Real-Time Sign Language Recognition and Translation with Glove and Mobile Integration
E Unnathi, Akkati Sreeja, Ravi Teja, Vinutha LV · 2024
This paper presents a system integrating real-time sign language detection with a smart glove for gesture-to-text and voice output. It employs a hybrid CNN and RNN algorithm for hand landmark detection, using MediaPipe for real-time processing. The CNN extracts features and patterns, which are then passed to an RNN for temporal analysis. The resulting text is converted to speech via Google Text-to-Speech. The smart glove, with two accelerometers, detects gestures and displays the results on an Arduino-connected LCD. Bluetooth HC05 and mobile device integration facilitate effective exchange of words for those with listening and speech impairments, offering a comprehensive solution for enhanced accessibility and interaction.