MUMBAI, India, Aug. 12 -- Intellectual Property India has published a patent application (202641091856 A) filed by Peri College Of Arts And Science on July 29, 2026, for Tamil Meet Ai: Real-Time Multilingual Speech Translation System For Smart Meetings.
Inventors include Dhivya R; Prathiba G; Prathiksha T; Ranjani Devi R; Reena K; and Hariharan K.
The application for the patent was published on August 07, 2026, under issue no. 32/2026.
Abstract: The present invention relates to an Artificial Intelligence (Al) based real-time multilingual speech translation system particularly aimed at enabling communication between the parties in different languages for an online meeting attended by government, organization or individual participants who speak the Tamil language in the online meeting. The existing translation services mostly cater to the dominant languages across the globe and the invention provides an intelligent meeting translation system that translates Tamil voice into multiple target languages and vice versa in minimal latency maintaining the meaning, technical terminologies and speaker inient of the conversation within the meeting. The system comprises of Automatic Speech Recognition (ASR), Natural Language Understanding (NLU), Neural Machine Translation (NMT), Context-Aware Translation Engine, Voice Cloning based Text-to-Speech (TTS), and adaptive learning modules. One of the innovations is the Meeting Context Memory Engine which adapts to the needs of the meeting, learning new words, acronyms, technical and speakers' terms as the meeting progresses to make the translation more accurate. The invention also includes an Al Confidence Analyzer to automatically identify translations that are not entirely certain and provide alternative versions for the listeners. Intelligent Terminology Repository supports domain-specific vocabulary updates for education, healthcare, engineering discussions, legal discussions and government meetings dynamically. The platform can be used as an independent application or as a plug-in that can be installed on MS Teams, Zoom, Google Meet, Webex and other major video communication services. The proposed system is different from the conventional one which works on sentence level translation, where semantic-level multilingual communication keeps the speaker's emotion, conversational flow and linguistic nuance. The invention greatly enhances access to and inclusiveness of Tamil, or other Indian regional language speakers in multilingual meetings, allowing them to participate equally to other participants in professional, academic, governmental and international collaborative settings.
Disclaimer: Curated by HT Syndication.
