Bridging Communication Gaps: A Low-Cost, Real-Time Sign Language Recognition Platform

The communication gap between deaf and hearing individuals remains a persistent challenge, limiting participation in daily life and access to essential services. The difficulty of learning sign language and the lack of dedicated tools for translating between signed and spoken languages create barriers to full inclusion in social, educational, and healthcare settings. This paper presents a low-cost, real-time mobile application to facilitate smooth, natural bidirectional communication between spoken and signed language during video calls. The platform incorporates two dedicated translation pipelines: one that converts user gestures captured by the smartphone camera into synthesized speech, and another that transforms spoken input into animated sign language. The mobile application is powered by a backend system built on Node.js and Mediasoup for media management, with Python microservices responsible for running the translation models. Experimental results confirm that the proposed solution achieves real-time responsiveness and translation accuracy, offering a practical and scalable tool to support inclusive communication between deaf and hearing individuals.