<![CDATA[Advances in internet-based communication technology have driven the widespread use of video conferencing across education, industry, and public services. Although offering real-time audio and visual communication, existing systems do not fully support accessibility, particularly for deaf participants facing barriers in receiving verbal information. Integrating a speech-to-text (STT) feature offers a practical solution. Previous research implemented a WebRTC-based video conferencing application focusing on screen sharing and recording, but lacked a real-time STT module. Therefore, this study aims to develop a WebRTC-based web video conferencing system utilizing the Web Speech API to generate automated conversation transcriptions in supported languages. The system was developed using the Waterfall method, covering analysis, design, implementation, and testing phases. System design was modeled using Unified Modeling Language (UML) with use case and class diagrams, alongside mid-fidelity wireframes. The application was implemented using HTML, CSS, JavaScript, Node.js, and supporting libraries including Express, EJS, PeerJS, Socket.IO, UUID, and Nodemon. Following hosting deployment, system testing was conducted using black-box testing. Test results demonstrate that all core features function properly, with STT successfully converting audio into interactive, multilingual text. Both host and participants can view real-time conversation transcriptions directly on the conference screen. This system is expected to enhance communication inclusivity and accessibility, especially for deaf users.]]>
Copyrights © 2026
