• Bandlamudi Sri Tarak Ram
• Bojedla Jaswanth
• T Meena

Speech-impaired individuals are facing significant challenges in social interactions due to the lack of widespread knowledge of sign language among the general population. To address this issue, this research proposes a real-time communication system that leverages Graph Neural Networks (GNNs) to translate hand gestures into meaningful sentences, thereby bridging the communication gap between normal people and broader population effectively. The system captures live video of hand gestures using a webcam, pre-processes the images to enhance quality, and utilizes GNNs to classify graph-structured data, ensuring efficient and accurate gesture recognition. The recognized gestures are then converted into coherent sentences using Natural Language Generation (NLG), with the final out- put delivered as synthesized speech through a speaker. Unlike traditional hand gesture recognition systems that often translate gestures into isolated words, this system focuses on generating meaningful sentences, enabling natural and seamless communication. By integrating advanced techniques like GNNs for structured data processing and NLG for sentence formation, this proposed solution offers a significant improvement over existing methods, fostering inclusivity and enhancing the quality of life for speech-impaired individuals.

Hand gesture recognition, graph Neural Net- works, natural language generation, real-time communication, sign language translation, Indian Sign Language, assistive communication.

"Communication System For Speech-Impaired People using Graph Neural Networks", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 8, page no.c82-c86, August-2025, Available :http://www.jetir.org/papers/JETIR2508213.pdf

"Communication System For Speech-Impaired People using Graph Neural Networks", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 8, page no. ppc82-c86, August-2025, Available at : http://www.jetir.org/papers/JETIR2508213.pdf