Sign-Language-Detection

This project is dedicated to developing a Sign Language Recognition System (SLRS) that facilitates communication between the deaf and hard-of-hearing community and the hearing population. By employing advanced artificial intel- ligence, including a multi-headed convolutional neural network tailored for rec- ognizing Indian sign language letters , in which the system aims to translate sign language gestures into written text with high accuracy and reduced latency. We encountered several challenges, including over-fitting, class imbalance, choosing a proper dataset, and the complexities of gesture recognition in diverse en- vironments. Nevertheless, our preliminary results demonstrate promising im- provements, highlighting the system’s potential in educational, healthcare, and communicative contexts. Our ongoing efforts are focused on enhancing the tech- nical robustness of the system through iterative preprocessing, Incorporating real time detection using object detection API, and expanding past just In- dian sign language and using other languages such as American sign language. This work underscores the transformative potential of artificial intelligence in creating more inclusive communication tools for society.