alibabasglab commited on
Commit
9eb22f6
Β·
verified Β·
1 Parent(s): 3b7b7eb

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +11 -12
README.md CHANGED
@@ -1,13 +1,14 @@
1
  ---
2
  license: apache-2.0
3
  ---
4
- LJSpeech-1.1 High-Resolution Dataset (48,000 Hz)
5
 
6
- Overview
7
 
8
- The LJSpeech-1.1 dataset, widely recognized for its utility in text-to-speech (TTS) and other speech processing tasks, has now been enhanced through a cutting-edge speech super-resolution algorithm. The original dataset, which featured a sampling rate of 22,050 Hz, has been upscaled to 48,000 Hz, providing a high-fidelity version suitable for advanced audio processing tasks.
 
 
9
 
10
- Key Features
11
 
12
  - High-Resolution Audio: The dataset now offers audio files at a sampling rate of 48,000 Hz, delivering enhanced perceptual quality with richer high-frequency details.
13
 
@@ -17,9 +18,7 @@ Key Features
17
 
18
  - Open Source: Freely available for academic and research purposes, fostering innovation in the speech and audio domains.
19
 
20
- Enhancement Details
21
-
22
- Original Dataset
23
 
24
  - Source: The original LJSpeech-1.1 dataset contains 13,100 audio clips of a single female speaker reading passages from public domain books.
25
 
@@ -27,7 +26,7 @@ Original Dataset
27
 
28
  - Annotations: Each audio clip is paired with a corresponding text transcript.
29
 
30
- Super-Resolution Processing
31
 
32
  The original 22,050 Hz audio recordings were processed using a state-of-the-art MossFormer2-based speech super-resolution model. This model employs:
33
 
@@ -37,7 +36,7 @@ The original 22,050 Hz audio recordings were processed using a state-of-the-art
37
 
38
  - High-Frequency Reconstruction: Algorithms specifically tuned to recover lost high-frequency components, ensuring smooth and artifact-free enhancement.
39
 
40
- Output Format
41
 
42
  - Sampling Rate: 48,000 Hz
43
 
@@ -47,7 +46,7 @@ Output Format
47
 
48
  - Channel Configuration: Mono
49
 
50
- Use Cases
51
 
52
  1. Text-to-Speech (TTS) Synthesis
53
  β—‹ Train high-fidelity TTS systems capable of generating human-like speech.
@@ -61,7 +60,7 @@ Use Cases
61
  β—‹ Restore low-resolution or degraded speech signals for professional applications.
62
  β—‹ Create high-quality voiceovers and narration for multimedia projects.
63
 
64
- File Structure
65
 
66
  The dataset retains the original LJSpeech-1.1 structure, ensuring ease of use:
67
 
@@ -71,6 +70,6 @@ The dataset retains the original LJSpeech-1.1 structure, ensuring ease of use:
71
  β”œβ”€β”€ wavs/ # Directory containing 48,000 Hz WAV files
72
  └── LICENSE.txt # License information
73
  ```
74
- Licensing
75
 
76
  The LJSpeech-1.1 High-Resolution Dataset is released under the same open license as the original LJSpeech-1.1 dataset (https://keithito.com/LJ-Speech-Dataset/). Users are free to use, modify, and share the dataset for academic and non-commercial purposes, provided proper attribution is given.
 
1
  ---
2
  license: apache-2.0
3
  ---
4
+ # LJSpeech-1.1 High-Resolution Dataset (48,000 Hz)
5
 
 
6
 
7
+ The LJSpeech-1.1 dataset, widely recognized for its utility in text-to-speech (TTS) and other speech processing tasks, has now been enhanced through a cutting-edge speech
8
+ super-resolution algorithm. The original dataset, which featured a sampling rate of 22,050 Hz, has been upscaled to 48,000 Hz using [**ClearerVoice-Studio**](https://github.com/modelscope/ClearerVoice-Studio), providing a high-fidelity version suitable
9
+ for advanced audio processing tasks.
10
 
11
+ **Key Features**
12
 
13
  - High-Resolution Audio: The dataset now offers audio files at a sampling rate of 48,000 Hz, delivering enhanced perceptual quality with richer high-frequency details.
14
 
 
18
 
19
  - Open Source: Freely available for academic and research purposes, fostering innovation in the speech and audio domains.
20
 
21
+ **Original Dataset**
 
 
22
 
23
  - Source: The original LJSpeech-1.1 dataset contains 13,100 audio clips of a single female speaker reading passages from public domain books.
24
 
 
26
 
27
  - Annotations: Each audio clip is paired with a corresponding text transcript.
28
 
29
+ **Super-Resolution Processing**
30
 
31
  The original 22,050 Hz audio recordings were processed using a state-of-the-art MossFormer2-based speech super-resolution model. This model employs:
32
 
 
36
 
37
  - High-Frequency Reconstruction: Algorithms specifically tuned to recover lost high-frequency components, ensuring smooth and artifact-free enhancement.
38
 
39
+ **Output Format**
40
 
41
  - Sampling Rate: 48,000 Hz
42
 
 
46
 
47
  - Channel Configuration: Mono
48
 
49
+ **Use Cases**
50
 
51
  1. Text-to-Speech (TTS) Synthesis
52
  β—‹ Train high-fidelity TTS systems capable of generating human-like speech.
 
60
  β—‹ Restore low-resolution or degraded speech signals for professional applications.
61
  β—‹ Create high-quality voiceovers and narration for multimedia projects.
62
 
63
+ **File Structure**
64
 
65
  The dataset retains the original LJSpeech-1.1 structure, ensuring ease of use:
66
 
 
70
  β”œβ”€β”€ wavs/ # Directory containing 48,000 Hz WAV files
71
  └── LICENSE.txt # License information
72
  ```
73
+ **Licensing**
74
 
75
  The LJSpeech-1.1 High-Resolution Dataset is released under the same open license as the original LJSpeech-1.1 dataset (https://keithito.com/LJ-Speech-Dataset/). Users are free to use, modify, and share the dataset for academic and non-commercial purposes, provided proper attribution is given.