TEXTIFY – LIP READING TO TEXT

Deeksha Shetty K, Kshama D, Rakshitha, Swathi
A J Institute of Engineering and Technology, Mangalore

Mrs. Sharanya P S
(Project Guide)

Assistant Professor, Department of Information Science and Engineering,

 A J Institute of Engineering and Technology, Mangalore

Email: sharanya@ajiet.edu.in

Abstract-- Textify is the process of identifying spoken words by watching lip movement. Visual listening techniques include lip reading. The ability of Textify to extract text from streaming or recorded video may present opportunities for solving connectivity and practical problems. This entails attempting to decipher the speaker's meaning from the expression on their face. For the conversion, the deep learning approach can be used to analyze lip movement. The system that is being proposed in this project transforms video data into text data by lip-reading. Text conversion is the most effective way to speak to people who are deaf or dumb. The application will receive frames from a live or recorded video as input. Additionally, the application must track down and understand the feature patterns.

Keywords- lip reading, deep learning, convolutional neural network.