Intelligent Sign Language Translation Solution for inclusive hearing

Communication barriers for people with hearing and speech impairments pose challenges in education, healthcare, and everyday social interaction. This project, SignSpeak: Real-Time Sign Language Translator, tackles this issue by creating a gesture recognition system that translates sign language into text and speech directly in the browser. The system uses ml5.js along with MobileNet and KNN Classifier for machine learning in the browser. This setup allows for lightweight deployment without needing backend support. Key features include real-time gesture training, prediction, and classification. Users can save and load trained models in JSON format for reuse and offline access. The design prioritizes user-friendliness and accessibility, offering tutorials, gesture-to-text, and gesture-to-speech modules. A gesture dictionary helps users manage, retrain, and review their gestures effectively. The system was tested with various users, including members of the deaf community, and was refined based on their feedback. By providing an affordable, portable, and inclusive communication tool, this project helps connect sign language users with non-signers, promoting accessibility and social inclusion in real-life situations.