Indian Sign Language Recogniser with Text and Speech Translation

Sign Language is important to bridge the gap between communication for the deaf and hard-of-hearing community. Here, we work with ISL in the context of the Indian deaf community. However, one of the major challenges here lies in making the interaction of ISL users with non-sign language users smooth. The system introduces a strong vision-based Indian Sign Language Recogniser with real-time text and speech translation for better communication and inclusiveness. It is made up of a system that uses deep learning models constructed on a TensorFlow backend, which classifies static ISL gestures captured through real-time video streams. OpenCV is utilized for hand detection and feature extraction to allow for accurate and efficient hand gesture recognition. The system translates recognised gestures into their corresponding textual representations and into audible speech, allowing the user for thorough comprehension by non-ISL users. It offers a user-friendly interface, including functionality such as activation of recognition, textual display, and speech synthesis.