Design and Creation of an App for Text and Voice Recognition System of Sign Language Using Deep Learning Technique - Sign Aloud

This is a project proposal for a Sign Language Recognition System – Sign Aloud. In non-verbal communication hand gesture is very important. Speech impairment is a disability of an individual to communicate using speech and hearing. Sign Language is the only solution for these kinds of people. There exists a challenge for non-signer to communicate with signer. Various Sign Language system had been developed by many developers but they are neither cost effective nor flexible for the end users. The major focus of this project is to create a deep learning-based android application that facilitate sign language translation to text and speech thereby aiding communication with general public. We use a custom CNN (Convolution Neural Network) EfficientNetB0 for recognizing the sign from a live video frame. This system can help in public speaking and students with speech impairment.