MUMBAI, India, Sept. 28 -- Intellectual Property India has published a patent application (202641112208 A) filed by Lingesh N; Mohammed Thameem Sulthan M; and Sam V George on September 18, 2026, for Generative System For Real-Time Emotion-Driven, Lip-Synchronized Conversational Chat Avatars With Facial Animation.

Inventor includes Sam V George.

The application for the patent was published on September 25, 2026, under issue no. 39/2026.

Abstract: An intelligent computer-implemented chatbot system is disclosed, capable of receiving a user's query in either textual or spoken voice form, and generating, in response, a coordinated output comprising a textual answer, a spoken voice answer, and a live three-dimensional avatar that visually and expressively delivers the same response in real time. The system generates the avatar's response using either a retrieval-augmented answer-generation subsystem grounded in a stored knowledge base, or a live web-search-grounded subsystem that draws upon current information available on the web, both of which feed into a common downstream response-delivery pipeline such that the manner in which the answer is generated does not alter how it is spoken or animated. As the spoken voice answer is delivered, the avatar's mouth, lip, and jaw movements are generated in accurate, real-time correspondence to the audible speech, such that the avatar visibly appears to be speaking the actual words being heard. Concurrently, the avatar's facial expression is animated to reflect an emotional quality consistent with the content and tone of the response at any given moment, such that the avatar's face conveys both the literal spoken content and its emotional character simultaneously, in real time, producing the effect of a live, expressive presence explaining or conversing with the user. In certain embodiments, the avatar further exhibits natural, lifelike facial behavior, such as blinking and subtle head movement, during pauses in speech, so as to maintain a continuously living presence throughout the interaction. The disclosed system is adaptable across domains and use cases, and is suitable for virtual teaching assistants, customer support agents, informational kiosks, digital-human interfaces, and other real-time conversational-agent applications.

Disclaimer: Curated by HT Syndication.