Indian Sign Language Detection for Real-Time Translation using Machine Learning

Gestural Language used by deaf and mute communities to communicate by using hand gestures and body movements that rely on visual and spatial patterns known as sign languages. Communication is an important part of human relations, but people in the deaf community face serious problems due to lack of good interpreting equipment and resources. Research on Indian Sign Language (ISL) recognition through machine learning aims to bridge the communication gap between the deaf and hard-of-hearing people in India. The proposed model is based on Indian Sign Language Detection using Convolutional Neural Networks (CNN) and achieved 99.95% accuracy on (ISL) dataset. Accuracy, F1 score, and precision-recall are used as metrics to evaluate the model for improving communication among deaf and mutes. Mediapipe Object detection is used in the proposed system for hand motion detection. The research elaborates the Model architecture, preprocessing, and classification methodologies for enhancing communication in deaf and mute communities.