Sign Language Interpreter Using Machine Learning

Abstract

Communication is the key point for interaction and we humans use different languages through which we share our views and opinions. But for certain people i. e., the hearing impaired or the deaf and dumb they are not able communicate well because of their disability. According to World Health Organization (WHO) it is estimated that hearing loss affects approximately four hundred sixty-six million people globally with thirty-four millions of these being children. Sign language is one of the possibilities through which they can communicate. It is a coded language in which they use hand gestures. Communication with the rest of the world is very difficult for hearing-impaired people and hence they are isolated, in order to overcome these circumstances, we present a project that is helpful to such people. This project presents a solution with the use of Machine Learning that will identify hand gestures and translate it into speech/text and the vice versa that is it convert from speech/text to gesture. A webcam will be used to capture the region of interest i.e, recognize the hand motion as well as the gesture that the hand is indicating. Based on the recognized gestures, the recorded soundtrack will be played.