Sign Language Conversion into Text Using Machine Learning and CNN

The Sign Language Recognition project focuses on developing an efficient system to interpret and comprehend sign language gestures, crucial for the communication of the deaf and hearing-impaired community. By leveraging computer vision and machine learning techniques, the project aims to capture and analyze hand movements through Google's Python library, Media Pipe, and OpenCV. The process involves extracting joint coordinates from images, mapping them to gestures in American Sign Language (ASL), and translating these into textual language using a feed-forward neural network. This innovative approach utilizes key points rather than high-resolution images, significantly reducing the training dataset's size and overcoming variations in background and lighting conditions. By storing coordinate points in a CSV file, the system enhances space efficiency while addressing the challenges of gesture classification. Successful implementation holds promise in fostering inclusivity and accessibility, empowering individuals with hearing impairments to actively engage in education, employment, and social interactions within diverse aspects of everyday life.