Real-Time Sign Language Recognition System

Hearing and speech impairments affect more than 36 million of the world's population. A sign-language recognition system is a crucial step towards the enhancement of communication among visually impaired people. The main idea behind this project is to create a real-time sign language recognition deep learning model which can work on pre-trained image and video dataset and give us the output in real-time. Our system would take the input as an ASL alphabet and save it and then show the string in text or audio format according to the user's choice. It involves accurate extraction of hand gestures using appropriate sensing devices in our model for the smooth communication between the normal and visually-impaired people. This project can serve as a perfect service to the educational industry.