whisper-medium-french Open-Source French Speech Recognition Model - Precise Recognition, Better Performance than the Original Version

Whisper Medium French

Developed by pierreguillou

A French speech recognition model fine-tuned on the common_voice_11_0 dataset based on openai/whisper-medium, achieving a standardized WER of 11.1406, outperforming the original model.

Speech Recognition

Transformers

FrenchOpen Source License:Apache-2.0 #French Speech Recognition #Low Word Error Rate #Common Voice Fine-tuning

Downloads 260

Release Time : 12/6/2022

Model Overview

This model is an automatic speech recognition (ASR) model optimized specifically for French, capable of accurately transcribing French audio into text.

Model Features

High-Performance French Recognition

Achieves a standardized WER of 11.1406 on the Common Voice 11.0 test set, outperforming the original Whisper-medium model.

Optimized Training

Fine-tuned for 5000 steps on the Common Voice 11.0 dataset using linear learning rate scheduling and the Adam optimizer.

Standardized Output

Provides both standardized and non-standardized Word Error Rate (WER) evaluations to ensure transcription results better meet practical application needs.

Model Capabilities

French Speech Recognition

Audio Transcription

Automatic Speech-to-Text

Use Cases

Speech Transcription

French Meeting Minutes

Automatically transcribe French meeting recordings into written records

High accuracy with standardized WER 11.1406

French Media Subtitle Generation

Automatically generate subtitles for French video content

🚀 French Medium Whisper

This model is a fine - tuned version of [openai/whisper - medium](https://huggingface.co/openai/whisper - medium) on the common_voice_11_0 dataset, achieving excellent results in French speech recognition.

🚀 Quick Start

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 (without normalization): 15.8969
Wer (with normalization): 11.1406

✨ Features

Blog post

All information about this model in this blog post: [Speech - to - Text & IA | Transcreva qualquer áudio para o português com o Whisper (OpenAI)... sem nenhum custo!](https://medium.com/@pierre_guillou/speech - to - text - ia - transcreva - qualquer - %C3%A1udio - para - o - português - com - o - whisper - openai - sem - ad0c17384681).

New SOTA

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

As this test dataset is similar to the [Common Voice 11.0](https://huggingface.co/datasets/mozilla - foundation/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](https://huggingface.co/openai/whisper - medium) model at transcribing audios French in 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)

🔧 Technical Details

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

📄 License

This model is licensed under the Apache 2.0 license.

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご