Abstract:

This work presents the development of an AI-driven web-based application for real-time recognition and learning of the American Sign Language (ASL) alphabet. The system integrates a Convolutional Neural Network (CNN) model trained on a dataset comprising over 80,000 labeled images of ASL hand gestures, in an interactive educational platform. The web interface captures live video from the user’s camera, processes the hand region in real time, and provides instant visual feedback by predicting the corresponding ASL letter. This interactive environment promotes independent, self-paced learning and enables users to assess their performance dynamically. The platform’s architecture combines deep learning and web technologies to deliver an accessible, browser-based tool that supports inclusive communication and education. Beyond its pedagogical value, results demonstrate efficient real-time inference with negligible latency, underscoring the potential of AI to support inclusive education and accessibility.

Keywords:

American Sign Language, Deep Learning, Computer Vision, Accessible Education, Real-Time Recognition.