Video Chat using WebRTC with In-built Sign Language Detection

Many research efforts into video conferencing technology have focused on audio-only interactions. Text chat, on the other hand, has many uses and is often used in conjunction with the audio channel in multimedia conferencing systems because of the many benefits it offers to users in a variety of contexts. Video chat on the other hand has been providing participants to participate not only through voice but also through visual clues, creating a richer and more immersive communication experience. With the development of technology, video chat has transformed how people cooperate and communicate across distances by becoming more accessible and user-friendly. This study details a prototype program that uses video chat with in-built sign language identification to help the hearing-impaired and the speech-impaired. The prototype is constructed from WebRTC to provide realtime communication for the application, on top of which the object detection model from Tensor Flow is integrated to provide hand sign detection and recognition for the user. We focus our testing on the bandwidth and sign language detection needs of customers in order to provide improved support for video chat. We then talk about how this evaluation's findings can inform the evolution of video chatting software.