AI Role in Sign Language Interpretation

Yashraj P.  Chavan; Paras P.  Wala; Nihal S.  Gavandi; Piyush N.  Patil; Babeetta  Bbhagat

Abstract

People who are mute or deaf encounter communication challenges when interacting with others. It can be difficult for individuals with similar conditions to effectively convey their thoughts, especially when not everyone understands sign language. This article aims to create a Data Acquisition and Control (DAC) system capable of translating sign language into written text that can be comprehended by a wider audience. This system is referred to as the “Sign Language Translator and Gesture Recognition”. We have designed an intelligent web interface that captures hand gestures and translates them into readable text. The content within this textbook can be wirelessly transmitted to a smartphone or displayed on an embedded TV screen. It is evident from the experimental outcomes that affordable sensors can capture gestures, measuring finger positions and their movements. The current system's performance demonstrates its ability to accurately interpret 20 out of 26 letters, achieving a recognition accuracy of 96%. Sign languages rely on hand and finger shapes, motion, body language, and facial expressions to convey meanings, and they vary from one country to another. As an example, Irish deaf individuals use Irish Sign Language, while in India, people use Indian Sign Language. Different sign languages possess their distinct alphabets, semantics, and vocabularies. Variations in dialects among regions are prevalent, even within a single nation. Variations in some of the signals are employed by various creative groups.

Abstract

Keywords