Welcome to the LIP READING-AI repository! This project focuses on developing an artificial intelligence system capable of reading lips from video inputs. Leveraging state-of-the-art machine learning techniques and deep learning models, our system aims to improve communication accessibility, particularly for the hearing-impaired community.
- Deep Learning Models: Utilizes advanced neural networks for accurate lip reading.
- Video Processing: Efficient video preprocessing and feature extraction.
- Pre-trained Models: Includes pre-trained models for immediate use.
- Custom Training: Provides tools for training models on custom datasets.
- Extensive Documentation: Comprehensive guides and tutorials for setup, usage, and customization.
- Python 3.7+
- TensorFlow 2.x
- OpenCV
- Other dependencies listed in requirements.txt
- You can download the dataset and the pretrained model from here 👇
- Dataset Link: https://drive.google.com/drive/folders/1zPIXGsSnC7NTh4E40a2QwdpNa-1FRn3A?usp=sharing
- Pre-Trained Model: https://drive.google.com/file/d/1u7ClC7z99SC4X9azEGG4zZhJkpkoYtlY/view?usp=sharing
Clone the repository and install the required dependencies:
In bash:
- Copy code -> git clone https://github.com/sindhujagopu/LIP-READING-AI.git
- cd LIP-READING-AI
- pip install -r requirements.txt
- Facilitates communication in masked situations by focusing on visible lip movements for speech interpretation, helping both the hearing-impaired and deaf.
- Augments surveillance systems by interpreting silent video footage to understand conversations without audio.
- Aids in monitoring sensitive areas by reading lips from video feeds where audio is unclear or unavailable.
- Enhances virtual assistants and customer service systems by incorporating lip-reading capabilities for better user interaction, especially for the deaf.
- Supports language learners and individuals in speech therapy, including deaf people, by demonstrating the formation of words on lips.
- Assists law enforcement in decoding silent video footage to gather crucial evidence from lip movements.
- Improves user experience in media consumption by providing subtitles based on lip reading in noisy environments.
- and a lot more ...
We would like to thank the contributors and the open-source community for their valuable work and support.