signcall: bridging communication gaps in virtual meetings with sign language recognition

Sign language is a visual mode of communication that uses hand gestures and movements to convey meaning, serving as an essential communication tool for individuals with hearing or speech impairments. Despite its importance, many virtual platforms lack the ability to recognize and interpret sign language, creating significant barriers to inclusivity in digital communication. As virtual meetings become more integral to professional and personal communication, the need for inclusivity in these spaces has grown. Current meeting platforms often fail to accommodate users who rely on sign language, limiting their ability to engage fully in discussions. This project aims to address this gap by integrating real-time sign language recognition into video calling platforms, ensuring accessibility for all participants. The proposed system employs the Video Calling Vision Transformer (VCViT) to accurately recognize word-level hand gestures. The system captures live video streams from participants, focusing on hand gestures, and translates them into text or speech in real time. By utilizing advanced video processing techniques, gesture segmentation, and the VCViT's ability to model spatial relationships, the system achieves high recognition accuracy, adapting to different signing styles and environmental conditions. This project strives to create inclusive virtual meeting environments, allowing hearing-impaired individuals to actively participate in discussions. Through AI-driven solutions, it ensures seamless communication, fosters equity, and enhances digital collaboration.