longdnk commited on
Commit
2b4ebca
·
verified ·
1 Parent(s): 56085c8

first commit

Browse files
config.json ADDED
@@ -0,0 +1,114 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/wav2vec2-conformer-rel-pos-large-100h-ft",
3
+ "activation_dropout": 0.1,
4
+ "adapter_kernel_size": 3,
5
+ "adapter_stride": 2,
6
+ "add_adapter": false,
7
+ "apply_spec_augment": true,
8
+ "architectures": [
9
+ "Wav2Vec2ConformerForCTC"
10
+ ],
11
+ "attention_dropout": 0.1,
12
+ "bos_token_id": 1,
13
+ "classifier_proj_size": 256,
14
+ "codevector_dim": 768,
15
+ "conformer_conv_dropout": 0.1,
16
+ "contrastive_logits_temperature": 0.1,
17
+ "conv_bias": true,
18
+ "conv_depthwise_kernel_size": 31,
19
+ "conv_dim": [
20
+ 512,
21
+ 512,
22
+ 512,
23
+ 512,
24
+ 512,
25
+ 512,
26
+ 512
27
+ ],
28
+ "conv_kernel": [
29
+ 10,
30
+ 3,
31
+ 3,
32
+ 3,
33
+ 3,
34
+ 2,
35
+ 2
36
+ ],
37
+ "conv_stride": [
38
+ 5,
39
+ 2,
40
+ 2,
41
+ 2,
42
+ 2,
43
+ 2,
44
+ 2
45
+ ],
46
+ "ctc_loss_reduction": "mean",
47
+ "ctc_zero_infinity": false,
48
+ "diversity_loss_weight": 0.1,
49
+ "do_stable_layer_norm": true,
50
+ "eos_token_id": 2,
51
+ "feat_extract_activation": "gelu",
52
+ "feat_extract_dropout": 0.0,
53
+ "feat_extract_norm": "layer",
54
+ "feat_proj_dropout": 0.1,
55
+ "feat_quantizer_dropout": 0.0,
56
+ "final_dropout": 0.1,
57
+ "hidden_act": "swish",
58
+ "hidden_dropout": 0.1,
59
+ "hidden_dropout_prob": 0.1,
60
+ "hidden_size": 1024,
61
+ "initializer_range": 0.02,
62
+ "intermediate_size": 4096,
63
+ "layer_norm_eps": 1e-05,
64
+ "layerdrop": 0.0,
65
+ "mask_feature_length": 10,
66
+ "mask_feature_min_masks": 0,
67
+ "mask_feature_prob": 0.0,
68
+ "mask_time_length": 10,
69
+ "mask_time_min_masks": 2,
70
+ "mask_time_prob": 0.05,
71
+ "max_source_positions": 5000,
72
+ "model_type": "wav2vec2-conformer",
73
+ "num_adapter_layers": 3,
74
+ "num_attention_heads": 16,
75
+ "num_codevector_groups": 2,
76
+ "num_codevectors_per_group": 320,
77
+ "num_conv_pos_embedding_groups": 16,
78
+ "num_conv_pos_embeddings": 128,
79
+ "num_feat_extract_layers": 7,
80
+ "num_hidden_layers": 24,
81
+ "num_negatives": 100,
82
+ "output_hidden_size": 1024,
83
+ "pad_token_id": 109,
84
+ "position_embeddings_type": "relative",
85
+ "proj_codevector_dim": 768,
86
+ "rotary_embedding_base": 10000,
87
+ "tdnn_dilation": [
88
+ 1,
89
+ 2,
90
+ 3,
91
+ 1,
92
+ 1
93
+ ],
94
+ "tdnn_dim": [
95
+ 512,
96
+ 512,
97
+ 512,
98
+ 512,
99
+ 1500
100
+ ],
101
+ "tdnn_kernel": [
102
+ 5,
103
+ 3,
104
+ 3,
105
+ 1,
106
+ 1
107
+ ],
108
+ "torch_dtype": "float32",
109
+ "transformers_version": "4.45.1",
110
+ "use_weighted_layer_sum": false,
111
+ "vocab_size": 112,
112
+ "xvector_output_dim": 512,
113
+ "gradient_checkpointing": true
114
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e51cc6359c2a0fc680add035a03da0c85c954123dc65ae49fe3fab1a4f06ace0
3
+ size 2475018768
preprocessor_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_normalize": true,
3
+ "feature_extractor_type": "Wav2Vec2FeatureExtractor",
4
+ "feature_size": 1,
5
+ "padding_side": "right",
6
+ "padding_value": 0.0,
7
+ "return_attention_mask": false,
8
+ "sampling_rate": 16000
9
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0b1fea49c5c927597b356c633fe273690cb0a3eb2596b6648a4efe2f70195e8f
3
+ size 2475225738
special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>"}
tokenizer_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "unk_token": "<unk>",
3
+ "bos_token": "<s>",
4
+ "eos_token": "</s>",
5
+ "pad_token": "<pad>",
6
+ "do_lower_case": false,
7
+ "word_delimiter_token": "|",
8
+ "tokenizer_class": "Wav2Vec2CTCTokenizer"
9
+ }
vocab.json ADDED
@@ -0,0 +1,112 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "ẻ": 0,
3
+ "6": 1,
4
+ "ụ": 2,
5
+ "í": 3,
6
+ "3": 4,
7
+ "ỹ": 5,
8
+ "ý": 6,
9
+ "ẩ": 7,
10
+ "ở": 8,
11
+ "ề": 9,
12
+ "õ": 10,
13
+ "7": 11,
14
+ "ê": 12,
15
+ "ứ": 13,
16
+ "ỏ": 14,
17
+ "v": 15,
18
+ "ỷ": 16,
19
+ "a": 17,
20
+ "l": 18,
21
+ "ự": 19,
22
+ "q": 20,
23
+ "ờ": 21,
24
+ "j": 22,
25
+ "ố": 23,
26
+ "à": 24,
27
+ "ỗ": 25,
28
+ "n": 26,
29
+ "é": 27,
30
+ "ủ": 28,
31
+ "у": 29,
32
+ "ô": 30,
33
+ "u": 31,
34
+ "y": 32,
35
+ "ằ": 33,
36
+ "4": 34,
37
+ "w": 35,
38
+ "b": 36,
39
+ "ệ": 37,
40
+ "ễ": 38,
41
+ "s": 39,
42
+ "ì": 40,
43
+ "ầ": 41,
44
+ "ỵ": 42,
45
+ "8": 43,
46
+ "d": 44,
47
+ "ể": 45,
48
+ "r": 47,
49
+ "ũ": 48,
50
+ "c": 49,
51
+ "ạ": 50,
52
+ "9": 51,
53
+ "ế": 52,
54
+ "ù": 53,
55
+ "ỡ": 54,
56
+ "2": 55,
57
+ "t": 56,
58
+ "i": 57,
59
+ "g": 58,
60
+ "́": 59,
61
+ "ử": 60,
62
+ "̀": 61,
63
+ "á": 62,
64
+ "0": 63,
65
+ "ậ": 64,
66
+ "e": 65,
67
+ "ộ": 66,
68
+ "m": 67,
69
+ "ẳ": 68,
70
+ "ợ": 69,
71
+ "ĩ": 70,
72
+ "h": 71,
73
+ "â": 72,
74
+ "ú": 73,
75
+ "ọ": 74,
76
+ "ồ": 75,
77
+ "ặ": 76,
78
+ "f": 77,
79
+ "ữ": 78,
80
+ "ắ": 79,
81
+ "ỳ": 80,
82
+ "x": 81,
83
+ "ó": 82,
84
+ "ã": 83,
85
+ "ổ": 84,
86
+ "ị": 85,
87
+ "̣": 86,
88
+ "z": 87,
89
+ "ả": 88,
90
+ "đ": 89,
91
+ "è": 90,
92
+ "ừ": 91,
93
+ "ò": 92,
94
+ "ẵ": 93,
95
+ "1": 94,
96
+ "ơ": 95,
97
+ "k": 96,
98
+ "ẫ": 97,
99
+ "p": 98,
100
+ "ấ": 99,
101
+ "ẽ": 100,
102
+ "ỉ": 101,
103
+ "ớ": 102,
104
+ "ẹ": 103,
105
+ "ă": 104,
106
+ "o": 105,
107
+ "ư": 106,
108
+ "5": 107,
109
+ "|": 46,
110
+ "<unk>": 108,
111
+ "<pad>": 109
112
+ }