Hugging Face's logo

pierreguillou
/
whisper-medium-french

Automatic Speech Recognition
Transformers
PyTorch
TensorBoard
French
whisper
Generated from Trainer
whisper-event
Eval Results
Inference Endpoints
Model card Files Metrics Community
4
whisper-medium-french / README.md
pierreguillou's picture
pierreguillou
Update README.md
a33134e
preview code
|
raw
history blame
3.02 kB
metadata
license: apache-2.0
tags:
  - generated_from_trainer
datasets:
  - common_voice_11_0
metrics:
  - wer
  - wer_norm
model-index:
  - name: openai/whisper-medium
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: common_voice_11_0
          type: common_voice_11_0
          config: fr
          split: test
          args: fr
        metrics:
          - name: Wer
            type: wer
            value: 11.1406
          - name: Wer (without normalization)
            type: wer_without_norm
            value: 15.89689189275029

French Medium Whisper

This model is a fine-tuned version of openai/whisper-medium on the common_voice_11_0 dataset. It achieves the following results on the evaluation set:

  • Loss: 0.2664
  • Wer (without normalization): 15.8969
  • Wer (with normalization): 11.1406

New SOTA

The Normalized WER in the OpenAI Whisper article with the Common Voice 9.0 test dataset is 16.0.

As this test dataset is similar to the Common Voice 11.0 test dataset used to evaluate our model (WER and WER Norm), it means that our French Medium Whisper is better than the Medium Whisper model at transcribing audios French in text.

OpenAI results with Whisper Medium and Test dataset of Commons Voice 9.0

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 1e-05
  • train_batch_size: 32
  • eval_batch_size: 16
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 500
  • training_steps: 5000
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Wer Wer Norm
0.2695 0.2 1000 0.3080 17.8083 12.9791
0.2099 0.4 2000 0.2981 17.4792 12.4242
0.1978 0.6 3000 0.2864 16.7767 12.0913
0.1455 0.8 4000 0.2752 16.4597 11.8966
0.1712 1.0 5000 0.2664 15.8969 11.1406

Framework versions

  • Transformers 4.26.0.dev0
  • Pytorch 1.13.0+cu117
  • Datasets 2.7.1.dev0
  • Tokenizers 0.13.2