AV HuBERT MuAViC Multilingual

🚀 Huggingface Implementation of AV-HuBERT on the MuAViC Dataset

This repository offers a Huggingface implementation of the AV-HuBERT (Audio-Visual Hidden Unit BERT) model, trained and tested on the MuAViC (Multilingual Audio-Visual Corpus) dataset. AV-HuBERT is a self - supervised model for audio - visual speech recognition, leveraging both audio and visual modalities for robust performance, especially in noisy environments.

🚀 Quick Start

This section provides a quick guide on how to get started with the AV - HuBERT implementation in this repository.

✨ Features

• Pre - trained Models: Access pre - trained AV - HuBERT models fine - tuned on the MuAViC dataset. The pre - trained models are exported from the MuAViC repository.
• Inference scripts: Easily set up pipelines using Huggingface’s interface.
• Data preprocessing scripts: These include normalizing frame rate, extracting lips and audio.

📦 Installation

To install the necessary dependencies and set up the environment, follow these steps:

git clone https://github.com/nguyenvulebinh/AV-HuBERT-S2S.git
cd AV-HuBERT-S2S
conda create -n avhuberts2s python=3.9
conda activate avhuberts2s
pip install -r requirements.txt

💻 Usage Examples

Basic Usage

Here is a simple example to run the inference code:

python run_example.py

Advanced Usage

The following Python code demonstrates a more detailed inference process:

from src.model.avhubert2text import AV2TextForConditionalGeneration
from src.dataset.load_data import load_feature
from transformers import Speech2TextTokenizer
import torch

if __name__ == "__main__":
    # Choose language to run example
    AVAILABEL_LANGUAGES = ["ar", "de", "el", "en", "es", "fr", "it", "pt", "ru", "multilingual"]
    language = "ru"
    assert language in AVAILABEL_LANGUAGES, f"Language {language} is not available, please choose one of {AVAILABEL_LANGUAGES}"
    
    
    # Load model and tokenizer
    model_name_or_path = f"nguyenvulebinh/AV-HuBERT-MuAViC-{language}"
    model = AV2TextForConditionalGeneration.from_pretrained(model_name_or_path, cache_dir='./model-bin')
    tokenizer = Speech2TextTokenizer.from_pretrained(model_name_or_path, cache_dir='./model-bin')
    
    model = model.cuda().eval()
    
    # Load example video and audio
    video_example = f"./example/video_processed/{language}_lip_movement.mp4"
    audio_example = f"./example/video_processed/{language}_audio.wav"
    if not os.path.exists(video_example) or not os.path.exists(audio_example):
        print(f"WARNING: Example video and audio for {language} is not available english will be used instead")
        video_example = f"./example/video_processed/en_lip_movement.mp4"
        audio_example = f"./example/video_processed/en_audio.wav"
    
    # Load and process example
    sample = load_feature(
        video_example,
        audio_example
    )
    
    audio_feats = sample['audio_source'].cuda()
    video_feats = sample['video_source'].cuda()
    attention_mask = torch.BoolTensor(audio_feats.size(0), audio_feats.size(-1)).fill_(False).cuda()
    
    # Generate text
    output = model.generate(
        audio_feats,
        attention_mask=attention_mask,
        video=video_feats,
        max_length=1024,
    )

    print(tokenizer.batch_decode(output, skip_special_tokens=True))

Data Preprocessing

The following commands are used for data preprocessing:

mkdir model-bin
cd model-bin
wget https://huggingface.co/nguyenvulebinh/AV-HuBERT/resolve/main/20words_mean_face.npy .
wget https://huggingface.co/nguyenvulebinh/AV-HuBERT/resolve/main/shape_predictor_68_face_landmarks.dat .

# raw video only support 4:3 ratio now
cp raw_video.mp4 ./example/ 

python src/dataset/video_to_audio_lips.py

📚 Documentation

Pretrained AVSR Model

The following table lists the available pre - trained AVSR models for different languages:

📄 License

This project is licensed under the CC - BY - NC 4.0 license.

📚 Citation

If you use this work, please cite the following paper:

@article{anwar2023muavic,
  title={MuAViC: A Multilingual Audio-Visual Corpus for Robust Speech Recognition and Robust Speech-to-Text Translation},
  author={Anwar, Mohamed and Shi, Bowen and Goswami, Vedanuj and Hsu, Wei-Ning and Pino, Juan and Wang, Changhan},
  journal={arXiv preprint arXiv:2303.00628},
  year={2023}
}

Acknowledgments

• AV - HuBERT: A significant portion of the codebase in this repository has been adapted from the original AV - HuBERT implementation.
• MuAViC Repository: We gratefully acknowledge the creators of the MuAViC dataset and repository for providing the pre - trained models used in this project.