surveillance
87
+
88
+ Speech-to-Text technologies may be misused to invade the privacy of others by recording and mining information from private conversations. This kind of individual privacy is protected by law in many countries. You should not assume consent to record and analyze private speech.
89
+
90
+
91
+ ## Troubleshooting issues
92
+
93
+ Machine learning models (like this acoustic model) perform best on data that is similar to the data on which they were trained.
94
+
95
+ The primary sources of variability in forced alignment will be the applicability of the pronunciation dictionary and how similar the speech, demographics, and recording conditions are. If you encounter issues in alignment, there are couple of avenues to improve performance:
96
+
97
+ 1. [Increase the beam size of MFA](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/configuration/#configuring-specific-commands)
98
+
99
+ * MFA defaults to a narrow beam to ensure quick alignment and also as a way to detect potential issues in your dataset, but depending on your data, you might benefit from boosting the beam to 100 or higher.
100
+
101
+ 2. Add pronunciations to the pronunciation dictionary
102
+
103
+ * This model was trained a particular dialect/style, and so adding pronunciations more representative of the variety spoken in your dataset will help alignment.
104
+
105
+ 3. Check the quality of your data
106
+
107
+ * MFA includes a [validator utility](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/data_validation.html), which aims to detect issues in the dataset.
108
+ * Use MFA's [anchor utility](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/workflows/anchor.html) to visually inspect your data as MFA sees it and correct issues in transcription or OOV items.
109
+
110
+ 4. Adapt the model to your data
111
+
112
+ * MFA has an [adaptation command](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/workflows/adapt_acoustic_model.html) to adapt some of the model to your data based on an initial alignment, and then run another alignment with the adapted model.
113
+
114
+ ## Training data
115
+
116
+ This model was trained on the following corpora:
117
+
118
+
119
+
120
+ * [Common Voice Serbian](../../../../corpus/croatian/common_voice_serbian/8.0/README.md):
121
+ * **Hours:** `1.50`
122
+ * **Speakers:** `50`
123
+ * **Utterances:** `1,878`
124
+
125
+ * [GlobalPhone Croatian](../../../../corpus/croatian/globalphone_croatian/3.1/README.md):
126
+ * **Hours:** `15.89`
127
+ * **Speakers:** `92`
128
+ * **Utterances:** `4,499`
acoustic/croatian/mfa/v2.0.0/meta.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "name": "croatian_mfa",
3
+ "language": "Croatian",
4
+ "dialect": "",
5
+ "phone_set": "MFA",
6
+ "version": "2.0.0",
7
+ "maintainer": "[Montreal Forced Aligner](https://montreal-forced-aligner.readthedocs.io/)",
8
+ "citation": "@techreport{mfa_croatian_mfa_acoustic_2022,\n\tauthor={McAuliffe, Michael and Sonderegger, Morgan},\n\ttitle={Croatian MFA acoustic model v2.0.0},\n\taddress={\\url{https://mfa-models.readthedocs.io/acoustic/Croatian/Croatian MFA acoustic model v2_0_0.html}},\n\tyear={2022},\n\tmonth={Mar},\n}",
9
+ "license": "CC BY 4.0",
10
+ "license_link": "[CC BY 4.0](https://github.com/MontrealCorpusTools/mfa-models/tree/main/acoustic/croatian/mfa/v2.0.0/LICENSE)",
11
+ "architecture": "gmm-hmm",
12
+ "features": "MFCC + pitch",
13
+ "evaluation": {},
14
+ "decode": {},
15
+ "train_date": "2022-03-21"
16
+ }
acoustic/croatian/mfa/v2.0.0a/LICENSE ADDED
@@ -0,0 +1,395 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ Attribution 4.0 International
2
+
3
+ =======================================================================
4
+
5
+ Creative Commons Corporation ("Creative Commons") is not a law firm and
6
+ does not provide legal services or legal advice. Distribution of
7
+ Creative Commons public licenses does not create a lawyer-client or
8
+ other relationship. Creative Commons makes its licenses and related
9
+ information available on an "as-is" basis. Creative Commons gives no
10
+ warranties regarding its licenses, any material licensed under their
11
+ terms and conditions, or any related information. Creative Commons
12
+ disclaims all liability for damages resulting from their use to the
13
+ fullest extent possible.
14
+
15
+ Using Creative Commons Public Licenses
16
+
17
+ Creative Commons public licenses provide a standard set of terms and
18
+ conditions that creators and other rights holders may use to share
19
+ original works of authorship and other material subject to copyright
20
+ and certain other rights specified in the public license below. The
21
+ following considerations are for informational purposes only, are not
22
+ exhaustive, and do not form part of our licenses.
23
+
24
+ Considerations for licensors: Our public licenses are
25
+ intended for use by those authorized to give the public
26
+ permission to use material in ways otherwise restricted by
27
+ copyright and certain other rights. Our licenses are
28
+ irrevocable. Licensors should read and understand the terms
29
+ and conditions of the license they choose before applying it.
30
+ Licensors should also secure all rights necessary before
31
+ applying our licenses so that the public can reuse the
32
+ material as expected. Licensors should clearly mark any
33
+ material not subject to the license. This includes other CC-
34
+ licensed material, or material used under an exception or
35
+ limitation to copyright. More considerations for licensors:
36
+ wiki.creativecommons.org/Considerations_for_licensors
37
+
38
+ Considerations for the public: By using one of our public
39
+ licenses, a licensor grants the public permission to use the
40
+ licensed material under specified terms and conditions. If
41
+ the licensor's permission is not necessary for any reason--for
42
+ example, because of any applicable exception or limitation to
43
+ copyright--then that use is not regulated by the license. Our
44
+ licenses grant only permissions under copyright and certain
45
+ other rights that a licensor has authority to grant. Use of
46
+ the licensed material may still be restricted for other
47
+ reasons, including because others have copyright or other
48
+ rights in the material. A licensor may make special requests,
49
+ such as asking that all changes be marked or described.
50
+ Although not required by our licenses, you are encouraged to
51
+ respect those requests where reasonable. More_considerations
52
+ for the public:
53
+ wiki.creativecommons.org/Considerations_for_licensees
54
+
55
+ =======================================================================
56
+
57
+ Creative Commons Attribution 4.0 International Public License
58
+
59
+ By exercising the Licensed Rights (defined below), You accept and agree
60
+ to be bound by the terms and conditions of this Creative Commons
61
+ Attribution 4.0 International Public License ("Public License"). To the
62
+ extent this Public License may be interpreted as a contract, You are
63
+ granted the Licensed Rights in consideration of Your acceptance of
64
+ these terms and conditions, and the Licensor grants You such rights in
65
+ consideration of benefits the Licensor receives from making the
66
+ Licensed Material available under these terms and conditions.
67
+
68
+
69
+ Section 1 -- Definitions.
70
+
71
+ a. Adapted Material means material subject to Copyright and Similar
72
+ Rights that is derived from or based upon the Licensed Material
73
+ and in which the Licensed Material is translated, altered,
74
+ arranged, transformed, or otherwise modified in a manner requiring
75
+ permission under the Copyright and Similar Rights held by the
76
+ Licensor. For purposes of this Public License, where the Licensed
77
+ Material is a musical work, performance, or sound recording,
78
+ Adapted Material is always produced where the Licensed Material is
79
+ synched in timed relation with a moving image.
80
+
81
+ b. Adapter's License means the license You apply to Your Copyright
82
+ and Similar Rights in Your contributions to Adapted Material in
83
+ accordance with the terms and conditions of this Public License.
84
+
85
+ c. Copyright and Similar Rights means copyright and/or similar rights
86
+ closely related to copyright including, without limitation,
87
+ performance, broadcast, sound recording, and Sui Generis Database
88
+ Rights, without regard to how the rights are labeled or
89
+ categorized. For purposes of this Public License, the rights
90
+ specified in Section 2(b)(1)-(2) are not Copyright and Similar
91
+ Rights.
92
+
93
+ d. Effective Technological Measures means those measures that, in the
94
+ absence of proper authority, may not be circumvented under laws
95
+ fulfilling obligations under Article 11 of the WIPO Copyright
96
+ Treaty adopted on December 20, 1996, and/or similar international
97
+ agreements.
98
+
99
+ e. Exceptions and Limitations means fair use, fair dealing, and/or
100
+ any other exception or limitation to Copyright and Similar Rights
101
+ that applies to Your use of the Licensed Material.
102
+
103
+ f. Licensed Material means the artistic or literary work, database,
104
+ or other material to which the Licensor applied this Public
105
+ License.
106
+
107
+ g. Licensed Rights means the rights granted to You subject to the
108
+ terms and conditions of this Public License, which are limited to
109
+ all Copyright and Similar Rights that apply to Your use of the
110
+ Licensed Material and that the Licensor has authority to license.
111
+
112
+ h. Licensor means the individual(s) or entity(ies) granting rights
113
+ under this Public License.
114
+
115
+ i. Share means to provide material to the public by any means or
116
+ process that requires permission under the Licensed Rights, such
117
+ as reproduction, public display, public performance, distribution,
118
+ dissemination, communication, or importation, and to make material
119
+ available to the public including in ways that members of the
120
+ public may access the material from a place and at a time
121
+ individually chosen by them.
122
+
123
+ j. Sui Generis Database Rights means rights other than copyright
124
+ resulting from Directive 96/9/EC of the European Parliament and of
125
+ the Council of 11 March 1996 on the legal protection of databases,
126
+ as amended and/or succeeded, as well as other essentially
127
+ equivalent rights anywhere in the world.
128
+
129
+ k. You means the individual or entity exercising the Licensed Rights
130
+ under this Public License. Your has a corresponding meaning.
131
+
132
+
133
+ Section 2 -- Scope.
134
+
135
+ a. License grant.
136
+
137
+ 1. Subject to the terms and conditions of this Public License,
138
+ the Licensor hereby grants You a worldwide, royalty-free,
139
+ non-sublicensable, non-exclusive, irrevocable license to
140
+ exercise the Licensed Rights in the Licensed Material to:
141
+
142
+ a. reproduce and Share the Licensed Material, in whole or
143
+ in part; and
144
+
145
+ b. produce, reproduce, and Share Adapted Material.
146
+
147
+ 2. Exceptions and Limitations. For the avoidance of doubt, where
148
+ Exceptions and Limitations apply to Your use, this Public
149
+ License does not apply, and You do not need to comply with
150
+ its terms and conditions.
151
+
152
+ 3. Term. The term of this Public License is specified in Section
153
+ 6(a).
154
+
155
+ 4. Media and formats; technical modifications allowed. The
156
+ Licensor authorizes You to exercise the Licensed Rights in
157
+ all media and formats whether now known or hereafter created,
158
+ and to make technical modifications necessary to do so. The
159
+ Licensor waives and/or agrees not to assert any right or
160
+ authority to forbid You from making technical modifications
161
+ necessary to exercise the Licensed Rights, including
162
+ technical modifications necessary to circumvent Effective
163
+ Technological Measures. For purposes of this Public License,
164
+ simply making modifications authorized by this Section 2(a)
165
+ (4) never produces Adapted Material.
166
+
167
+ 5. Downstream recipients.
168
+
169
+ a. Offer from the Licensor -- Licensed Material. Every
170
+ recipient of the Licensed Material automatically
171
+ receives an offer from the Licensor to exercise the
172
+ Licensed Rights under the terms and conditions of this
173
+ Public License.
174
+
175
+ b. No downstream restrictions. You may not offer or impose
176
+ any additional or different terms or conditions on, or
177
+ apply any Effective Technological Measures to, the
178
+ Licensed Material if doing so restricts exercise of the
179
+ Licensed Rights by any recipient of the Licensed
180
+ Material.
181
+
182
+ 6. No endorsement. Nothing in this Public License constitutes or
183
+ may be construed as permission to assert or imply that You
184
+ are, or that Your use of the Licensed Material is, connected
185
+ with, or sponsored, endorsed, or granted official status by,
186
+ the Licensor or others designated to receive attribution as
187
+ provided in Section 3(a)(1)(A)(i).
188
+
189
+ b. Other rights.
190
+
191
+ 1. Moral rights, such as the right of integrity, are not
192
+ licensed under this Public License, nor are publicity,
193
+ privacy, and/or other similar personality rights; however, to
194
+ the extent possible, the Licensor waives and/or agrees not to
195
+ assert any such rights held by the Licensor to the limited
196
+ extent necessary to allow You to exercise the Licensed
197
+ Rights, but not otherwise.
198
+
199
+ 2. Patent and trademark rights are not licensed under this
200
+ Public License.
201
+
202
+ 3. To the extent possible, the Licensor waives any right to
203
+ collect royalties from You for the exercise of the Licensed
204
+ Rights, whether directly or through a collecting society
205
+ under any voluntary or waivable statutory or compulsory
206
+ licensing scheme. In all other cases the Licensor expressly
207
+ reserves any right to collect such royalties.
208
+
209
+
210
+ Section 3 -- License Conditions.
211
+
212
+ Your exercise of the Licensed Rights is expressly made subject to the
213
+ following conditions.
214
+
215
+ a. Attribution.
216
+
217
+ 1. If You Share the Licensed Material (including in modified
218
+ form), You must:
219
+
220
+ a. retain the following if it is supplied by the Licensor
221
+ with the Licensed Material:
222
+
223
+ i. identification of the creator(s) of the Licensed
224
+ Material and any others designated to receive
225
+ attribution, in any reasonable manner requested by
226
+ the Licensor (including by pseudonym if
227
+ designated);
228
+
229
+ ii. a copyright notice;
230
+
231
+ iii. a notice that refers to this Public License;
232
+
233
+ iv. a notice that refers to the disclaimer of
234
+ warranties;
235
+
236
+ v. a URI or hyperlink to the Licensed Material to the
237
+ extent reasonably practicable;
238
+
239
+ b. indicate if You modified the Licensed Material and
240
+ retain an indication of any previous modifications; and
241
+
242
+ c. indicate the Licensed Material is licensed under this
243
+ Public License, and include the text of, or the URI or
244
+ hyperlink to, this Public License.
245
+
246
+ 2. You may satisfy the conditions in Section 3(a)(1) in any
247
+ reasonable manner based on the medium, means, and context in
248
+ which You Share the Licensed Material. For example, it may be
249
+ reasonable to satisfy the conditions by providing a URI or
250
+ hyperlink to a resource that includes the required
251
+ information.
252
+
253
+ 3. If requested by the Licensor, You must remove any of the
254
+ information required by Section 3(a)(1)(A) to the extent
255
+ reasonably practicable.
256
+
257
+ 4. If You Share Adapted Material You produce, the Adapter's
258
+ License You apply must not prevent recipients of the Adapted
259
+ Material from complying with this Public License.
260
+
261
+
262
+ Section 4 -- Sui Generis Database Rights.
263
+
264
+ Where the Licensed Rights include Sui Generis Database Rights that
265
+ apply to Your use of the Licensed Material:
266
+
267
+ a. for the avoidance of doubt, Section 2(a)(1) grants You the right
268
+ to extract, reuse, reproduce, and Share all or a substantial
269
+ portion of the contents of the database;
270
+
271
+ b. if You include all or a substantial portion of the database
272
+ contents in a database in which You have Sui Generis Database
273
+ Rights, then the database in which You have Sui Generis Database
274
+ Rights (but not its individual contents) is Adapted Material; and
275
+
276
+ c. You must comply with the conditions in Section 3(a) if You Share
277
+ all or a substantial portion of the contents of the database.
278
+
279
+ For the avoidance of doubt, this Section 4 supplements and does not
280
+ replace Your obligations under this Public License where the Licensed
281
+ Rights include other Copyright and Similar Rights.
282
+
283
+
284
+ Section 5 -- Disclaimer of Warranties and Limitation of Liability.
285
+
286
+ a. UNLESS OTHERWISE SEPARATELY UNDERTAKEN BY THE LICENSOR, TO THE
287
+ EXTENT POSSIBLE, THE LICENSOR OFFERS THE LICENSED MATERIAL AS-IS
288
+ AND AS-AVAILABLE, AND MAKES NO REPRESENTATIONS OR WARRANTIES OF
289
+ ANY KIND CONCERNING THE LICENSED MATERIAL, WHETHER EXPRESS,
290
+ IMPLIED, STATUTORY, OR OTHER. THIS INCLUDES, WITHOUT LIMITATION,
291
+ WARRANTIES OF TITLE, MERCHANTABILITY, FITNESS FOR A PARTICULAR
292
+ PURPOSE, NON-INFRINGEMENT, ABSENCE OF LATENT OR OTHER DEFECTS,
293
+ ACCURACY, OR THE PRESENCE OR ABSENCE OF ERRORS, WHETHER OR NOT
294
+ KNOWN OR DISCOVERABLE. WHERE DISCLAIMERS OF WARRANTIES ARE NOT
295
+ ALLOWED IN FULL OR IN PART, THIS DISCLAIMER MAY NOT APPLY TO YOU.
296
+
297
+ b. TO THE EXTENT POSSIBLE, IN NO EVENT WILL THE LICENSOR BE LIABLE
298
+ TO YOU ON ANY LEGAL THEORY (INCLUDING, WITHOUT LIMITATION,
299
+ NEGLIGENCE) OR OTHERWISE FOR ANY DIRECT, SPECIAL, INDIRECT,
300
+ INCIDENTAL, CONSEQUENTIAL, PUNITIVE, EXEMPLARY, OR OTHER LOSSES,
301
+ COSTS, EXPENSES, OR DAMAGES ARISING OUT OF THIS PUBLIC LICENSE OR
302
+ USE OF THE LICENSED MATERIAL, EVEN IF THE LICENSOR HAS BEEN
303
+ ADVISED OF THE POSSIBILITY OF SUCH LOSSES, COSTS, EXPENSES, OR
304
+ DAMAGES. WHERE A LIMITATION OF LIABILITY IS NOT ALLOWED IN FULL OR
305
+ IN PART, THIS LIMITATION MAY NOT APPLY TO YOU.
306
+
307
+ c. The disclaimer of warranties and limitation of liability provided
308
+ above shall be interpreted in a manner that, to the extent
309
+ possible, most closely approximates an absolute disclaimer and
310
+ waiver of all liability.
311
+
312
+
313
+ Section 6 -- Term and Termination.
314
+
315
+ a. This Public License applies for the term of the Copyright and
316
+ Similar Rights licensed here. However, if You fail to comply with
317
+ this Public License, then Your rights under this Public License
318
+ terminate automatically.
319
+
320
+ b. Where Your right to use the Licensed Material has terminated under
321
+ Section 6(a), it reinstates:
322
+
323
+ 1. automatically as of the date the violation is cured, provided
324
+ it is cured within 30 days of Your discovery of the
325
+ violation; or
326
+
327
+ 2. upon express reinstatement by the Licensor.
328
+
329
+ For the avoidance of doubt, this Section 6(b) does not affect any
330
+ right the Licensor may have to seek remedies for Your violations
331
+ of this Public License.
332
+
333
+ c. For the avoidance of doubt, the Licensor may also offer the
334
+ Licensed Material under separate terms or conditions or stop
335
+ distributing the Licensed Material at any time; however, doing so
336
+ will not terminate this Public License.
337
+
338
+ d. Sections 1, 5, 6, 7, and 8 survive termination of this Public
339
+ License.
340
+
341
+
342
+ Section 7 -- Other Terms and Conditions.
343
+
344
+ a. The Licensor shall not be bound by any additional or different
345
+ terms or conditions communicated by You unless expressly agreed.
346
+
347
+ b. Any arrangements, understandings, or agreements regarding the
348
+ Licensed Material not stated herein are separate from and
349
+ independent of the terms and conditions of this Public License.
350
+
351
+
352
+ Section 8 -- Interpretation.
353
+
354
+ a. For the avoidance of doubt, this Public License does not, and
355
+ shall not be interpreted to, reduce, limit, restrict, or impose
356
+ conditions on any use of the Licensed Material that could lawfully
357
+ be made without permission under this Public License.
358
+
359
+ b. To the extent possible, if any provision of this Public License is
360
+ deemed unenforceable, it shall be automatically reformed to the
361
+ minimum extent necessary to make it enforceable. If the provision
362
+ cannot be reformed, it shall be severed from this Public License
363
+ without affecting the enforceability of the remaining terms and
364
+ conditions.
365
+
366
+ c. No term or condition of this Public License will be waived and no
367
+ failure to comply consented to unless expressly agreed to by the
368
+ Licensor.
369
+
370
+ d. Nothing in this Public License constitutes or may be interpreted
371
+ as a limitation upon, or waiver of, any privileges and immunities
372
+ that apply to the Licensor or You, including from the legal
373
+ processes of any jurisdiction or authority.
374
+
375
+
376
+ =======================================================================
377
+
378
+ Creative Commons is not a party to its public
379
+ licenses. Notwithstanding, Creative Commons may elect to apply one of
380
+ its public licenses to material it publishes and in those instances
381
+ will be considered the “Licensor.” The text of the Creative Commons
382
+ public licenses is dedicated to the public domain under the CC0 Public
383
+ Domain Dedication. Except for the limited purpose of indicating that
384
+ material is shared under a Creative Commons public license or as
385
+ otherwise permitted by the Creative Commons policies published at
386
+ creativecommons.org/policies, Creative Commons does not authorize the
387
+ use of the trademark "Creative Commons" or any other trademark or logo
388
+ of Creative Commons without its prior written consent including,
389
+ without limitation, in connection with any unauthorized modifications
390
+ to any of its public licenses or any other arrangements,
391
+ understandings, or agreements concerning use of licensed material. For
392
+ the avoidance of doubt, this paragraph does not form part of the
393
+ public licenses.
394
+
395
+ Creative Commons may be contacted at creativecommons.org.
acoustic/croatian/mfa/v2.0.0a/README.md ADDED
@@ -0,0 +1,128 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+
2
+ # Croatian MFA acoustic model v2.0.0a
3
+
4
+ [Link to documentation on mfa-models](https://mfa-models.readthedocs.io/en/main/acoustic/croatian_mfa.html)
5
+
6
+ Jump to section:
7
+
8
+ - [Model details](#model-details)
9
+ - [Installation](#installation)
10
+ - [Intended use](#intended-use)
11
+ - [Performance Factors](#performance-factors)
12
+ - [Metrics](#metrics)
13
+ - [Ethical considerations](#ethical-considerations)
14
+ - [Troubleshooting issues](#troubleshooting-issues)
15
+ - [Training data](#training-data)
16
+ - [Evaluation data](#evaluation-data)
17
+
18
+ ## Model details
19
+
20
+ - **Maintainer:** [Montreal Forced Aligner](https://montreal-forced-aligner.readthedocs.io/)
21
+ - **Language:** [Serbo-Croatian](https://en.wikipedia.org/wiki/Serbo-Croatian)
22
+ - **Dialect:** N/A
23
+ - **Phone set:** [MFA](https://mfa-models.readthedocs.io/en/refactor/mfa_phone_set.html#croatian)
24
+ - **Model type:** `Acoustic`
25
+ - **Features:** `MFCC + pitch`
26
+ - **Architecture:** `gmm-hmm`
27
+ - **Model version:** `v2.0.0a`
28
+ - **Trained date:** `2022-05-18`
29
+ - **Compatible MFA version:** `v2.0.0`
30
+ - **License:** [CC BY 4.0](https://github.com/MontrealCorpusTools/mfa-models/tree/main/acoustic/croatian/mfa/v2.0.0a/LICENSE)
31
+ - **Citation:**
32
+
33
+ ```bibtex
34
+ @techreport{mfa_croatian_mfa_acoustic_2022,
35
+ author={McAuliffe, Michael and Sonderegger, Morgan},
36
+ title={Croatian MFA acoustic model v2.0.0a},
37
+ address={\url{https://mfa-models.readthedocs.io/acoustic/Croatian/Croatian MFA acoustic model v2_0_0a.html}},
38
+ year={2022},
39
+ month={May},
40
+ }
41
+ ```
42
+
43
+ - If you have comments or questions about this model, you can check [previous MFA model discussion posts](https://github.com/MontrealCorpusTools/mfa-models/discussions?discussions_q=Croatian+MFA+acoustic+model+v2.0.0a) or create [a new one](https://github.com/MontrealCorpusTools/mfa-models/discussions/new).
44
+
45
+ ## Installation
46
+
47
+ Install from the [MFA command line](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/models/index.html):
48
+
49
+ ```
50
+ mfa model download acoustic croatian_mfa
51
+ ```
52
+
53
+ Or download from [the release page](https://github.com/MontrealCorpusTools/mfa-models/releases/tag/acoustic-croatian_mfa-v2.0.0a).
54
+
55
+ ## Intended use
56
+
57
+ This model is intended for forced alignment of [Serbo-Croatian](https://en.wikipedia.org/wiki/Serbo-Croatian) transcripts.
58
+
59
+ This model uses the [MFA](https://mfa-models.readthedocs.io/en/refactor/mfa_phone_set.html#croatian) phone set for Croatian, and was trained with the pronunciation dictionaries above. Pronunciations can be added on top of the dictionary, as long as no additional phones are introduced.
60
+
61
+ ## Performance Factors
62
+
63
+ As forced alignment is a relatively well-constrained problem (given accurate transcripts), this model should be applicable to a range of recording conditions and speakers. However, please note that it was trained on read speech in low-noise environments, so as your data diverges from that, you may run into alignment issues or need to [increase the beam size of MFA](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/configuration/#configuring-specific-commands) or see other recommendations in the [troubleshooting section below](#troubleshooting-issues).
64
+
65
+ Please note as well that MFA does not use state-of-the-art ASR models for forced alignment. You may get better performance (especially on speech-to-text tasks) using other frameworks like [Coqui](https://coqui.ai/).
66
+
67
+ ## Metrics
68
+
69
+ Acoustic models are typically generated as one component of a larger ASR system where the metric is word error rate (WER). For forced alignment, there is typically not the same sort of gold standard measure for most languages.
70
+
71
+ As a rough approximation of the acoustic model quality, we evaluated it against the corpus it was trained on alongside a language model trained from the same data. Key caveat here is that this is not a typical WER measure on held out data, so it should not be taken as a hard measure of how well an acoustic model will generalize to your data, but rather is more of a sanity check that the training data quality was sufficiently high.
72
+
73
+ Using the pronunciation dictionary above and a language model trained on the GlobalPhone Croatian corpus (see the [GlobalPhone benchmark analysis page](https://montreal-forced-aligner.readthedocs.io/en/latest/benchmarks/globalphone_transcriptions.html) for more details):
74
+
75
+ - **WER:** `6.4%`
76
+ - **CER:** `3.4%`
77
+
78
+ ## Ethical considerations
79
+
80
+ Deploying any Speech-to-Text model into any production setting has ethical implications. You should consider these implications before use.
81
+
82
+ ### Demographic Bias
83
+
84
+ You should assume every machine learning model has demographic bias unless proven otherwise. For STT models, it is often the case that transcription accuracy is better for men than it is for women. If you are using this model in production, you should acknowledge this as a potential issue.
85
+
86
+ ### Surveillance
87
+
88
+ Speech-to-Text technologies may be misused to invade the privacy of others by recording and mining information from private conversations. This kind of individual privacy is protected by law in many countries. You should not assume consent to record and analyze private speech.
89
+
90
+
91
+ ## Troubleshooting issues
92
+
93
+ Machine learning models (like this acoustic model) perform best on data that is similar to the data on which they were trained.
94
+
95
+ The primary sources of variability in forced alignment will be the applicability of the pronunciation dictionary and how similar the speech, demographics, and recording conditions are. If you encounter issues in alignment, there are couple of avenues to improve performance:
96
+
97
+ 1. [Increase the beam size of MFA](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/configuration/#configuring-specific-commands)
98
+
99
+ * MFA defaults to a narrow beam to ensure quick alignment and also as a way to detect potential issues in your dataset, but depending on your data, you might benefit from boosting the beam to 100 or higher.
100
+
101
+ 2. Add pronunciations to the pronunciation dictionary
102
+
103
+ * This model was trained a particular dialect/style, and so adding pronunciations more representative of the variety spoken in your dataset will help alignment.
104
+
105
+ 3. Check the quality of your data
106
+
107
+ * MFA includes a [validator utility](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/data_validation.html), which aims to detect issues in the dataset.
108
+ * Use MFA's [anchor utility](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/workflows/anchor.html) to visually inspect your data as MFA sees it and correct issues in transcription or OOV items.
109
+
110
+ 4. Adapt the model to your data
111
+
112
+ * MFA has an [adaptation command](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/workflows/adapt_acoustic_model.html) to adapt some of the model to your data based on an initial alignment, and then run another alignment with the adapted model.
113
+
114
+ ## Training data
115
+
116
+ This model was trained on the following corpora:
117
+
118
+
119
+
120
+ * [Common Voice Serbian](../../../../corpus/croatian/common_voice_serbian/8.0/README.md):
121
+ * **Hours:** `1.50`
122
+ * **Speakers:** `50`
123
+ * **Utterances:** `1,878`
124
+
125
+ * [GlobalPhone Croatian](../../../../corpus/croatian/globalphone_croatian/3.1/README.md):
126
+ * **Hours:** `15.89`
127
+ * **Speakers:** `92`
128
+ * **Utterances:** `4,499`
acoustic/croatian/mfa/v2.0.0a/meta.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "name": "croatian_mfa",
3
+ "language": "Croatian",
4
+ "dialect": "",
5
+ "phone_set": "MFA",
6
+ "version": "2.0.0a",
7
+ "maintainer": "[Montreal Forced Aligner](https://montreal-forced-aligner.readthedocs.io/)",
8
+ "citation": "@techreport{mfa_croatian_mfa_acoustic_2022,\n\tauthor={McAuliffe, Michael and Sonderegger, Morgan},\n\ttitle={Croatian MFA acoustic model v2.0.0a},\n\taddress={\\url{https://mfa-models.readthedocs.io/acoustic/Croatian/Croatian MFA acoustic model v2_0_0a.html}},\n\tyear={2022},\n\tmonth={May},\n}",
9
+ "license": "CC BY 4.0",
10
+ "license_link": "[CC BY 4.0](https://github.com/MontrealCorpusTools/mfa-models/tree/main/acoustic/croatian/mfa/v2.0.0a/LICENSE)",
11
+ "architecture": "gmm-hmm",
12
+ "features": "MFCC + pitch",
13
+ "evaluation": {},
14
+ "decode": {},
15
+ "train_date": "2022-05-18"
16
+ }
acoustic/czech.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b8e7112879afccb6cdbb16864087ad632c7cd18dd9fc45e85b4e457fc8d721f7
3
+ size 14670658
acoustic/czech/cv/v2.0.0/README.md ADDED
@@ -0,0 +1,110 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+
2
+ # Czech CV acoustic model v2.0.0
3
+
4
+ [Link to documentation on mfa-models](https://mfa-models.readthedocs.io/en/main/acoustic/czech_cv.html)
5
+
6
+ Jump to section:
7
+
8
+ - [Model details](#model-details)
9
+ - [Installation](#installation)
10
+ - [Intended use](#intended-use)
11
+ - [Performance Factors](#performance-factors)
12
+ - [Training data](#training-data)
13
+ - [Ethical considerations](#ethical-considerations)
14
+ - [Troubleshooting issues](#troubleshooting-issues)
15
+
16
+ ## Model details
17
+
18
+ - **Maintainer:** [Vox Communis](https://osf.io/t957v/)
19
+ - **Language:** [Czech](https://en.wikipedia.org/wiki/Czech_language)
20
+ - **Dialect:** N/A
21
+ - **Phone set:** [XPF](https://github.com/CohenPr-XPF/XPF)
22
+ - **Model type:** `Acoustic model`
23
+ - **Features:** `MFCC`
24
+ - **Architecture:** `gmm-hmm`
25
+ - **Model version:** `v2.0.0`
26
+ - **Trained date:** `02-11-2022`
27
+ - **Compatible MFA version:** `v2.0.0`
28
+ - **License:** [CC-0](https://creativecommons.org/publicdomain/zero/1.0/)
29
+ - **Citation:**
30
+
31
+ ```bibtex
32
+ @misc{Ahn_Chodroff_2022,
33
+ author={Ahn, Emily and Chodroff, Eleanor},
34
+ title={VoxCommunis Corpus},
35
+ address={\url{https://osf.io/t957v}},
36
+ publisher={OSF},
37
+ year={2022},
38
+ month={Jan}
39
+ }
40
+ ```
41
+
42
+ - If you have comments or questions about this model, you can check [previous MFA model discussion posts](https://github.com/MontrealCorpusTools/mfa-models/discussions?discussions_q=Czech+CV+acoustic+model+v2.0.0) or create [a new one](https://github.com/MontrealCorpusTools/mfa-models/discussions/new).
43
+
44
+ ## Installation
45
+
46
+ Install from the [MFA command line](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/models/index.html):
47
+
48
+ ```
49
+ mfa model download acoustic czech_cv
50
+ ```
51
+
52
+ Or download from [the release page](https://github.com/MontrealCorpusTools/mfa-models/releases/tag/acoustic-czech_cv-v2.0.0).
53
+
54
+ ## Intended use
55
+
56
+ This model is intended for forced alignment of [Czech](https://en.wikipedia.org/wiki/Czech_language) transcripts.
57
+
58
+ This model uses the [XPF](https://github.com/CohenPr-XPF/XPF) phone set for Czech, and was trained with the pronunciation dictionaries above. Pronunciations can be added on top of the dictionary, as long as no additional phones are introduced.
59
+
60
+ ## Performance Factors
61
+
62
+ As forced alignment is a relatively well-constrained problem (given accurate transcripts), this model should be applicable to a range of recording conditions and speakers. However, please note that it was trained on read speech in low-noise environments, so as your data diverges from that, you may run into alignment issues or need to [increase the beam size of MFA](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/configuration/#configuring-specific-commands) or see other recommendations in the [troubleshooting section below](#troubleshooting-issues).
63
+
64
+ Please note as well that MFA does not use state-of-the-art ASR models for forced alignment. You may get better performance (especially on speech-to-text tasks) using other frameworks like [Coqui](https://coqui.ai/).
65
+
66
+ ## Ethical considerations
67
+
68
+ Deploying any Speech-to-Text model into any production setting has ethical implications. You should consider these implications before use.
69
+
70
+ ### Demographic Bias
71
+
72
+ You should assume every machine learning model has demographic bias unless proven otherwise. For STT models, it is often the case that transcription accuracy is better for men than it is for women. If you are using this model in production, you should acknowledge this as a potential issue.
73
+
74
+ ### Surveillance
75
+
76
+ Speech-to-Text technologies may be misused to invade the privacy of others by recording and mining information from private conversations. This kind of individual privacy is protected by law in many countries. You should not assume consent to record and analyze private speech.
77
+
78
+ ## Troubleshooting issues
79
+
80
+ Machine learning models (like this acoustic model) perform best on data that is similar to the data on which they were trained.
81
+
82
+ The primary sources of variability in forced alignment will be the applicability of the pronunciation dictionary and how similar the speech, demographics, and recording conditions are. If you encounter issues in alignment, there are couple of avenues to improve performance:
83
+
84
+ 1. [Increase the beam size of MFA](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/configuration/#configuring-specific-commands)
85
+
86
+ * MFA defaults to a narrow beam to ensure quick alignment and also as a way to detect potential issues in your dataset, but depending on your data, you might benefit from boosting the beam to 100 or higher.
87
+
88
+ 2. Add pronunciations to the pronunciation dictionary
89
+
90
+ * This model was trained a particular dialect/style, and so adding pronunciations more representative of the variety spoken in your dataset will help alignment.
91
+
92
+ 3. Check the quality of your data
93
+
94
+ * MFA includes a [validator utility](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/data_validation.html), which aims to detect issues in the dataset.
95
+ * Use MFA's [anchor utility](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/workflows/anchor.html) to visually inspect your data as MFA sees it and correct issues in transcription or OOV items.
96
+
97
+ 4. Adapt the model to your data
98
+
99
+ * MFA has an [adaptation command](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/workflows/adapt_acoustic_model.html) to adapt some of the model to your data based on an initial alignment, and then run another alignment with the adapted model.
100
+
101
+ ## Training data
102
+
103
+ This model was trained on the following corpora:
104
+
105
+
106
+
107
+ * [Common Voice Czech](../../../../corpus/czech/common_voice_czech/7.0/README.md):
108
+ * **Hours:** `49.00`
109
+ * **Speakers:** `475`
110
+ * **Utterances:** `41,567`
acoustic/czech/cv/v2.0.0/meta.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "name": "czech_cv",
3
+ "language": "Czech",
4
+ "dialect": "",
5
+ "phone_set": "CV",
6
+ "version": "2.0.0",
7
+ "maintainer": "[Vox Communis](https://osf.io/t957v/)",
8
+ "citation": "@misc{Ahn_Chodroff_2022,\n\tauthor={Ahn, Emily and Chodroff, Eleanor},\n\ttitle={VoxCommunis Corpus},\n\taddress={\\url{https://osf.io/t957v}},\n\tpublisher={OSF},\n\tyear={2022}, \n\tmonth={Jan}\n}",
9
+ "license": "CC-0",
10
+ "license_link": "[CC-0](https://creativecommons.org/publicdomain/zero/1.0/)",
11
+ "architecture": "gmm-hmm",
12
+ "features": "MFCC",
13
+ "evaluation": {},
14
+ "decode": {},
15
+ "train_date": "02-11-2022"
16
+ }
acoustic/czech/mfa/v2.0.0/LICENSE ADDED
@@ -0,0 +1,395 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ Attribution 4.0 International
2
+
3
+ =======================================================================
4
+
5
+ Creative Commons Corporation ("Creative Commons") is not a law firm and
6
+ does not provide legal services or legal advice. Distribution of
7
+ Creative Commons public licenses does not create a lawyer-client or
8
+ other relationship. Creative Commons makes its licenses and related
9
+ information available on an "as-is" basis. Creative Commons gives no
10
+ warranties regarding its licenses, any material licensed under their
11
+ terms and conditions, or any related information. Creative Commons
12
+ disclaims all liability for damages resulting from their use to the
13
+ fullest extent possible.
14
+
15
+ Using Creative Commons Public Licenses
16
+
17
+ Creative Commons public licenses provide a standard set of terms and
18
+ conditions that creators and other rights holders may use to share
19
+ original works of authorship and other material subject to copyright
20
+ and certain other rights specified in the public license below. The
21
+ following considerations are for informational purposes only, are not
22
+ exhaustive, and do not form part of our licenses.
23
+
24
+ Considerations for licensors: Our public licenses are
25
+ intended for use by those authorized to give the public
26
+ permission to use material in ways otherwise restricted by
27
+ copyright and certain other rights. Our licenses are
28
+ irrevocable. Licensors should read and understand the terms
29
+ and conditions of the license they choose before applying it.
30
+ Licensors should also secure all rights necessary before
31
+ applying our licenses so that the public can reuse the
32
+ material as expected. Licensors should clearly mark any
33
+ material not subject to the license. This includes other CC-
34
+ licensed material, or material used under an exception or
35
+ limitation to copyright. More considerations for licensors:
36
+ wiki.creativecommons.org/Considerations_for_licensors
37
+
38
+ Considerations for the public: By using one of our public
39
+ licenses, a licensor grants the public permission to use the
40
+ licensed material under specified terms and conditions. If
41
+ the licensor's permission is not necessary for any reason--for
42
+ example, because of any applicable exception or limitation to
43
+ copyright--then that use is not regulated by the license. Our
44
+ licenses grant only permissions under copyright and certain
45
+ other rights that a licensor has authority to grant. Use of
46
+ the licensed material may still be restricted for other
47
+ reasons, including because others have copyright or other
48
+ rights in the material. A licensor may make special requests,
49
+ such as asking that all changes be marked or described.
50
+ Although not required by our licenses, you are encouraged to
51
+ respect those requests where reasonable. More_considerations
52
+ for the public:
53
+ wiki.creativecommons.org/Considerations_for_licensees
54
+
55
+ =======================================================================
56
+
57
+ Creative Commons Attribution 4.0 International Public License
58
+
59
+ By exercising the Licensed Rights (defined below), You accept and agree
60
+ to be bound by the terms and conditions of this Creative Commons
61
+ Attribution 4.0 International Public License ("Public License"). To the
62
+ extent this Public License may be interpreted as a contract, You are
63
+ granted the Licensed Rights in consideration of Your acceptance of
64
+ these terms and conditions, and the Licensor grants You such rights in
65
+ consideration of benefits the Licensor receives from making the
66
+ Licensed Material available under these terms and conditions.
67
+
68
+
69
+ Section 1 -- Definitions.
70
+
71
+ a. Adapted Material means material subject to Copyright and Similar
72
+ Rights that is derived from or based upon the Licensed Material
73
+ and in which the Licensed Material is translated, altered,
74
+ arranged, transformed, or otherwise modified in a manner requiring
75
+ permission under the Copyright and Similar Rights held by the
76
+ Licensor. For purposes of this Public License, where the Licensed
77
+ Material is a musical work, performance, or sound recording,
78
+ Adapted Material is always produced where the Licensed Material is
79
+ synched in timed relation with a moving image.
80
+
81
+ b. Adapter's License means the license You apply to Your Copyright
82
+ and Similar Rights in Your contributions to Adapted Material in
83
+ accordance with the terms and conditions of this Public License.
84
+
85
+ c. Copyright and Similar Rights means copyright and/or similar rights
86
+ closely related to copyright including, without limitation,
87
+ performance, broadcast, sound recording, and Sui Generis Database
88
+ Rights, without regard to how the rights are labeled or
89
+ categorized. For purposes of this Public License, the rights
90
+ specified in Section 2(b)(1)-(2) are not Copyright and Similar
91
+ Rights.
92
+
93
+ d. Effective Technological Measures means those measures that, in the
94
+ absence of proper authority, may not be circumvented under laws
95
+ fulfilling obligations under Article 11 of the WIPO Copyright
96
+ Treaty adopted on December 20, 1996, and/or similar international
97
+ agreements.
98
+
99
+ e. Exceptions and Limitations means fair use, fair dealing, and/or
100
+ any other exception or limitation to Copyright and Similar Rights
101
+ that applies to Your use of the Licensed Material.
102
+
103
+ f. Licensed Material means the artistic or literary work, database,
104
+ or other material to which the Licensor applied this Public
105
+ License.
106
+
107
+ g. Licensed Rights means the rights granted to You subject to the
108
+ terms and conditions of this Public License, which are limited to
109
+ all Copyright and Similar Rights that apply to Your use of the
110
+ Licensed Material and that the Licensor has authority to license.
111
+
112
+ h. Licensor means the individual(s) or entity(ies) granting rights
113
+ under this Public License.
114
+
115
+ i. Share means to provide material to the public by any means or
116
+ process that requires permission under the Licensed Rights, such
117
+ as reproduction, public display, public performance, distribution,
118
+ dissemination, communication, or importation, and to make material
119
+ available to the public including in ways that members of the
120
+ public may access the material from a place and at a time
121
+ individually chosen by them.
122
+
123
+ j. Sui Generis Database Rights means rights other than copyright
124
+ resulting from Directive 96/9/EC of the European Parliament and of
125
+ the Council of 11 March 1996 on the legal protection of databases,
126
+ as amended and/or succeeded, as well as other essentially
127
+ equivalent rights anywhere in the world.
128
+
129
+ k. You means the individual or entity exercising the Licensed Rights
130
+ under this Public License. Your has a corresponding meaning.
131
+
132
+
133
+ Section 2 -- Scope.
134
+
135
+ a. License grant.
136
+
137
+ 1. Subject to the terms and conditions of this Public License,
138
+ the Licensor hereby grants You a worldwide, royalty-free,
139
+ non-sublicensable, non-exclusive, irrevocable license to
140
+ exercise the Licensed Rights in the Licensed Material to:
141
+
142
+ a. reproduce and Share the Licensed Material, in whole or
143
+ in part; and
144
+
145
+ b. produce, reproduce, and Share Adapted Material.
146
+
147
+ 2. Exceptions and Limitations. For the avoidance of doubt, where
148
+ Exceptions and Limitations apply to Your use, this Public
149
+ License does not apply, and You do not need to comply with
150
+ its terms and conditions.
151
+
152
+ 3. Term. The term of this Public License is specified in Section
153
+ 6(a).
154
+
155
+ 4. Media and formats; technical modifications allowed. The
156
+ Licensor authorizes You to exercise the Licensed Rights in
157
+ all media and formats whether now known or hereafter created,
158
+ and to make technical modifications necessary to do so. The
159
+ Licensor waives and/or agrees not to assert any right or
160
+ authority to forbid You from making technical modifications
161
+ necessary to exercise the Licensed Rights, including
162
+ technical modifications necessary to circumvent Effective
163
+ Technological Measures. For purposes of this Public License,
164
+ simply making modifications authorized by this Section 2(a)
165
+ (4) never produces Adapted Material.
166
+
167
+ 5. Downstream recipients.
168
+
169
+ a. Offer from the Licensor -- Licensed Material. Every
170
+ recipient of the Licensed Material automatically
171
+ receives an offer from the Licensor to exercise the
172
+ Licensed Rights under the terms and conditions of this
173
+ Public License.
174
+
175
+ b. No downstream restrictions. You may not offer or impose
176
+ any additional or different terms or conditions on, or
177
+ apply any Effective Technological Measures to, the
178
+ Licensed Material if doing so restricts exercise of the
179
+ Licensed Rights by any recipient of the Licensed
180
+ Material.
181
+
182
+ 6. No endorsement. Nothing in this Public License constitutes or
183
+ may be construed as permission to assert or imply that You
184
+ are, or that Your use of the Licensed Material is, connected
185
+ with, or sponsored, endorsed, or granted official status by,
186
+ the Licensor or others designated to receive attribution as
187
+ provided in Section 3(a)(1)(A)(i).
188
+
189
+ b. Other rights.
190
+
191
+ 1. Moral rights, such as the right of integrity, are not
192
+ licensed under this Public License, nor are publicity,
193
+ privacy, and/or other similar personality rights; however, to
194
+ the extent possible, the Licensor waives and/or agrees not to
195
+ assert any such rights held by the Licensor to the limited
196
+ extent necessary to allow You to exercise the Licensed
197
+ Rights, but not otherwise.
198
+
199
+ 2. Patent and trademark rights are not licensed under this
200
+ Public License.
201
+
202
+ 3. To the extent possible, the Licensor waives any right to
203
+ collect royalties from You for the exercise of the Licensed
204
+ Rights, whether directly or through a collecting society
205
+ under any voluntary or waivable statutory or compulsory
206
+ licensing scheme. In all other cases the Licensor expressly
207
+ reserves any right to collect such royalties.
208
+
209
+
210
+ Section 3 -- License Conditions.
211
+
212
+ Your exercise of the Licensed Rights is expressly made subject to the
213
+ following conditions.
214
+
215
+ a. Attribution.
216
+
217
+ 1. If You Share the Licensed Material (including in modified
218
+ form), You must:
219
+
220
+ a. retain the following if it is supplied by the Licensor
221
+ with the Licensed Material:
222
+
223
+ i. identification of the creator(s) of the Licensed
224
+ Material and any others designated to receive
225
+ attribution, in any reasonable manner requested by
226
+ the Licensor (including by pseudonym if
227
+ designated);
228
+
229
+ ii. a copyright notice;
230
+
231
+ iii. a notice that refers to this Public License;
232
+
233
+ iv. a notice that refers to the disclaimer of
234
+ warranties;
235
+
236
+ v. a URI or hyperlink to the Licensed Material to the
237
+ extent reasonably practicable;
238
+
239
+ b. indicate if You modified the Licensed Material and
240
+ retain an indication of any previous modifications; and
241
+
242
+ c. indicate the Licensed Material is licensed under this
243
+ Public License, and include the text of, or the URI or
244
+ hyperlink to, this Public License.
245
+
246
+ 2. You may satisfy the conditions in Section 3(a)(1) in any
247
+ reasonable manner based on the medium, means, and context in
248
+ which You Share the Licensed Material. For example, it may be
249
+ reasonable to satisfy the conditions by providing a URI or
250
+ hyperlink to a resource that includes the required
251
+ information.
252
+
253
+ 3. If requested by the Licensor, You must remove any of the
254
+ information required by Section 3(a)(1)(A) to the extent
255
+ reasonably practicable.
256
+
257
+ 4. If You Share Adapted Material You produce, the Adapter's
258
+ License You apply must not prevent recipients of the Adapted
259
+ Material from complying with this Public License.
260
+
261
+
262
+ Section 4 -- Sui Generis Database Rights.
263
+
264
+ Where the Licensed Rights include Sui Generis Database Rights that
265
+ apply to Your use of the Licensed Material:
266
+
267
+ a. for the avoidance of doubt, Section 2(a)(1) grants You the right
268
+ to extract, reuse, reproduce, and Share all or a substantial
269
+ portion of the contents of the database;
270
+
271
+ b. if You include all or a substantial portion of the database
272
+ contents in a database in which You have Sui Generis Database
273
+ Rights, then the database in which You have Sui Generis Database
274
+ Rights (but not its individual contents) is Adapted Material; and
275
+
276
+ c. You must comply with the conditions in Section 3(a) if You Share
277
+ all or a substantial portion of the contents of the database.
278
+
279
+ For the avoidance of doubt, this Section 4 supplements and does not
280
+ replace Your obligations under this Public License where the Licensed
281
+ Rights include other Copyright and Similar Rights.
282
+
283
+
284
+ Section 5 -- Disclaimer of Warranties and Limitation of Liability.
285
+
286
+ a. UNLESS OTHERWISE SEPARATELY UNDERTAKEN BY THE LICENSOR, TO THE
287
+ EXTENT POSSIBLE, THE LICENSOR OFFERS THE LICENSED MATERIAL AS-IS
288
+ AND AS-AVAILABLE, AND MAKES NO REPRESENTATIONS OR WARRANTIES OF
289
+ ANY KIND CONCERNING THE LICENSED MATERIAL, WHETHER EXPRESS,
290
+ IMPLIED, STATUTORY, OR OTHER. THIS INCLUDES, WITHOUT LIMITATION,
291
+ WARRANTIES OF TITLE, MERCHANTABILITY, FITNESS FOR A PARTICULAR
292
+ PURPOSE, NON-INFRINGEMENT, ABSENCE OF LATENT OR OTHER DEFECTS,
293
+ ACCURACY, OR THE PRESENCE OR ABSENCE OF ERRORS, WHETHER OR NOT
294
+ KNOWN OR DISCOVERABLE. WHERE DISCLAIMERS OF WARRANTIES ARE NOT
295
+ ALLOWED IN FULL OR IN PART, THIS DISCLAIMER MAY NOT APPLY TO YOU.
296
+
297
+ b. TO THE EXTENT POSSIBLE, IN NO EVENT WILL THE LICENSOR BE LIABLE
298
+ TO YOU ON ANY LEGAL THEORY (INCLUDING, WITHOUT LIMITATION,
299
+ NEGLIGENCE) OR OTHERWISE FOR ANY DIRECT, SPECIAL, INDIRECT,
300
+ INCIDENTAL, CONSEQUENTIAL, PUNITIVE, EXEMPLARY, OR OTHER LOSSES,
301
+ COSTS, EXPENSES, OR DAMAGES ARISING OUT OF THIS PUBLIC LICENSE OR
302
+ USE OF THE LICENSED MATERIAL, EVEN IF THE LICENSOR HAS BEEN
303
+ ADVISED OF THE POSSIBILITY OF SUCH LOSSES, COSTS, EXPENSES, OR
304
+ DAMAGES. WHERE A LIMITATION OF LIABILITY IS NOT ALLOWED IN FULL OR
305
+ IN PART, THIS LIMITATION MAY NOT APPLY TO YOU.
306
+
307
+ c. The disclaimer of warranties and limitation of liability provided
308
+ above shall be interpreted in a manner that, to the extent
309
+ possible, most closely approximates an absolute disclaimer and
310
+ waiver of all liability.
311
+
312
+
313
+ Section 6 -- Term and Termination.
314
+
315
+ a. This Public License applies for the term of the Copyright and
316
+ Similar Rights licensed here. However, if You fail to comply with
317
+ this Public License, then Your rights under this Public License
318
+ terminate automatically.
319
+
320
+ b. Where Your right to use the Licensed Material has terminated under
321
+ Section 6(a), it reinstates:
322
+
323
+ 1. automatically as of the date the violation is cured, provided
324
+ it is cured within 30 days of Your discovery of the
325
+ violation; or
326
+
327
+ 2. upon express reinstatement by the Licensor.
328
+
329
+ For the avoidance of doubt, this Section 6(b) does not affect any
330
+ right the Licensor may have to seek remedies for Your violations
331
+ of this Public License.
332
+
333
+ c. For the avoidance of doubt, the Licensor may also offer the
334
+ Licensed Material under separate terms or conditions or stop
335
+ distributing the Licensed Material at any time; however, doing so
336
+ will not terminate this Public License.
337
+
338
+ d. Sections 1, 5, 6, 7, and 8 survive termination of this Public
339
+ License.
340
+
341
+
342
+ Section 7 -- Other Terms and Conditions.
343
+
344
+ a. The Licensor shall not be bound by any additional or different
345
+ terms or conditions communicated by You unless expressly agreed.
346
+
347
+ b. Any arrangements, understandings, or agreements regarding the
348
+ Licensed Material not stated herein are separate from and
349
+ independent of the terms and conditions of this Public License.
350
+
351
+
352
+ Section 8 -- Interpretation.
353
+
354
+ a. For the avoidance of doubt, this Public License does not, and
355
+ shall not be interpreted to, reduce, limit, restrict, or impose
356
+ conditions on any use of the Licensed Material that could lawfully
357
+ be made without permission under this Public License.
358
+
359
+ b. To the extent possible, if any provision of this Public License is
360
+ deemed unenforceable, it shall be automatically reformed to the
361
+ minimum extent necessary to make it enforceable. If the provision
362
+ cannot be reformed, it shall be severed from this Public License
363
+ without affecting the enforceability of the remaining terms and
364
+ conditions.
365
+
366
+ c. No term or condition of this Public License will be waived and no
367
+ failure to comply consented to unless expressly agreed to by the
368
+ Licensor.
369
+
370
+ d. Nothing in this Public License constitutes or may be interpreted
371
+ as a limitation upon, or waiver of, any privileges and immunities
372
+ that apply to the Licensor or You, including from the legal
373
+ processes of any jurisdiction or authority.
374
+
375
+
376
+ =======================================================================
377
+
378
+ Creative Commons is not a party to its public
379
+ licenses. Notwithstanding, Creative Commons may elect to apply one of
380
+ its public licenses to material it publishes and in those instances
381
+ will be considered the “Licensor.” The text of the Creative Commons
382
+ public licenses is dedicated to the public domain under the CC0 Public
383
+ Domain Dedication. Except for the limited purpose of indicating that
384
+ material is shared under a Creative Commons public license or as
385
+ otherwise permitted by the Creative Commons policies published at
386
+ creativecommons.org/policies, Creative Commons does not authorize the
387
+ use of the trademark "Creative Commons" or any other trademark or logo
388
+ of Creative Commons without its prior written consent including,
389
+ without limitation, in connection with any unauthorized modifications
390
+ to any of its public licenses or any other arrangements,
391
+ understandings, or agreements concerning use of licensed material. For
392
+ the avoidance of doubt, this paragraph does not form part of the
393
+ public licenses.
394
+
395
+ Creative Commons may be contacted at creativecommons.org.
acoustic/czech/mfa/v2.0.0/README.md ADDED
@@ -0,0 +1,138 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+
2
+ # Czech MFA acoustic model v2.0.0
3
+
4
+ [Link to documentation on mfa-models](https://mfa-models.readthedocs.io/en/main/acoustic/czech_mfa.html)
5
+
6
+ Jump to section:
7
+
8
+ - [Model details](#model-details)
9
+ - [Installation](#installation)
10
+ - [Intended use](#intended-use)
11
+ - [Performance Factors](#performance-factors)
12
+ - [Metrics](#metrics)
13
+ - [Ethical considerations](#ethical-considerations)
14
+ - [Troubleshooting issues](#troubleshooting-issues)
15
+ - [Training data](#training-data)
16
+ - [Evaluation data](#evaluation-data)
17
+
18
+ ## Model details
19
+
20
+ - **Maintainer:** [Montreal Forced Aligner](https://montreal-forced-aligner.readthedocs.io/)
21
+ - **Language:** [Czech](https://en.wikipedia.org/wiki/Czech_language)
22
+ - **Dialect:** N/A
23
+ - **Phone set:** [MFA](https://mfa-models.readthedocs.io/en/refactor/mfa_phone_set.html#czech)
24
+ - **Model type:** `Acoustic`
25
+ - **Features:** `MFCC + pitch`
26
+ - **Architecture:** `gmm-hmm`
27
+ - **Model version:** `v2.0.0`
28
+ - **Trained date:** `2022-03-31`
29
+ - **Compatible MFA version:** `v2.0.0`
30
+ - **License:** [CC BY 4.0](https://github.com/MontrealCorpusTools/mfa-models/tree/main/acoustic/czech/mfa/v2.0.0/LICENSE)
31
+ - **Citation:**
32
+
33
+ ```bibtex
34
+ @techreport{mfa_czech_mfa_acoustic_2022,
35
+ author={McAuliffe, Michael and Sonderegger, Morgan},
36
+ title={Czech MFA acoustic model v2.0.0},
37
+ address={\url{https://mfa-models.readthedocs.io/acoustic/Czech/Czech MFA acoustic model v2_0_0.html}},
38
+ year={2022},
39
+ month={Mar},
40
+ }
41
+ ```
42
+
43
+ - If you have comments or questions about this model, you can check [previous MFA model discussion posts](https://github.com/MontrealCorpusTools/mfa-models/discussions?discussions_q=Czech+MFA+acoustic+model+v2.0.0) or create [a new one](https://github.com/MontrealCorpusTools/mfa-models/discussions/new).
44
+
45
+ ## Installation
46
+
47
+ Install from the [MFA command line](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/models/index.html):
48
+
49
+ ```
50
+ mfa model download acoustic czech_mfa
51
+ ```
52
+
53
+ Or download from [the release page](https://github.com/MontrealCorpusTools/mfa-models/releases/tag/acoustic-czech_mfa-v2.0.0).
54
+
55
+ ## Intended use
56
+
57
+ This model is intended for forced alignment of [Czech](https://en.wikipedia.org/wiki/Czech_language) transcripts.
58
+
59
+ This model uses the [MFA](https://mfa-models.readthedocs.io/en/refactor/mfa_phone_set.html#czech) phone set for Czech, and was trained with the pronunciation dictionaries above. Pronunciations can be added on top of the dictionary, as long as no additional phones are introduced.
60
+
61
+ ## Performance Factors
62
+
63
+ As forced alignment is a relatively well-constrained problem (given accurate transcripts), this model should be applicable to a range of recording conditions and speakers. However, please note that it was trained on read speech in low-noise environments, so as your data diverges from that, you may run into alignment issues or need to [increase the beam size of MFA](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/configuration/#configuring-specific-commands) or see other recommendations in the [troubleshooting section below](#troubleshooting-issues).
64
+
65
+ Please note as well that MFA does not use state-of-the-art ASR models for forced alignment. You may get better performance (especially on speech-to-text tasks) using other frameworks like [Coqui](https://coqui.ai/).
66
+
67
+ ## Metrics
68
+
69
+ Acoustic models are typically generated as one component of a larger ASR system where the metric is word error rate (WER). For forced alignment, there is typically not the same sort of gold standard measure for most languages.
70
+
71
+ As a rough approximation of the acoustic model quality, we evaluated it against the corpus it was trained on alongside a language model trained from the same data. Key caveat here is that this is not a typical WER measure on held out data, so it should not be taken as a hard measure of how well an acoustic model will generalize to your data, but rather is more of a sanity check that the training data quality was sufficiently high.
72
+
73
+ Using the pronunciation dictionary above and a language model trained on the GlobalPhone Czech corpus (see the [GlobalPhone benchmark analysis page](https://montreal-forced-aligner.readthedocs.io/en/latest/benchmarks/globalphone_transcriptions.html) for more details):
74
+
75
+ - **WER:** `3.9%`
76
+ - **CER:** `1.5%`
77
+
78
+ ## Ethical considerations
79
+
80
+ Deploying any Speech-to-Text model into any production setting has ethical implications. You should consider these implications before use.
81
+
82
+ ### Demographic Bias
83
+
84
+ You should assume every machine learning model has demographic bias unless proven otherwise. For STT models, it is often the case that transcription accuracy is better for men than it is for women. If you are using this model in production, you should acknowledge this as a potential issue.
85
+
86
+ ### Surveillance
87
+
88
+ Speech-to-Text technologies may be misused to invade the privacy of others by recording and mining information from private conversations. This kind of individual privacy is protected by law in many countries. You should not assume consent to record and analyze private speech.
89
+
90
+
91
+ ## Troubleshooting issues
92
+
93
+ Machine learning models (like this acoustic model) perform best on data that is similar to the data on which they were trained.
94
+
95
+ The primary sources of variability in forced alignment will be the applicability of the pronunciation dictionary and how similar the speech, demographics, and recording conditions are. If you encounter issues in alignment, there are couple of avenues to improve performance:
96
+
97
+ 1. [Increase the beam size of MFA](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/configuration/#configuring-specific-commands)
98
+
99
+ * MFA defaults to a narrow beam to ensure quick alignment and also as a way to detect potential issues in your dataset, but depending on your data, you might benefit from boosting the beam to 100 or higher.
100
+
101
+ 2. Add pronunciations to the pronunciation dictionary
102
+
103
+ * This model was trained a particular dialect/style, and so adding pronunciations more representative of the variety spoken in your dataset will help alignment.
104
+
105
+ 3. Check the quality of your data
106
+
107
+ * MFA includes a [validator utility](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/data_validation.html), which aims to detect issues in the dataset.
108
+ * Use MFA's [anchor utility](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/workflows/anchor.html) to visually inspect your data as MFA sees it and correct issues in transcription or OOV items.
109
+
110
+ 4. Adapt the model to your data
111
+
112
+ * MFA has an [adaptation command](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/workflows/adapt_acoustic_model.html) to adapt some of the model to your data based on an initial alignment, and then run another alignment with the adapted model.
113
+
114
+ ## Training data
115
+
116
+ This model was trained on the following corpora:
117
+
118
+
119
+
120
+ * [Common Voice Czech](../../../../corpus/czech/common_voice_czech/8.0/README.md):
121
+ * **Hours:** `66.33`
122
+ * **Speakers:** `523`
123
+ * **Utterances:** `55,391`
124
+
125
+ * [GlobalPhone Czech](../../../../corpus/czech/globalphone_czech/3.1/README.md):
126
+ * **Hours:** `31.67`
127
+ * **Speakers:** `101`
128
+ * **Utterances:** `12,322`
129
+
130
+ * [Large Corpus of Czech Parliament Plenary Hearings](../../../../corpus/czech/large_corpus_of_czech_parliament_plenary_hearings/README.md):
131
+ * **Hours:** `450.95`
132
+ * **Speakers:** `84`
133
+ * **Utterances:** `194,181`
134
+
135
+ * [Czech Parliament Meetings](../../../../corpus/czech/czech_parliament_meetings/README.md):
136
+ * **Hours:** `88.81`
137
+ * **Speakers:** `67`
138
+ * **Utterances:** `51,494`
acoustic/czech/mfa/v2.0.0/meta.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "name": "czech_mfa",
3
+ "language": "Czech",
4
+ "dialect": "",
5
+ "phone_set": "MFA",
6
+ "version": "2.0.0",
7
+ "maintainer": "[Montreal Forced Aligner](https://montreal-forced-aligner.readthedocs.io/)",
8
+ "citation": "@techreport{mfa_czech_mfa_acoustic_2022,\n\tauthor={McAuliffe, Michael and Sonderegger, Morgan},\n\ttitle={Czech MFA acoustic model v2.0.0},\n\taddress={\\url{https://mfa-models.readthedocs.io/acoustic/Czech/Czech MFA acoustic model v2_0_0.html}},\n\tyear={2022},\n\tmonth={Mar},\n}",
9
+ "license": "CC BY 4.0",
10
+ "license_link": "[CC BY 4.0](https://github.com/MontrealCorpusTools/mfa-models/tree/main/acoustic/czech/mfa/v2.0.0/LICENSE)",
11
+ "architecture": "gmm-hmm",
12
+ "features": "MFCC + pitch",
13
+ "evaluation": {},
14
+ "decode": {},
15
+ "train_date": "2022-03-31"
16
+ }
acoustic/czech/mfa/v2.0.0a/LICENSE ADDED
@@ -0,0 +1,395 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ Attribution 4.0 International
2
+
3
+ =======================================================================
4
+
5
+ Creative Commons Corporation ("Creative Commons") is not a law firm and
6
+ does not provide legal services or legal advice. Distribution of
7
+ Creative Commons public licenses does not create a lawyer-client or
8
+ other relationship. Creative Commons makes its licenses and related
9
+ information available on an "as-is" basis. Creative Commons gives no
10
+ warranties regarding its licenses, any material licensed under their
11
+ terms and conditions, or any related information. Creative Commons
12
+ disclaims all liability for damages resulting from their use to the
13
+ fullest extent possible.
14
+
15
+ Using Creative Commons Public Licenses
16
+
17
+ Creative Commons public licenses provide a standard set of terms and
18
+ conditions that creators and other rights holders may use to share
19
+ original works of authorship and other material subject to copyright
20
+ and certain other rights specified in the public license below. The
21
+ following considerations are for informational purposes only, are not
22
+ exhaustive, and do not form part of our licenses.
23
+
24
+ Considerations for licensors: Our public licenses are
25
+ intended for use by those authorized to give the public
26
+ permission to use material in ways otherwise restricted by
27
+ copyright and certain other rights. Our licenses are
28
+ irrevocable. Licensors should read and understand the terms
29
+ and conditions of the license they choose before applying it.
30
+ Licensors should also secure all rights necessary before
31
+ applying our licenses so that the public can reuse the
32
+ material as expected. Licensors should clearly mark any
33
+ material not subject to the license. This includes other CC-
34
+ licensed material, or material used under an exception or
35
+ limitation to copyright. More considerations for licensors:
36
+ wiki.creativecommons.org/Considerations_for_licensors
37
+
38
+ Considerations for the public: By using one of our public
39
+ licenses, a licensor grants the public permission to use the
40
+ licensed material under specified terms and conditions. If
41
+ the licensor's permission is not necessary for any reason--for
42
+ example, because of any applicable exception or limitation to
43
+ copyright--then that use is not regulated by the license. Our
44
+ licenses grant only permissions under copyright and certain
45
+ other rights that a licensor has authority to grant. Use of
46
+ the licensed material may still be restricted for other
47
+ reasons, including because others have copyright or other
48
+ rights in the material. A licensor may make special requests,
49
+ such as asking that all changes be marked or described.
50
+ Although not required by our licenses, you are encouraged to
51
+ respect those requests where reasonable. More_considerations
52
+ for the public:
53
+ wiki.creativecommons.org/Considerations_for_licensees
54
+
55
+ =======================================================================
56
+
57
+ Creative Commons Attribution 4.0 International Public License
58
+
59
+ By exercising the Licensed Rights (defined below), You accept and agree
60
+ to be bound by the terms and conditions of this Creative Commons
61
+ Attribution 4.0 International Public License ("Public License"). To the
62
+ extent this Public License may be interpreted as a contract, You are
63
+ granted the Licensed Rights in consideration of Your acceptance of
64
+ these terms and conditions, and the Licensor grants You such rights in
65
+ consideration of benefits the Licensor receives from making the
66
+ Licensed Material available under these terms and conditions.
67
+
68
+
69
+ Section 1 -- Definitions.
70
+
71
+ a. Adapted Material means material subject to Copyright and Similar
72
+ Rights that is derived from or based upon the Licensed Material
73
+ and in which the Licensed Material is translated, altered,
74
+ arranged, transformed, or otherwise modified in a manner requiring
75
+ permission under the Copyright and Similar Rights held by the
76
+ Licensor. For purposes of this Public License, where the Licensed
77
+ Material is a musical work, performance, or sound recording,
78
+ Adapted Material is always produced where the Licensed Material is
79
+ synched in timed relation with a moving image.
80
+
81
+ b. Adapter's License means the license You apply to Your Copyright
82
+ and Similar Rights in Your contributions to Adapted Material in
83
+ accordance with the terms and conditions of this Public License.
84
+
85
+ c. Copyright and Similar Rights means copyright and/or similar rights
86
+ closely related to copyright including, without limitation,
87
+ performance, broadcast, sound recording, and Sui Generis Database
88
+ Rights, without regard to how the rights are labeled or
89
+ categorized. For purposes of this Public License, the rights
90
+ specified in Section 2(b)(1)-(2) are not Copyright and Similar
91
+ Rights.
92
+
93
+ d. Effective Technological Measures means those measures that, in the
94
+ absence of proper authority, may not be circumvented under laws
95
+ fulfilling obligations under Article 11 of the WIPO Copyright
96
+ Treaty adopted on December 20, 1996, and/or similar international
97
+ agreements.
98
+
99
+ e. Exceptions and Limitations means fair use, fair dealing, and/or
100
+ any other exception or limitation to Copyright and Similar Rights
101
+ that applies to Your use of the Licensed Material.
102
+
103
+ f. Licensed Material means the artistic or literary work, database,
104
+ or other material to which the Licensor applied this Public
105
+ License.
106
+
107
+ g. Licensed Rights means the rights granted to You subject to the
108
+ terms and conditions of this Public License, which are limited to
109
+ all Copyright and Similar Rights that apply to Your use of the
110
+ Licensed Material and that the Licensor has authority to license.
111
+
112
+ h. Licensor means the individual(s) or entity(ies) granting rights
113
+ under this Public License.
114
+
115
+ i. Share means to provide material to the public by any means or
116
+ process that requires permission under the Licensed Rights, such
117
+ as reproduction, public display, public performance, distribution,
118
+ dissemination, communication, or importation, and to make material
119
+ available to the public including in ways that members of the
120
+ public may access the material from a place and at a time
121
+ individually chosen by them.
122
+
123
+ j. Sui Generis Database Rights means rights other than copyright
124
+ resulting from Directive 96/9/EC of the European Parliament and of
125
+ the Council of 11 March 1996 on the legal protection of databases,
126
+ as amended and/or succeeded, as well as other essentially
127
+ equivalent rights anywhere in the world.
128
+
129
+ k. You means the individual or entity exercising the Licensed Rights
130
+ under this Public License. Your has a corresponding meaning.
131
+
132
+
133
+ Section 2 -- Scope.
134
+
135
+ a. License grant.
136
+
137
+ 1. Subject to the terms and conditions of this Public License,
138
+ the Licensor hereby grants You a worldwide, royalty-free,
139
+ non-sublicensable, non-exclusive, irrevocable license to
140
+ exercise the Licensed Rights in the Licensed Material to:
141
+
142
+ a. reproduce and Share the Licensed Material, in whole or
143
+ in part; and
144
+
145
+ b. produce, reproduce, and Share Adapted Material.
146
+
147
+ 2. Exceptions and Limitations. For the avoidance of doubt, where
148
+ Exceptions and Limitations apply to Your use, this Public
149
+ License does not apply, and You do not need to comply with
150
+ its terms and conditions.
151
+
152
+ 3. Term. The term of this Public License is specified in Section
153
+ 6(a).
154
+
155
+ 4. Media and formats; technical modifications allowed. The
156
+ Licensor authorizes You to exercise the Licensed Rights in
157
+ all media and formats whether now known or hereafter created,
158
+ and to make technical modifications necessary to do so. The
159
+ Licensor waives and/or agrees not to assert any right or
160
+ authority to forbid You from making technical modifications
161
+ necessary to exercise the Licensed Rights, including
162
+ technical modifications necessary to circumvent Effective
163
+ Technological Measures. For purposes of this Public License,
164
+ simply making modifications authorized by this Section 2(a)
165
+ (4) never produces Adapted Material.
166
+
167
+ 5. Downstream recipients.
168
+
169
+ a. Offer from the Licensor -- Licensed Material. Every
170
+ recipient of the Licensed Material automatically
171
+ receives an offer from the Licensor to exercise the
172
+ Licensed Rights under the terms and conditions of this
173
+ Public License.
174
+
175
+ b. No downstream restrictions. You may not offer or impose
176
+ any additional or different terms or conditions on, or
177
+ apply any Effective Technological Measures to, the
178
+ Licensed Material if doing so restricts exercise of the
179
+ Licensed Rights by any recipient of the Licensed
180
+ Material.
181
+
182
+ 6. No endorsement. Nothing in this Public License constitutes or
183
+ may be construed as permission to assert or imply that You
184
+ are, or that Your use of the Licensed Material is, connected
185
+ with, or sponsored, endorsed, or granted official status by,
186
+ the Licensor or others designated to receive attribution as
187
+ provided in Section 3(a)(1)(A)(i).
188
+
189
+ b. Other rights.
190
+
191
+ 1. Moral rights, such as the right of integrity, are not
192
+ licensed under this Public License, nor are publicity,
193
+ privacy, and/or other similar personality rights; however, to
194
+ the extent possible, the Licensor waives and/or agrees not to
195
+ assert any such rights held by the Licensor to the limited
196
+ extent necessary to allow You to exercise the Licensed
197
+ Rights, but not otherwise.
198
+
199
+ 2. Patent and trademark rights are not licensed under this
200
+ Public License.
201
+
202
+ 3. To the extent possible, the Licensor waives any right to
203
+ collect royalties from You for the exercise of the Licensed
204
+ Rights, whether directly or through a collecting society
205
+ under any voluntary or waivable statutory or compulsory
206
+ licensing scheme. In all other cases the Licensor expressly
207
+ reserves any right to collect such royalties.
208
+
209
+
210
+ Section 3 -- License Conditions.
211
+
212
+ Your exercise of the Licensed Rights is expressly made subject to the
213
+ following conditions.
214
+
215
+ a. Attribution.
216
+
217
+ 1. If You Share the Licensed Material (including in modified
218
+ form), You must:
219
+
220
+ a. retain the following if it is supplied by the Licensor
221
+ with the Licensed Material:
222
+
223
+ i. identification of the creator(s) of the Licensed
224
+ Material and any others designated to receive
225
+ attribution, in any reasonable manner requested by
226
+ the Licensor (including by pseudonym if
227
+ designated);
228
+
229
+ ii. a copyright notice;
230
+
231
+ iii. a notice that refers to this Public License;
232
+
233
+ iv. a notice that refers to the disclaimer of
234
+ warranties;
235
+
236
+ v. a URI or hyperlink to the Licensed Material to the
237
+ extent reasonably practicable;
238
+
239
+ b. indicate if You modified the Licensed Material and
240
+ retain an indication of any previous modifications; and
241
+
242
+ c. indicate the Licensed Material is licensed under this
243
+ Public License, and include the text of, or the URI or
244
+ hyperlink to, this Public License.
245
+
246
+ 2. You may satisfy the conditions in Section 3(a)(1) in any
247
+ reasonable manner based on the medium, means, and context in
248
+ which You Share the Licensed Material. For example, it may be
249
+ reasonable to satisfy the conditions by providing a URI or
250
+ hyperlink to a resource that includes the required
251
+ information.
252
+
253
+ 3. If requested by the Licensor, You must remove any of the
254
+ information required by Section 3(a)(1)(A) to the extent
255
+ reasonably practicable.
256
+
257
+ 4. If You Share Adapted Material You produce, the Adapter's
258
+ License You apply must not prevent recipients of the Adapted
259
+ Material from complying with this Public License.
260
+
261
+
262
+ Section 4 -- Sui Generis Database Rights.
263
+
264
+ Where the Licensed Rights include Sui Generis Database Rights that
265
+ apply to Your use of the Licensed Material:
266
+
267
+ a. for the avoidance of doubt, Section 2(a)(1) grants You the right
268
+ to extract, reuse, reproduce, and Share all or a substantial
269
+ portion of the contents of the database;
270
+
271
+ b. if You include all or a substantial portion of the database
272
+ contents in a database in which You have Sui Generis Database
273
+ Rights, then the database in which You have Sui Generis Database
274
+ Rights (but not its individual contents) is Adapted Material; and
275
+
276
+ c. You must comply with the conditions in Section 3(a) if You Share
277
+ all or a substantial portion of the contents of the database.
278
+
279
+ For the avoidance of doubt, this Section 4 supplements and does not
280
+ replace Your obligations under this Public License where the Licensed
281
+ Rights include other Copyright and Similar Rights.
282
+
283
+
284
+ Section 5 -- Disclaimer of Warranties and Limitation of Liability.
285
+
286
+ a. UNLESS OTHERWISE SEPARATELY UNDERTAKEN BY THE LICENSOR, TO THE
287
+ EXTENT POSSIBLE, THE LICENSOR OFFERS THE LICENSED MATERIAL AS-IS
288
+ AND AS-AVAILABLE, AND MAKES NO REPRESENTATIONS OR WARRANTIES OF
289
+ ANY KIND CONCERNING THE LICENSED MATERIAL, WHETHER EXPRESS,
290
+ IMPLIED, STATUTORY, OR OTHER. THIS INCLUDES, WITHOUT LIMITATION,
291
+ WARRANTIES OF TITLE, MERCHANTABILITY, FITNESS FOR A PARTICULAR
292
+ PURPOSE, NON-INFRINGEMENT, ABSENCE OF LATENT OR OTHER DEFECTS,
293
+ ACCURACY, OR THE PRESENCE OR ABSENCE OF ERRORS, WHETHER OR NOT
294
+ KNOWN OR DISCOVERABLE. WHERE DISCLAIMERS OF WARRANTIES ARE NOT
295
+ ALLOWED IN FULL OR IN PART, THIS DISCLAIMER MAY NOT APPLY TO YOU.
296
+
297
+ b. TO THE EXTENT POSSIBLE, IN NO EVENT WILL THE LICENSOR BE LIABLE
298
+ TO YOU ON ANY LEGAL THEORY (INCLUDING, WITHOUT LIMITATION,
299
+ NEGLIGENCE) OR OTHERWISE FOR ANY DIRECT, SPECIAL, INDIRECT,
300
+ INCIDENTAL, CONSEQUENTIAL, PUNITIVE, EXEMPLARY, OR OTHER LOSSES,
301
+ COSTS, EXPENSES, OR DAMAGES ARISING OUT OF THIS PUBLIC LICENSE OR
302
+ USE OF THE LICENSED MATERIAL, EVEN IF THE LICENSOR HAS BEEN
303
+ ADVISED OF THE POSSIBILITY OF SUCH LOSSES, COSTS, EXPENSES, OR
304
+ DAMAGES. WHERE A LIMITATION OF LIABILITY IS NOT ALLOWED IN FULL OR
305
+ IN PART, THIS LIMITATION MAY NOT APPLY TO YOU.
306
+
307
+ c. The disclaimer of warranties and limitation of liability provided
308
+ above shall be interpreted in a manner that, to the extent
309
+ possible, most closely approximates an absolute disclaimer and
310
+ waiver of all liability.
311
+
312
+
313
+ Section 6 -- Term and Termination.
314
+
315
+ a. This Public License applies for the term of the Copyright and
316
+ Similar Rights licensed here. However, if You fail to comply with
317
+ this Public License, then Your rights under this Public License
318
+ terminate automatically.
319
+
320
+ b. Where Your right to use the Licensed Material has terminated under
321
+ Section 6(a), it reinstates:
322
+
323
+ 1. automatically as of the date the violation is cured, provided
324
+ it is cured within 30 days of Your discovery of the
325
+ violation; or
326
+
327
+ 2. upon express reinstatement by the Licensor.
328
+
329
+ For the avoidance of doubt, this Section 6(b) does not affect any
330
+ right the Licensor may have to seek remedies for Your violations
331
+ of this Public License.
332
+
333
+ c. For the avoidance of doubt, the Licensor may also offer the
334
+ Licensed Material under separate terms or conditions or stop
335
+ distributing the Licensed Material at any time; however, doing so
336
+ will not terminate this Public License.
337
+
338
+ d. Sections 1, 5, 6, 7, and 8 survive termination of this Public
339
+ License.
340
+
341
+
342
+ Section 7 -- Other Terms and Conditions.
343
+
344
+ a. The Licensor shall not be bound by any additional or different
345
+ terms or conditions communicated by You unless expressly agreed.
346
+
347
+ b. Any arrangements, understandings, or agreements regarding the
348
+ Licensed Material not stated herein are separate from and
349
+ independent of the terms and conditions of this Public License.
350
+
351
+
352
+ Section 8 -- Interpretation.
353
+
354
+ a. For the avoidance of doubt, this Public License does not, and
355
+ shall not be interpreted to, reduce, limit, restrict, or impose
356
+ conditions on any use of the Licensed Material that could lawfully
357
+ be made without permission under this Public License.
358
+
359
+ b. To the extent possible, if any provision of this Public License is
360
+ deemed unenforceable, it shall be automatically reformed to the
361
+ minimum extent necessary to make it enforceable. If the provision
362
+ cannot be reformed, it shall be severed from this Public License
363
+ without affecting the enforceability of the remaining terms and
364
+ conditions.
365
+
366
+ c. No term or condition of this Public License will be waived and no
367
+ failure to comply consented to unless expressly agreed to by the
368
+ Licensor.
369
+
370
+ d. Nothing in this Public License constitutes or may be interpreted
371
+ as a limitation upon, or waiver of, any privileges and immunities
372
+ that apply to the Licensor or You, including from the legal
373
+ processes of any jurisdiction or authority.
374
+
375
+
376
+ =======================================================================
377
+
378
+ Creative Commons is not a party to its public
379
+ licenses. Notwithstanding, Creative Commons may elect to apply one of
380
+ its public licenses to material it publishes and in those instances
381
+ will be considered the “Licensor.” The text of the Creative Commons
382
+ public licenses is dedicated to the public domain under the CC0 Public
383
+ Domain Dedication. Except for the limited purpose of indicating that
384
+ material is shared under a Creative Commons public license or as
385
+ otherwise permitted by the Creative Commons policies published at
386
+ creativecommons.org/policies, Creative Commons does not authorize the
387
+ use of the trademark "Creative Commons" or any other trademark or logo
388
+ of Creative Commons without its prior written consent including,
389
+ without limitation, in connection with any unauthorized modifications
390
+ to any of its public licenses or any other arrangements,
391
+ understandings, or agreements concerning use of licensed material. For
392
+ the avoidance of doubt, this paragraph does not form part of the
393
+ public licenses.
394
+
395
+ Creative Commons may be contacted at creativecommons.org.
acoustic/czech/mfa/v2.0.0a/README.md ADDED
@@ -0,0 +1,138 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+
2
+ # Czech MFA acoustic model v2.0.0a
3
+
4
+ [Link to documentation on mfa-models](https://mfa-models.readthedocs.io/en/main/acoustic/czech_mfa.html)
5
+
6
+ Jump to section:
7
+
8
+ - [Model details](#model-details)
9
+ - [Installation](#installation)
10
+ - [Intended use](#intended-use)
11
+ - [Performance Factors](#performance-factors)
12
+ - [Metrics](#metrics)
13
+ - [Ethical considerations](#ethical-considerations)
14
+ - [Troubleshooting issues](#troubleshooting-issues)
15
+ - [Training data](#training-data)
16
+ - [Evaluation data](#evaluation-data)
17
+
18
+ ## Model details
19
+
20
+ - **Maintainer:** [Montreal Forced Aligner](https://montreal-forced-aligner.readthedocs.io/)
21
+ - **Language:** [Czech](https://en.wikipedia.org/wiki/Czech_language)
22
+ - **Dialect:** N/A
23
+ - **Phone set:** [MFA](https://mfa-models.readthedocs.io/en/refactor/mfa_phone_set.html#czech)
24
+ - **Model type:** `Acoustic`
25
+ - **Features:** `MFCC + pitch`
26
+ - **Architecture:** `gmm-hmm`
27
+ - **Model version:** `v2.0.0a`
28
+ - **Trained date:** `2022-05-19`
29
+ - **Compatible MFA version:** `v2.0.0`
30
+ - **License:** [CC BY 4.0](https://github.com/MontrealCorpusTools/mfa-models/tree/main/acoustic/czech/mfa/v2.0.0a/LICENSE)
31
+ - **Citation:**
32
+
33
+ ```bibtex
34
+ @techreport{mfa_czech_mfa_acoustic_2022,
35
+ author={McAuliffe, Michael and Sonderegger, Morgan},
36
+ title={Czech MFA acoustic model v2.0.0a},
37
+ address={\url{https://mfa-models.readthedocs.io/acoustic/Czech/Czech MFA acoustic model v2_0_0a.html}},
38
+ year={2022},
39
+ month={May},
40
+ }
41
+ ```
42
+
43
+ - If you have comments or questions about this model, you can check [previous MFA model discussion posts](https://github.com/MontrealCorpusTools/mfa-models/discussions?discussions_q=Czech+MFA+acoustic+model+v2.0.0a) or create [a new one](https://github.com/MontrealCorpusTools/mfa-models/discussions/new).
44
+
45
+ ## Installation
46
+
47
+ Install from the [MFA command line](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/models/index.html):
48
+
49
+ ```
50
+ mfa model download acoustic czech_mfa
51
+ ```
52
+
53
+ Or download from [the release page](https://github.com/MontrealCorpusTools/mfa-models/releases/tag/acoustic-czech_mfa-v2.0.0a).
54
+
55
+ ## Intended use
56
+
57
+ This model is intended for forced alignment of [Czech](https://en.wikipedia.org/wiki/Czech_language) transcripts.
58
+
59
+ This model uses the [MFA](https://mfa-models.readthedocs.io/en/refactor/mfa_phone_set.html#czech) phone set for Czech, and was trained with the pronunciation dictionaries above. Pronunciations can be added on top of the dictionary, as long as no additional phones are introduced.
60
+
61
+ ## Performance Factors
62
+
63
+ As forced alignment is a relatively well-constrained problem (given accurate transcripts), this model should be applicable to a range of recording conditions and speakers. However, please note that it was trained on read speech in low-noise environments, so as your data diverges from that, you may run into alignment issues or need to [increase the beam size of MFA](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/configuration/#configuring-specific-commands) or see other recommendations in the [troubleshooting section below](#troubleshooting-issues).
64
+
65
+ Please note as well that MFA does not use state-of-the-art ASR models for forced alignment. You may get better performance (especially on speech-to-text tasks) using other frameworks like [Coqui](https://coqui.ai/).
66
+
67
+ ## Metrics
68
+
69
+ Acoustic models are typically generated as one component of a larger ASR system where the metric is word error rate (WER). For forced alignment, there is typically not the same sort of gold standard measure for most languages.
70
+
71
+ As a rough approximation of the acoustic model quality, we evaluated it against the corpus it was trained on alongside a language model trained from the same data. Key caveat here is that this is not a typical WER measure on held out data, so it should not be taken as a hard measure of how well an acoustic model will generalize to your data, but rather is more of a sanity check that the training data quality was sufficiently high.
72
+
73
+ Using the pronunciation dictionary above and a language model trained on the GlobalPhone Czech corpus (see the [GlobalPhone benchmark analysis page](https://montreal-forced-aligner.readthedocs.io/en/latest/benchmarks/globalphone_transcriptions.html) for more details):
74
+
75
+ - **WER:** `3.1%`
76
+ - **CER:** `1.2%`
77
+
78
+ ## Ethical considerations
79
+
80
+ Deploying any Speech-to-Text model into any production setting has ethical implications. You should consider these implications before use.
81
+
82
+ ### Demographic Bias
83
+
84
+ You should assume every machine learning model has demographic bias unless proven otherwise. For STT models, it is often the case that transcription accuracy is better for men than it is for women. If you are using this model in production, you should acknowledge this as a potential issue.
85
+
86
+ ### Surveillance
87
+
88
+ Speech-to-Text technologies may be misused to invade the privacy of others by recording and mining information from private conversations. This kind of individual privacy is protected by law in many countries. You should not assume consent to record and analyze private speech.
89
+
90
+
91
+ ## Troubleshooting issues
92
+
93
+ Machine learning models (like this acoustic model) perform best on data that is similar to the data on which they were trained.
94
+
95
+ The primary sources of variability in forced alignment will be the applicability of the pronunciation dictionary and how similar the speech, demographics, and recording conditions are. If you encounter issues in alignment, there are couple of avenues to improve performance:
96
+
97
+ 1. [Increase the beam size of MFA](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/configuration/#configuring-specific-commands)
98
+
99
+ * MFA defaults to a narrow beam to ensure quick alignment and also as a way to detect potential issues in your dataset, but depending on your data, you might benefit from boosting the beam to 100 or higher.
100
+
101
+ 2. Add pronunciations to the pronunciation dictionary
102
+
103
+ * This model was trained a particular dialect/style, and so adding pronunciations more representative of the variety spoken in your dataset will help alignment.
104
+
105
+ 3. Check the quality of your data
106
+
107
+ * MFA includes a [validator utility](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/data_validation.html), which aims to detect issues in the dataset.
108
+ * Use MFA's [anchor utility](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/workflows/anchor.html) to visually inspect your data as MFA sees it and correct issues in transcription or OOV items.
109
+
110
+ 4. Adapt the model to your data
111
+
112
+ * MFA has an [adaptation command](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/workflows/adapt_acoustic_model.html) to adapt some of the model to your data based on an initial alignment, and then run another alignment with the adapted model.
113
+
114
+ ## Training data
115
+
116
+ This model was trained on the following corpora:
117
+
118
+
119
+
120
+ * [Common Voice Czech](../../../../corpus/czech/common_voice_czech/8.0/README.md):
121
+ * **Hours:** `66.33`
122
+ * **Speakers:** `523`
123
+ * **Utterances:** `55,391`
124
+
125
+ * [GlobalPhone Czech](../../../../corpus/czech/globalphone_czech/3.1/README.md):
126
+ * **Hours:** `31.67`
127
+ * **Speakers:** `101`
128
+ * **Utterances:** `12,322`
129
+
130
+ * [Large Corpus of Czech Parliament Plenary Hearings](../../../../corpus/czech/large_corpus_of_czech_parliament_plenary_hearings/README.md):
131
+ * **Hours:** `450.95`
132
+ * **Speakers:** `84`
133
+ * **Utterances:** `194,181`
134
+
135
+ * [Czech Parliament Meetings](../../../../corpus/czech/czech_parliament_meetings/README.md):
136
+ * **Hours:** `88.81`
137
+ * **Speakers:** `67`
138
+ * **Utterances:** `51,494`
acoustic/czech/mfa/v2.0.0a/meta.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "name": "czech_mfa",
3
+ "language": "Czech",
4
+ "dialect": "",
5
+ "phone_set": "MFA",
6
+ "version": "2.0.0a",
7
+ "maintainer": "[Montreal Forced Aligner](https://montreal-forced-aligner.readthedocs.io/)",
8
+ "citation": "@techreport{mfa_czech_mfa_acoustic_2022,\n\tauthor={McAuliffe, Michael and Sonderegger, Morgan},\n\ttitle={Czech MFA acoustic model v2.0.0a},\n\taddress={\\url{https://mfa-models.readthedocs.io/acoustic/Czech/Czech MFA acoustic model v2_0_0a.html}},\n\tyear={2022},\n\tmonth={May},\n}",
9
+ "license": "CC BY 4.0",
10
+ "license_link": "[CC BY 4.0](https://github.com/MontrealCorpusTools/mfa-models/tree/main/acoustic/czech/mfa/v2.0.0a/LICENSE)",
11
+ "architecture": "gmm-hmm",
12
+ "features": "MFCC + pitch",
13
+ "evaluation": {},
14
+ "decode": {},
15
+ "train_date": "2022-05-19"
16
+ }
acoustic/czech/mfa/v3.3.0/LICENSE ADDED
@@ -0,0 +1,395 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ Attribution 4.0 International
2
+
3
+ =======================================================================
4
+
5
+ Creative Commons Corporation ("Creative Commons") is not a law firm and
6
+ does not provide legal services or legal advice. Distribution of
7
+ Creative Commons public licenses does not create a lawyer-client or
8
+ other relationship. Creative Commons makes its licenses and related
9
+ information available on an "as-is" basis. Creative Commons gives no
10
+ warranties regarding its licenses, any material licensed under their
11
+ terms and conditions, or any related information. Creative Commons
12
+ disclaims all liability for damages resulting from their use to the
13
+ fullest extent possible.
14
+
15
+ Using Creative Commons Public Licenses
16
+
17
+ Creative Commons public licenses provide a standard set of terms and
18
+ conditions that creators and other rights holders may use to share
19
+ original works of authorship and other material subject to copyright
20
+ and certain other rights specified in the public license below. The
21
+ following considerations are for informational purposes only, are not
22
+ exhaustive, and do not form part of our licenses.
23
+
24
+ Considerations for licensors: Our public licenses are
25
+ intended for use by those authorized to give the public
26
+ permission to use material in ways otherwise restricted by
27
+ copyright and certain other rights. Our licenses are
28
+ irrevocable. Licensors should read and understand the terms
29
+ and conditions of the license they choose before applying it.
30
+ Licensors should also secure all rights necessary before
31
+ applying our licenses so that the public can reuse the
32
+ material as expected. Licensors should clearly mark any
33
+ material not subject to the license. This includes other CC-
34
+ licensed material, or material used under an exception or
35
+ limitation to copyright. More considerations for licensors:
36
+ wiki.creativecommons.org/Considerations_for_licensors
37
+
38
+ Considerations for the public: By using one of our public
39
+ licenses, a licensor grants the public permission to use the
40
+ licensed material under specified terms and conditions. If
41
+ the licensor's permission is not necessary for any reason--for
42
+ example, because of any applicable exception or limitation to
43
+ copyright--then that use is not regulated by the license. Our
44
+ licenses grant only permissions under copyright and certain
45
+ other rights that a licensor has authority to grant. Use of
46
+ the licensed material may still be restricted for other
47
+ reasons, including because others have copyright or other
48
+ rights in the material. A licensor may make special requests,
49
+ such as asking that all changes be marked or described.
50
+ Although not required by our licenses, you are encouraged to
51
+ respect those requests where reasonable. More_considerations
52
+ for the public:
53
+ wiki.creativecommons.org/Considerations_for_licensees
54
+
55
+ =======================================================================
56
+
57
+ Creative Commons Attribution 4.0 International Public License
58
+
59
+ By exercising the Licensed Rights (defined below), You accept and agree
60
+ to be bound by the terms and conditions of this Creative Commons
61
+ Attribution 4.0 International Public License ("Public License"). To the
62
+ extent this Public License may be interpreted as a contract, You are
63
+ granted the Licensed Rights in consideration of Your acceptance of
64
+ these terms and conditions, and the Licensor grants You such rights in
65
+ consideration of benefits the Licensor receives from making the
66
+ Licensed Material available under these terms and conditions.
67
+
68
+
69
+ Section 1 -- Definitions.
70
+
71
+ a. Adapted Material means material subject to Copyright and Similar
72
+ Rights that is derived from or based upon the Licensed Material
73
+ and in which the Licensed Material is translated, altered,
74
+ arranged, transformed, or otherwise modified in a manner requiring
75
+ permission under the Copyright and Similar Rights held by the
76
+ Licensor. For purposes of this Public License, where the Licensed
77
+ Material is a musical work, performance, or sound recording,
78
+ Adapted Material is always produced where the Licensed Material is
79
+ synched in timed relation with a moving image.
80
+
81
+ b. Adapter's License means the license You apply to Your Copyright
82
+ and Similar Rights in Your contributions to Adapted Material in
83
+ accordance with the terms and conditions of this Public License.
84
+
85
+ c. Copyright and Similar Rights means copyright and/or similar rights
86
+ closely related to copyright including, without limitation,
87
+ performance, broadcast, sound recording, and Sui Generis Database
88
+ Rights, without regard to how the rights are labeled or
89
+ categorized. For purposes of this Public License, the rights
90
+ specified in Section 2(b)(1)-(2) are not Copyright and Similar
91
+ Rights.
92
+
93
+ d. Effective Technological Measures means those measures that, in the
94
+ absence of proper authority, may not be circumvented under laws
95
+ fulfilling obligations under Article 11 of the WIPO Copyright
96
+ Treaty adopted on December 20, 1996, and/or similar international
97
+ agreements.
98
+
99
+ e. Exceptions and Limitations means fair use, fair dealing, and/or
100
+ any other exception or limitation to Copyright and Similar Rights
101
+ that applies to Your use of the Licensed Material.
102
+
103
+ f. Licensed Material means the artistic or literary work, database,
104
+ or other material to which the Licensor applied this Public
105
+ License.
106
+
107
+ g. Licensed Rights means the rights granted to You subject to the
108
+ terms and conditions of this Public License, which are limited to
109
+ all Copyright and Similar Rights that apply to Your use of the
110
+ Licensed Material and that the Licensor has authority to license.
111
+
112
+ h. Licensor means the individual(s) or entity(ies) granting rights
113
+ under this Public License.
114
+
115
+ i. Share means to provide material to the public by any means or
116
+ process that requires permission under the Licensed Rights, such
117
+ as reproduction, public display, public performance, distribution,
118
+ dissemination, communication, or importation, and to make material
119
+ available to the public including in ways that members of the
120
+ public may access the material from a place and at a time
121
+ individually chosen by them.
122
+
123
+ j. Sui Generis Database Rights means rights other than copyright
124
+ resulting from Directive 96/9/EC of the European Parliament and of
125
+ the Council of 11 March 1996 on the legal protection of databases,
126
+ as amended and/or succeeded, as well as other essentially
127
+ equivalent rights anywhere in the world.
128
+
129
+ k. You means the individual or entity exercising the Licensed Rights
130
+ under this Public License. Your has a corresponding meaning.
131
+
132
+
133
+ Section 2 -- Scope.
134
+
135
+ a. License grant.
136
+
137
+ 1. Subject to the terms and conditions of this Public License,
138
+ the Licensor hereby grants You a worldwide, royalty-free,
139
+ non-sublicensable, non-exclusive, irrevocable license to
140
+ exercise the Licensed Rights in the Licensed Material to:
141
+
142
+ a. reproduce and Share the Licensed Material, in whole or
143
+ in part; and
144
+
145
+ b. produce, reproduce, and Share Adapted Material.
146
+
147
+ 2. Exceptions and Limitations. For the avoidance of doubt, where
148
+ Exceptions and Limitations apply to Your use, this Public
149
+ License does not apply, and You do not need to comply with
150
+ its terms and conditions.
151
+
152
+ 3. Term. The term of this Public License is specified in Section
153
+ 6(a).
154
+
155
+ 4. Media and formats; technical modifications allowed. The
156
+ Licensor authorizes You to exercise the Licensed Rights in
157
+ all media and formats whether now known or hereafter created,
158
+ and to make technical modifications necessary to do so. The
159
+ Licensor waives and/or agrees not to assert any right or
160
+ authority to forbid You from making technical modifications
161
+ necessary to exercise the Licensed Rights, including
162
+ technical modifications necessary to circumvent Effective
163
+ Technological Measures. For purposes of this Public License,
164
+ simply making modifications authorized by this Section 2(a)
165
+ (4) never produces Adapted Material.
166
+
167
+ 5. Downstream recipients.
168
+
169
+ a. Offer from the Licensor -- Licensed Material. Every
170
+ recipient of the Licensed Material automatically
171
+ receives an offer from the Licensor to exercise the
172
+ Licensed Rights under the terms and conditions of this
173
+ Public License.
174
+
175
+ b. No downstream restrictions. You may not offer or impose
176
+ any additional or different terms or conditions on, or
177
+ apply any Effective Technological Measures to, the
178
+ Licensed Material if doing so restricts exercise of the
179
+ Licensed Rights by any recipient of the Licensed
180
+ Material.
181
+
182
+ 6. No endorsement. Nothing in this Public License constitutes or
183
+ may be construed as permission to assert or imply that You
184
+ are, or that Your use of the Licensed Material is, connected
185
+ with, or sponsored, endorsed, or granted official status by,
186
+ the Licensor or others designated to receive attribution as
187
+ provided in Section 3(a)(1)(A)(i).
188
+
189
+ b. Other rights.
190
+
191
+ 1. Moral rights, such as the right of integrity, are not
192
+ licensed under this Public License, nor are publicity,
193
+ privacy, and/or other similar personality rights; however, to
194
+ the extent possible, the Licensor waives and/or agrees not to
195
+ assert any such rights held by the Licensor to the limited
196
+ extent necessary to allow You to exercise the Licensed
197
+ Rights, but not otherwise.
198
+
199
+ 2. Patent and trademark rights are not licensed under this
200
+ Public License.
201
+
202
+ 3. To the extent possible, the Licensor waives any right to
203
+ collect royalties from You for the exercise of the Licensed
204
+ Rights, whether directly or through a collecting society
205
+ under any voluntary or waivable statutory or compulsory
206
+ licensing scheme. In all other cases the Licensor expressly
207
+ reserves any right to collect such royalties.
208
+
209
+
210
+ Section 3 -- License Conditions.
211
+
212
+ Your exercise of the Licensed Rights is expressly made subject to the
213
+ following conditions.
214
+
215
+ a. Attribution.
216
+
217
+ 1. If You Share the Licensed Material (including in modified
218
+ form), You must:
219
+
220
+ a. retain the following if it is supplied by the Licensor
221
+ with the Licensed Material:
222
+
223
+ i. identification of the creator(s) of the Licensed
224
+ Material and any others designated to receive
225
+ attribution, in any reasonable manner requested by
226
+ the Licensor (including by pseudonym if
227
+ designated);
228
+
229
+ ii. a copyright notice;
230
+
231
+ iii. a notice that refers to this Public License;
232
+
233
+ iv. a notice that refers to the disclaimer of
234
+ warranties;
235
+
236
+ v. a URI or hyperlink to the Licensed Material to the
237
+ extent reasonably practicable;
238
+
239
+ b. indicate if You modified the Licensed Material and
240
+ retain an indication of any previous modifications; and
241
+
242
+ c. indicate the Licensed Material is licensed under this
243
+ Public License, and include the text of, or the URI or
244
+ hyperlink to, this Public License.
245
+
246
+ 2. You may satisfy the conditions in Section 3(a)(1) in any
247
+ reasonable manner based on the medium, means, and context in
248
+ which You Share the Licensed Material. For example, it may be
249
+ reasonable to satisfy the conditions by providing a URI or
250
+ hyperlink to a resource that includes the required
251
+ information.
252
+
253
+ 3. If requested by the Licensor, You must remove any of the
254
+ information required by Section 3(a)(1)(A) to the extent
255
+ reasonably practicable.
256
+
257
+ 4. If You Share Adapted Material You produce, the Adapter's
258
+ License You apply must not prevent recipients of the Adapted
259
+ Material from complying with this Public License.
260
+
261
+
262
+ Section 4 -- Sui Generis Database Rights.
263
+
264
+ Where the Licensed Rights include Sui Generis Database Rights that
265
+ apply to Your use of the Licensed Material:
266
+
267
+ a. for the avoidance of doubt, Section 2(a)(1) grants You the right
268
+ to extract, reuse, reproduce, and Share all or a substantial
269
+ portion of the contents of the database;
270
+
271
+ b. if You include all or a substantial portion of the database
272
+ contents in a database in which You have Sui Generis Database
273
+ Rights, then the database in which You have Sui Generis Database
274
+ Rights (but not its individual contents) is Adapted Material; and
275
+
276
+ c. You must comply with the conditions in Section 3(a) if You Share
277
+ all or a substantial portion of the contents of the database.
278
+
279
+ For the avoidance of doubt, this Section 4 supplements and does not
280
+ replace Your obligations under this Public License where the Licensed
281
+ Rights include other Copyright and Similar Rights.
282
+
283
+
284
+ Section 5 -- Disclaimer of Warranties and Limitation of Liability.
285
+
286
+ a. UNLESS OTHERWISE SEPARATELY UNDERTAKEN BY THE LICENSOR, TO THE
287
+ EXTENT POSSIBLE, THE LICENSOR OFFERS THE LICENSED MATERIAL AS-IS
288
+ AND AS-AVAILABLE, AND MAKES NO REPRESENTATIONS OR WARRANTIES OF
289
+ ANY KIND CONCERNING THE LICENSED MATERIAL, WHETHER EXPRESS,
290
+ IMPLIED, STATUTORY, OR OTHER. THIS INCLUDES, WITHOUT LIMITATION,
291
+ WARRANTIES OF TITLE, MERCHANTABILITY, FITNESS FOR A PARTICULAR
292
+ PURPOSE, NON-INFRINGEMENT, ABSENCE OF LATENT OR OTHER DEFECTS,
293
+ ACCURACY, OR THE PRESENCE OR ABSENCE OF ERRORS, WHETHER OR NOT
294
+ KNOWN OR DISCOVERABLE. WHERE DISCLAIMERS OF WARRANTIES ARE NOT
295
+ ALLOWED IN FULL OR IN PART, THIS DISCLAIMER MAY NOT APPLY TO YOU.
296
+
297
+ b. TO THE EXTENT POSSIBLE, IN NO EVENT WILL THE LICENSOR BE LIABLE
298
+ TO YOU ON ANY LEGAL THEORY (INCLUDING, WITHOUT LIMITATION,
299
+ NEGLIGENCE) OR OTHERWISE FOR ANY DIRECT, SPECIAL, INDIRECT,
300
+ INCIDENTAL, CONSEQUENTIAL, PUNITIVE, EXEMPLARY, OR OTHER LOSSES,
301
+ COSTS, EXPENSES, OR DAMAGES ARISING OUT OF THIS PUBLIC LICENSE OR
302
+ USE OF THE LICENSED MATERIAL, EVEN IF THE LICENSOR HAS BEEN
303
+ ADVISED OF THE POSSIBILITY OF SUCH LOSSES, COSTS, EXPENSES, OR
304
+ DAMAGES. WHERE A LIMITATION OF LIABILITY IS NOT ALLOWED IN FULL OR
305
+ IN PART, THIS LIMITATION MAY NOT APPLY TO YOU.
306
+
307
+ c. The disclaimer of warranties and limitation of liability provided
308
+ above shall be interpreted in a manner that, to the extent
309
+ possible, most closely approximates an absolute disclaimer and
310
+ waiver of all liability.
311
+
312
+
313
+ Section 6 -- Term and Termination.
314
+
315
+ a. This Public License applies for the term of the Copyright and
316
+ Similar Rights licensed here. However, if You fail to comply with
317
+ this Public License, then Your rights under this Public License
318
+ terminate automatically.
319
+
320
+ b. Where Your right to use the Licensed Material has terminated under
321
+ Section 6(a), it reinstates:
322
+
323
+ 1. automatically as of the date the violation is cured, provided
324
+ it is cured within 30 days of Your discovery of the
325
+ violation; or
326
+
327
+ 2. upon express reinstatement by the Licensor.
328
+
329
+ For the avoidance of doubt, this Section 6(b) does not affect any
330
+ right the Licensor may have to seek remedies for Your violations
331
+ of this Public License.
332
+
333
+ c. For the avoidance of doubt, the Licensor may also offer the
334
+ Licensed Material under separate terms or conditions or stop
335
+ distributing the Licensed Material at any time; however, doing so
336
+ will not terminate this Public License.
337
+
338
+ d. Sections 1, 5, 6, 7, and 8 survive termination of this Public
339
+ License.
340
+
341
+
342
+ Section 7 -- Other Terms and Conditions.
343
+
344
+ a. The Licensor shall not be bound by any additional or different
345
+ terms or conditions communicated by You unless expressly agreed.
346
+
347
+ b. Any arrangements, understandings, or agreements regarding the
348
+ Licensed Material not stated herein are separate from and
349
+ independent of the terms and conditions of this Public License.
350
+
351
+
352
+ Section 8 -- Interpretation.
353
+
354
+ a. For the avoidance of doubt, this Public License does not, and
355
+ shall not be interpreted to, reduce, limit, restrict, or impose
356
+ conditions on any use of the Licensed Material that could lawfully
357
+ be made without permission under this Public License.
358
+
359
+ b. To the extent possible, if any provision of this Public License is
360
+ deemed unenforceable, it shall be automatically reformed to the
361
+ minimum extent necessary to make it enforceable. If the provision
362
+ cannot be reformed, it shall be severed from this Public License
363
+ without affecting the enforceability of the remaining terms and
364
+ conditions.
365
+
366
+ c. No term or condition of this Public License will be waived and no
367
+ failure to comply consented to unless expressly agreed to by the
368
+ Licensor.
369
+
370
+ d. Nothing in this Public License constitutes or may be interpreted
371
+ as a limitation upon, or waiver of, any privileges and immunities
372
+ that apply to the Licensor or You, including from the legal
373
+ processes of any jurisdiction or authority.
374
+
375
+
376
+ =======================================================================
377
+
378
+ Creative Commons is not a party to its public
379
+ licenses. Notwithstanding, Creative Commons may elect to apply one of
380
+ its public licenses to material it publishes and in those instances
381
+ will be considered the “Licensor.” The text of the Creative Commons
382
+ public licenses is dedicated to the public domain under the CC0 Public
383
+ Domain Dedication. Except for the limited purpose of indicating that
384
+ material is shared under a Creative Commons public license or as
385
+ otherwise permitted by the Creative Commons policies published at
386
+ creativecommons.org/policies, Creative Commons does not authorize the
387
+ use of the trademark "Creative Commons" or any other trademark or logo
388
+ of Creative Commons without its prior written consent including,
389
+ without limitation, in connection with any unauthorized modifications
390
+ to any of its public licenses or any other arrangements,
391
+ understandings, or agreements concerning use of licensed material. For
392
+ the avoidance of doubt, this paragraph does not form part of the
393
+ public licenses.
394
+
395
+ Creative Commons may be contacted at creativecommons.org.
acoustic/czech/mfa/v3.3.0/README.md ADDED
@@ -0,0 +1,137 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Czech MFA acoustic model v3.3.0
2
+
3
+ [Link to documentation on mfa-models](https://mfa-models.readthedocs.io/en/main/acoustic/czech_mfa.html)
4
+
5
+ Jump to section:
6
+
7
+ - [Model details](#model-details)
8
+ - [Installation](#installation)
9
+ - [Intended use](#intended-use)
10
+ - [Performance Factors](#performance-factors)
11
+ - [Metrics](#metrics)
12
+ - [Ethical considerations](#ethical-considerations)
13
+ - [Troubleshooting issues](#troubleshooting-issues)
14
+ - [Training data](#training-data)
15
+ - [Evaluation data](#evaluation-data)
16
+
17
+ ## Model details
18
+
19
+ - **Maintainer:** [Montreal Forced Aligner](https://montreal-forced-aligner.readthedocs.io/)
20
+ - **Language:** [Czech](https://en.wikipedia.org/wiki/Czech_language)
21
+ - **Dialect:** N/A
22
+ - **Phone set:** [MFA](https://mfa-models.readthedocs.io/en/refactor/mfa_phone_set.html#czech)
23
+ - **Model type:** `Acoustic`
24
+ - **Features:** `MFCC`
25
+ - **Architecture:** `gmm-hmm`
26
+ - **Model version:** `v3.3.0`
27
+ - **Trained date:** `2025-05-05`
28
+ - **Compatible MFA version:** `v3.3.0`
29
+ - **License:** [CC BY 4.0](https://github.com/MontrealCorpusTools/mfa-models/tree/main/acoustic/czech/mfa/v3.3.0/LICENSE)
30
+ - **Citation:**
31
+
32
+ ```bibtex
33
+ @techreport{mfa_czech_mfa_acoustic_2025,
34
+ author={McAuliffe, Michael and Sonderegger, Morgan},
35
+ title={Czech MFA acoustic model v3.3.0},
36
+ address={\url{https://mfa-models.readthedocs.io/acoustic/Czech/Czech MFA acoustic model v3_3_0.html}},
37
+ year={2025},
38
+ month={May},
39
+ }
40
+ ```
41
+
42
+ - If you have comments or questions about this model, you can check [previous MFA model discussion posts](https://github.com/MontrealCorpusTools/mfa-models/discussions?discussions_q=Czech+MFA+acoustic+model+v3.3.0) or create [a new one](https://github.com/MontrealCorpusTools/mfa-models/discussions/new).
43
+
44
+ ## Installation
45
+
46
+ Install from the [MFA command line](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/models/index.html):
47
+
48
+ ```
49
+ mfa model download acoustic czech_mfa
50
+ ```
51
+
52
+ Or download from [the release page](https://github.com/MontrealCorpusTools/mfa-models/releases/tag/acoustic-czech_mfa-v3.3.0).
53
+
54
+ ## Intended use
55
+
56
+ This model is intended for forced alignment of [Czech](https://en.wikipedia.org/wiki/Czech_language) transcripts.
57
+
58
+ This model uses the [MFA](https://mfa-models.readthedocs.io/en/refactor/mfa_phone_set.html#czech) phone set for Czech, and was trained with the pronunciation dictionaries above. Pronunciations can be added on top of the dictionary, as long as no additional phones are introduced.
59
+
60
+ ## Performance Factors
61
+
62
+ As forced alignment is a relatively well-constrained problem (given accurate transcripts), this model should be applicable to a range of recording conditions and speakers. However, please note that it was trained on read speech in low-noise environments, so as your data diverges from that, you may run into alignment issues or need to [increase the beam size of MFA](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/configuration/#configuring-specific-commands) or see other recommendations in the [troubleshooting section below](#troubleshooting-issues).
63
+
64
+ Please note as well that MFA does not use state-of-the-art ASR models for forced alignment. You may get better performance (especially on speech-to-text tasks) using other frameworks like [Coqui](https://coqui.ai/).
65
+
66
+ ## Metrics
67
+
68
+ Acoustic models are typically generated as one component of a larger ASR system where the metric is word error rate (WER). For forced alignment, there is typically not the same sort of gold standard measure for most languages.
69
+
70
+ As a rough approximation of the acoustic model quality, we evaluated it against the corpus it was trained on alongside a language model trained from the same data. Key caveat here is that this is not a typical WER measure on held out data, so it should not be taken as a hard measure of how well an acoustic model will generalize to your data, but rather is more of a sanity check that the training data quality was sufficiently high.
71
+
72
+ Using the pronunciation dictionaries and language models above:
73
+
74
+ - **WER:** `0%`
75
+ - **CER:** `0%`
76
+
77
+ ## Ethical considerations
78
+
79
+ Deploying any Speech-to-Text model into any production setting has ethical implications. You should consider these implications before use.
80
+
81
+ ### Demographic Bias
82
+
83
+ You should assume every machine learning model has demographic bias unless proven otherwise. For STT models, it is often the case that transcription accuracy is better for men than it is for women. If you are using this model in production, you should acknowledge this as a potential issue.
84
+
85
+ ### Surveillance
86
+
87
+ Speech-to-Text technologies may be misused to invade the privacy of others by recording and mining information from private conversations. This kind of individual privacy is protected by law in many countries. You should not assume consent to record and analyze private speech.
88
+
89
+
90
+ ## Troubleshooting issues
91
+
92
+ Machine learning models (like this acoustic model) perform best on data that is similar to the data on which they were trained.
93
+
94
+ The primary sources of variability in forced alignment will be the applicability of the pronunciation dictionary and how similar the speech, demographics, and recording conditions are. If you encounter issues in alignment, there are couple of avenues to improve performance:
95
+
96
+ 1. [Increase the beam size of MFA](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/configuration/#configuring-specific-commands)
97
+
98
+ * MFA defaults to a narrow beam to ensure quick alignment and also as a way to detect potential issues in your dataset, but depending on your data, you might benefit from boosting the beam to 100 or higher.
99
+
100
+ 2. Add pronunciations to the pronunciation dictionary
101
+
102
+ * This model was trained a particular dialect/style, and so adding pronunciations more representative of the variety spoken in your dataset will help alignment.
103
+
104
+ 3. Check the quality of your data
105
+
106
+ * MFA includes a [validator utility](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/data_validation.html), which aims to detect issues in the dataset.
107
+ * Use MFA's [anchor utility](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/workflows/anchor.html) to visually inspect your data as MFA sees it and correct issues in transcription or OOV items.
108
+
109
+ 4. Adapt the model to your data
110
+
111
+ * MFA has an [adaptation command](https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/workflows/adapt_acoustic_model.html) to adapt some of the model to your data based on an initial alignment, and then run another alignment with the adapted model.
112
+
113
+ ## Training data
114
+
115
+ This model was trained on the following corpora:
116
+
117
+
118
+
119
+ * [Common Voice Czech](../../../../corpus/czech/common_voice_czech/9.0/README.md):
120
+ * **Hours:** `66.33`
121
+ * **Speakers:** `523`
122
+ * **Utterances:** `55,391`
123
+
124
+ * [GlobalPhone Czech](../../../../corpus/czech/globalphone_czech/3.1/README.md):
125
+ * **Hours:** `31.67`
126
+ * **Speakers:** `101`
127
+ * **Utterances:** `12,322`
128
+
129
+ * [Large Corpus of Czech Parliament Plenary Hearings](../../../../corpus/czech/large_corpus_of_czech_parliament_plenary_hearings/README.md):
130
+ * **Hours:** `450.95`
131
+ * **Speakers:** `84`
132
+ * **Utterances:** `194,181`
133
+
134
+ * [Czech Parliament Meetings](../../../../corpus/czech/czech_parliament_meetings/README.md):
135
+ * **Hours:** `88.81`
136
+ * **Speakers:** `67`
137
+ * **Utterances:** `51,494`
acoustic/czech/mfa/v3.3.0/meta.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "name": "czech_mfa",
3
+ "language": "Czech",
4
+ "dialect": "",
5
+ "phone_set": "MFA",
6
+ "version": "3.3.0",
7
+ "maintainer": "[Montreal Forced Aligner](https://montreal-forced-aligner.readthedocs.io/)",
8
+ "citation": "@techreport{mfa_czech_mfa_acoustic_2025,\n\tauthor={McAuliffe, Michael and Sonderegger, Morgan},\n\ttitle={Czech MFA acoustic model v3.3.0},\n\taddress={\\url{https://mfa-models.readthedocs.io/acoustic/Czech/Czech MFA acoustic model v3_3_0.html}},\n\tyear={2025},\n\tmonth={May},\n}",
9
+ "license": "CC BY 4.0",
10
+ "license_link": "[CC BY 4.0](https://github.com/MontrealCorpusTools/mfa-models/tree/main/acoustic/czech/mfa/v3.3.0/LICENSE)",
11
+ "architecture": "gmm-hmm",
12
+ "features": "MFCC",
13
+ "evaluation": {},
14
+ "decode": {},
15
+ "train_date": "2025-05-05"
16
+ }