Commit
•
dd3f886
1
Parent(s):
1062137
Update README.md
Browse files
README.md
CHANGED
|
@@ -33,15 +33,18 @@ tags:
|
|
| 33 |
- [Contributions](#contributions)
|
| 34 |
|
| 35 |
## Dataset Description
|
|
|
|
| 36 |
- **Homepage:** [MultiLingual LibriSpeech ASR corpus](http://www.openslr.org/94)
|
| 37 |
- **Repository:** [Needs More Information]
|
| 38 |
- **Paper:** [MLS: A Large-Scale Multilingual Dataset for Speech Research](https://arxiv.org/abs/2012.03411)
|
| 39 |
- **Leaderboard:** [Paperswithcode Leaderboard](https://paperswithcode.com/dataset/multilingual-librispeech)
|
| 40 |
|
| 41 |
### Dataset Summary
|
|
|
|
| 42 |
Multilingual LibriSpeech (MLS) dataset is a large multilingual corpus suitable for speech research. The dataset is derived from read audiobooks from LibriVox and consists of 8 languages - English, German, Dutch, Spanish, French, Italian, Portuguese, Polish.
|
| 43 |
|
| 44 |
### Supported Tasks and Leaderboards
|
|
|
|
| 45 |
- `automatic-speech-recognition`, `audio-speaker-identification`: The dataset can be used to train a model for Automatic Speech Recognition (ASR). The model is presented with an audio file and asked to transcribe the audio file to written text. The most common evaluation metric is the word error rate (WER). The task has an active leaderboard which can be found at https://paperswithcode.com/dataset/multilingual-librispeech and ranks models based on their WER.
|
| 46 |
|
| 47 |
|
|
|
|
| 33 |
- [Contributions](#contributions)
|
| 34 |
|
| 35 |
## Dataset Description
|
| 36 |
+
|
| 37 |
- **Homepage:** [MultiLingual LibriSpeech ASR corpus](http://www.openslr.org/94)
|
| 38 |
- **Repository:** [Needs More Information]
|
| 39 |
- **Paper:** [MLS: A Large-Scale Multilingual Dataset for Speech Research](https://arxiv.org/abs/2012.03411)
|
| 40 |
- **Leaderboard:** [Paperswithcode Leaderboard](https://paperswithcode.com/dataset/multilingual-librispeech)
|
| 41 |
|
| 42 |
### Dataset Summary
|
| 43 |
+
|
| 44 |
Multilingual LibriSpeech (MLS) dataset is a large multilingual corpus suitable for speech research. The dataset is derived from read audiobooks from LibriVox and consists of 8 languages - English, German, Dutch, Spanish, French, Italian, Portuguese, Polish.
|
| 45 |
|
| 46 |
### Supported Tasks and Leaderboards
|
| 47 |
+
|
| 48 |
- `automatic-speech-recognition`, `audio-speaker-identification`: The dataset can be used to train a model for Automatic Speech Recognition (ASR). The model is presented with an audio file and asked to transcribe the audio file to written text. The most common evaluation metric is the word error rate (WER). The task has an active leaderboard which can be found at https://paperswithcode.com/dataset/multilingual-librispeech and ranks models based on their WER.
|
| 49 |
|
| 50 |
|