A prototype lip-reading system recognizing English phonemes, designed to assist hearing-impaired individuals in situations where sound is unavailable. Trained on a subset of the GRID corpus (~34,000 short videos) inspired by the LipNet architecture.
No image found.
Demo day video
Tech stack
Python
TensorFlow
OpenCV
Streamlit
Dlib
Meet the team
Gabriel Gress
Micolas Ogawa
Sandhiya Tamilvanan
Hidekazu Takayoshi
See similar projects
