International Journal of Scientific Research and Engineering Development

People who communicate using hand gestures, especially those who are deaf or unable to speak, face difficulties in talking with others. Real-time translation of hand gestures, which can then be converted to speeches, may overcome communication difficulties between people using hand gestures and normal people. This project presents a Real-time translation of Dynamic hand gestures into synthesized speech using the spatiotemporal deep neural networks, The system that captures the video/Photo of continuous hand gestures that convert into the frames. The CNN to extract the visuals from frames and the LSTM to understand the movement of the gestures. Then the gestures are converted into Text and immediately transformed into real time Speech. The result shows that produces the clear gestures of spatiotemporal deep neural networks converting the Text into the real time speech, The project that can be useful for the individuals with speech and hearing disabilities and human–computer interaction and the real time communication systems using the spatiotemporal deep neural networks.

Sandeep Kumar Arudra, P.NagaVenkata Dileep, K.Rakesh, R.Venkata Sai Lokesh, P.Irfan Khan,"Real-Time Translation of Dynamic Hand Gestures Sequences into Synthesized Speech Using Spatiotemporal Deep Neural Networks" International Journal of Scientific Research and Engineering Development, V9(2): Page(818-821) Mar-Apr 2026. ISSN: 2581-7175. www.ijsred.com. Published by Scientific and Academic Research Publishing.