MUMBAI, India, Oct. 5 -- Intellectual Property India has published a patent application (202541028876 A) filed by Chatbucket on March 27, 2025, for Multilingual Text-To-Speech (tts) System And Method Of Generating Multilingual Speech Output.
Inventor includes Singavaram Sunil Kumar Singh.
The application for the patent was published on October 02, 2026, under issue no. 40/2026.
Abstract: Disclosed is a multilingual text-to-speech (TTS) system (100) includes a user device (102) configured to receive text or speech input from a user and a processing circuitry (104) that is communicatively coupled with the user device (102) and configured to segment, tokenize, and normalize the input text using language-specific linguistic rules, automatically identify language transitions within mixed-language text input sequences, employ a multilingual hierarchical language model to extract phonetic, grammatical, and tonal features per language, integrate phoneme-based tokenization and self-supervised acoustic mapping to preserve pronunciation and tone consistency across one or more languages, captures tone, pitch, rhythm, and emotional inflection from reference speech or user-defined parameters, and integrates neural vocoders to generate naturalistic speech output with low-latency performance. The present disclosure also relates to a method (200) of generating multilingual speech output
Disclaimer: Curated by HT Syndication.
