A REAL TIME INDIAN SIGN LANGUAGE RECOGNITION USING TENSORFLOW

Abstract

Communication is the exchange of information, ideas, or emotions, typically through spoken or written language. However, for individuals who are deaf or mute, traditional communication methods may not be effective. Instead, they rely on sign language—a visual form of communication using gestures and movements. Unfortunately, many people are unfamiliar with sign language, creating a barrier between those who use it and those who do not. Machine learning offers a promising solution to this challenge. By training a model to recognize and translate sign language gestures into spoken or written language, we can bridge this communication gap. This study proposes a real-time Sign Language Recognition (SLR) system using transfer learning with TensorFlow. Our approach involves capturing Indian Sign Language (ISL) gestures via a webcam and continuously training a deep learning model for accurate, real-time recognition. To enhance usability, we integrate a text-to-audio translation feature, converting recognized gestures into spoken language. This functionality allows individuals proficient in sign language to communicate seamlessly with those who are not, fostering inclusivity and accessibility. By leveraging machine learning, our system aims to break communication barriers and create a more inclusive society.