Published Paper Details:

--Sign language serves as a crucial communication medium for the deaf and hard-of-hearing community. This research presents a real-time sign language recognition system that enhances accessibility through intuitive gesture-based inter? actions. The system consists of three key components: 1. Text-to-Sign Conversion Transforms input text into corre? sponding sign language images. 2. Gesture Recognition Predicts and displays recognized hand signs as text and speech output. 3. Smart Assistant Enables users to control system settings such as brightness and volume and launch applications like YouTube using predefned gestures. Unlike traditional machine learning-based approaches, our system leverages convolutional neural networks (CNN) along with rule-based gesture mapping and computer vision techniques for accurate recognition and execution of commands. The model is trained on American Sign Language (ASL) fnger spelling gestures, where hand position and orientation are used for training and testing. The captured images are preprocessed using flters and passed through a classifer to predict gesture classes before being fed into the CNN model for further refnement. Experimental results demonstrate high accuracy and effciency, providing a seamless and interactive experience for individuals relying on sign language for communication and accessibility.