Intelligent Sign Language Recognition for Real-Time Text Conversion to Aid Speech and Hearing Impaired
Abhishek Tripathi, Divya Kiran, D Jaya Chandra, M Bharath Kumar, Arjun Ram, Shubham Anjankar, K Suthendran, Nishit Malviya · Procedia Computer Science · 2025
This research presents an innovative system for converting sign language into text using an Arduino UNO-IoT platform and a custom-designed glove equipped with finger flex sensors. The flex sensors detect the bending angles of the fingers, generating real-time data corresponding to different sign language gestures. These signals are processed by the Arduino to interpret gestures, which are then displayed as text on a 16x2 LCD screen. The system is designed to improve communication for the hearing and speech impaired, providing a low-cost, portable, and user-friendly solution for converting hand gestures into readable text. Technical findings show that the system achieves an average accuracy of 85% for simple gestures and 75% for complex gestures, with consistent performance across multiple users. Results were analyzed in terms of angular displacement, sensor resistance, and gesture type recognition. The system effectively bridges the communication gap between sign language users and the general public, offering enhanced accessibility through IoT integration.