pierreguillou
/

whisper-medium-french

Automatic Speech Recognition

Generated from Trainer

Inference Endpoints

Model card Files Files and versions Metrics Training metrics Community

pierreguillou commited on Dec 7, 2022

Commit

dbbe639

•

1 Parent(s): 6597046

Update README.md

Files changed (1) hide show

README.md +9 -1

README.md CHANGED Viewed

@@ -37,7 +37,15 @@ This model is a fine-tuned version of [openai/whisper-medium](https://huggingfac
 It achieves the following results on the evaluation set:
 - Loss: 0.2664
 - Wer: 15.8969
-- Wer Norm (Normalized Wer): 11.1406
 ## Model description

 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