Wav2vec2 Large Xlsr 53 French

🚀 Fine-tuned XLSR-53 large model for speech recognition in French

This fine-tuned model is based on facebook/wav2vec2-large-xlsr-53 and trained on French data from Common Voice 6.1. It's designed for accurate French speech recognition.

🚀 Quick Start

This model is a fine-tuned version of facebook/wav2vec2-large-xlsr-53 on French, using the train and validation splits of Common Voice 6.1. When using this model, ensure that your speech input is sampled at 16kHz.

This model has been fine-tuned thanks to the GPU credits generously given by the OVHcloud :)

The script used for training can be found here: https://github.com/jonatasgrosman/wav2vec2-sprint

✨ Features

• Audio Processing: Specialized for French speech recognition.
• Fine-tuned Model: Based on a pre - trained large model and fine - tuned on French data.
• Multiple Metrics: Evaluated using WER (Word Error Rate) and CER (Character Error Rate).

📦 Installation

No specific installation steps are provided in the original document.

💻 Usage Examples

Basic Usage

Using the HuggingSound library:

from huggingsound import SpeechRecognitionModel

model = SpeechRecognitionModel("jonatasgrosman/wav2vec2-large-xlsr-53-french")
audio_paths = ["/path/to/file.mp3", "/path/to/another_file.wav"]

transcriptions = model.transcribe(audio_paths)

Advanced Usage

Writing your own inference script:

import torch
import librosa
from datasets import load_dataset
from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor

LANG_ID = "fr"
MODEL_ID = "jonatasgrosman/wav2vec2-large-xlsr-53-french"
SAMPLES = 10

test_dataset = load_dataset("common_voice", LANG_ID, split=f"test[:{SAMPLES}]")

processor = Wav2Vec2Processor.from_pretrained(MODEL_ID)
model = Wav2Vec2ForCTC.from_pretrained(MODEL_ID)

# Preprocessing the datasets.
# We need to read the audio files as arrays
def speech_file_to_array_fn(batch):
    speech_array, sampling_rate = librosa.load(batch["path"], sr=16_000)
    batch["speech"] = speech_array
    batch["sentence"] = batch["sentence"].upper()
    return batch

test_dataset = test_dataset.map(speech_file_to_array_fn)
inputs = processor(test_dataset["speech"], sampling_rate=16_000, return_tensors="pt", padding=True)

with torch.no_grad():
    logits = model(inputs.input_values, attention_mask=inputs.attention_mask).logits

predicted_ids = torch.argmax(logits, dim=-1)
predicted_sentences = processor.batch_decode(predicted_ids)

for i, predicted_sentence in enumerate(predicted_sentences):
    print("-" * 100)
    print("Reference:", test_dataset[i]["sentence"])
    print("Prediction:", predicted_sentence)

Usage Results

📚 Documentation

Evaluation

1. To evaluate on mozilla-foundation/common_voice_6_0 with split test

python eval.py --model_id jonatasgrosman/wav2vec2-large-xlsr-53-french --dataset mozilla-foundation/common_voice_6_0 --config fr --split test

2. To evaluate on speech-recognition-community-v2/dev_data

python eval.py --model_id jonatasgrosman/wav2vec2-large-xlsr-53-french --dataset speech-recognition-community-v2/dev_data --config fr --split validation --chunk_length_s 5.0 --stride_length_s 1.0

Model Index

• Name: XLSR Wav2Vec2 French by Jonatas Grosman
• Results:
  • Task: Automatic Speech Recognition
  • Datasets:
    • Common Voice fr:
      • Metrics:
        • Test WER: 17.65
        • Test CER: 4.89
        • Test WER (+LM): 13.59
        • Test CER (+LM): 3.91
    • Robust Speech Event - Dev Data:
      • Metrics:
        • Dev WER: 34.35
        • Dev CER: 14.09
        • Dev WER (+LM): 24.72
        • Dev CER (+LM): 12.33

Other Information

📄 License

This project is licensed under the Apache-2.0 license.

📖 Citation

If you want to cite this model you can use this:

@misc{grosman2021xlsr53-large-french,
  title={Fine-tuned {XLSR}-53 large model for speech recognition in {F}rench},
  author={Grosman, Jonatas},
  howpublished={\url{https://huggingface.co/jonatasgrosman/wav2vec2-large-xlsr-53-french}},
  year={2021}
}