speechbrainteam's picture
Update README.md
- fr
thumbnail: null
pipeline_tag: automatic-speech-recognition
- whisper
- pytorch
- speechbrain
- Transformer
- hf-asr-leaderboard
license: apache-2.0
- commonvoice
- wer
- cer
- name: asr-whisper-large-v2-commonvoice-fr
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
name: CommonVoice 10.0 (French)
type: mozilla-foundation/common_voice_10_0
config: fr
split: test
language: fr
- name: Test WER
type: wer
value: '10.62'
<iframe src="https://ghbtns.com/github-btn.html?user=speechbrain&repo=speechbrain&type=star&count=true&size=large&v=2" frameborder="0" scrolling="0" width="170" height="30" title="GitHub"></iframe>
# whisper large-v2 fine-tuned on CommonVoice French
This repository provides all the necessary tools to perform automatic speech
recognition from an end-to-end whisper model fine-tuned on CommonVoice (French Language) within
SpeechBrain. For a better experience, we encourage you to learn more about
The performance of the model is the following:
| Release | Test CER | Test WER | GPUs |
|:-------------:|:--------------:|:--------------:| :--------:|
| 01-02-23 | 3.83 | 10.62 | 1xV100 16GB |
## Pipeline description
This ASR system is composed of whisper encoder-decoder blocks:
- The pretrained whisper-large-v2 encoder is frozen.
- The pretrained Whisper tokenizer is used.
- A pretrained Whisper-large-v2 decoder ([openai/whisper-large-v2](https://huggingface.co/openai/whisper-large-v2)) is finetuned on CommonVoice FR.
The obtained final acoustic representation is given to the greedy decoder.
The system is trained with recordings sampled at 16kHz (single channel).
The code will automatically normalize your audio (i.e., resampling + mono channel selection) when calling *transcribe_file* if needed.
## Install SpeechBrain
First of all, please install tranformers and SpeechBrain with the following command:
pip install speechbrain transformers==4.28.0
Please notice that we encourage you to read our tutorials and learn more about
### Transcribing your own audio files (in French)
from speechbrain.pretrained import WhisperASR
asr_model = WhisperASR.from_hparams(source="speechbrain/asr-whisper-large-v2-commonvoice-fr", savedir="pretrained_models/asr-whisper-large-v2-commonvoice-fr")
### Inference on GPU
To perform inference on the GPU, add `run_opts={"device":"cuda"}` when calling the `from_hparams` method.
### Training
The model was trained with SpeechBrain.
To train it from scratch follow these steps:
1. Clone SpeechBrain:
git clone https://github.com/speechbrain/speechbrain/
2. Install it:
cd speechbrain
pip install -r requirements.txt
pip install -e .
3. Run Training:
cd recipes/CommonVoice/ASR/transformer/
python train_with_whisper.py hparams/train_fr_hf_whisper.yaml --data_folder=your_data_folder
You can find our training results (models, logs, etc) [here](https://drive.google.com/drive/folders/1_iI_G-pMYNeyLsvmHPgNR6gPi8zazkF4?usp=share_link).
### Limitations
The SpeechBrain team does not provide any warranty on the performance achieved by this model when used on other datasets.
#### Referencing SpeechBrain
author = {Ravanelli, Mirco and Parcollet, Titouan and Rouhe, Aku and Plantinga, Peter and Rastorgueva, Elena and Lugosch, Loren and Dawalatabad, Nauman and Ju-Chieh, Chou and Heba, Abdel and Grondin, Francois and Aris, William and Liao, Chien-Feng and Cornell, Samuele and Yeh, Sung-Lin and Na, Hwidong and Gao, Yan and Fu, Szu-Wei and Subakan, Cem and De Mori, Renato and Bengio, Yoshua },
title = {SpeechBrain},
year = {2021},
publisher = {GitHub},
journal = {GitHub repository},
howpublished = {\\\\url{https://github.com/speechbrain/speechbrain}},
#### About SpeechBrain
SpeechBrain is an open-source and all-in-one speech toolkit. It is designed to be simple, extremely flexible, and user-friendly. Competitive or state-of-the-art performance is obtained in various domains.
Website: https://speechbrain.github.io/
GitHub: https://github.com/speechbrain/speechbrain