---
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: 15.89689189275029
    - name: Wer norm
      type: wer
      value: 11.1406
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# openai/whisper-medium

This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on the common_voice_11_0 dataset.
It achieves the following results on the evaluation set:
- Loss: 0.2664
- Wer: 15.8969
- Wer Norm (Normalized Wer): **11.1406**

## New SOTA

The Normalized WER in the [OpenAI Whisper article](https://cdn.openai.com/papers/whisper.pdf) with the [Common Voice 9.0](https://huggingface.co/datasets/mozilla-foundation/common_voice_9_0) test dataset is 16.0. 

This means that our French Medium Whisper is better than the model [Medium Whisper]() at transcribing French audios into text.

![OpenAI results with Whisper Medium and Test dataset of Commons Voice 9.0](https://huggingface.co/pierreguillou/whisper-medium-french/resolve/main/whisper_medium_french_wer_commonvoice9.png)

## 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