• Raashid. A
• Dr Daphney Joann. J
• Dinesh Kumar. G
• Santhosh. V
• Mohammad Shamil

Abstract : Sharing ideas, feelings, and facts requires effective communication. However, there are major obstacles to inclusivity because virtual meeting systems like Zoom, Microsoft Teams, and Google Meet sometimes do not include accommodations for people with speech and hearing impairments. Deaf people can communicate through sign language, but non-signers still find it difficult to understand. Because of their poor accessibility and accuracy, current sign language recognition technologies are not well suited for smooth incorporation into virtual platforms. In order to improve communication between hearing and deaf participants in virtual meetings, this project presents an AI-driven solution. Real-time two-way communication is made possible by the system's use of deep learning innovations, specifically Temporal Convolutional Networks (TCNs). In order to improve communication between hearing and deaf participants in virtual meetings, this project presents an AI-driven solution. Real-time two-way communication is made possible by the system's use of deep learning innovations, specifically Temporal Convolutional Networks (TCNs). Its three main modules are the Speech Recognition and Synthesis Module (SRSM), which uses Hidden Markov Models to translate spoken language into text, the Sign Recognition Module (SRM), which uses TCN to interpret signs, and the Avatar Module (AM), which visually translates speech into corresponding signs. In order to ensure that non-signers can interact with sign language users in an engaging and efficient manner, the Avatar Module is crucial for visually displaying spoken language in sign language format. The technology, which was trained in Indian Sign Language, makes it easier for people who are deaf, mute, hard of hearing, visually handicapped, and non-signers to communicate with one other. It incorporates into well-known online meeting platforms via an intuitive web interface improves accessibility and engagement. In terms of promoting accessibility and inclusivity in online meeting settings, this method is a major breakthrough.

Sign Language Recognition, Temporal Convolutional Networks, Accessibility, Human-Computer Interaction

"AI Powered Accessibility for Enabling Effective Communication for Hearing and Speech Impaired in Virtual Platforms", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 5, page no.d348-d354, May-2025, Available :http://www.jetir.org/papers/JETIR2505379.pdf

"AI Powered Accessibility for Enabling Effective Communication for Hearing and Speech Impaired in Virtual Platforms", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 5, page no. ppd348-d354, May-2025, Available at : http://www.jetir.org/papers/JETIR2505379.pdf