REAL-TIME BIDIRECTIONAL TRANSLATION BETWEEN SIGN LANGUAGE AND VOICE USING GESTURE RECOGNITION

Abstract

Sign language is a major source of communication among the deaf or speech-challenged people in society; nevertheless, they do not interact with the rest of society because most people do not know how to communicate using sign language. To address this communication gap, this paper proposes a gesture-based translation system that enables bilateral communication between sign language gestures and spoken language. The suggested system removes the discriminative characteristics of the hand gestures and transforms them into the audio response, as well as the speech input to gesture output. The system shows good performance with a gesture-to-speech conversion of one second and a speech-to-gesture conversion in seven seconds. The outcomes of the experiments suggest that the suggested solution can become a useful assistive means of communication to be used by hearing- and speech-impaired people.