MUMBAI, India, June 22 -- Intellectual Property India has published a patent application (202621050120 A) filed by Sage University on April 20, 2026, for Apparatus And Method For Recognizing Speech In Plurality Of Languages.
Inventors include Dr. Prashant Jain; Ms. Sakshi Agrawal; Prof. Aishwarya Patidar; Veerandra Prajapati; and Dhruv Gupta.
The application for the patent was published on June 12, 2026, under issue no. 24/2026.
Abstract: ABSTRACT The present disclosure relates to speech recognition across multiple languages. The apparatus (102) includes a handheld enclosure (104) designed for single-hand operation, featuring an acoustic isolation chamber (106) with a directional port (108) that passively attenuates ambient noise and pre-conditions speech signals before electrical conversion. A multi-microphone array (110) positioned within the acoustic isolation chamber (106) generates digital audio signals from the pre- conditioned speech signals. A hardware processor (112), coupled to the multi-microphone array (110) and a storage medium (114) housed within the handheld enclosure (104), applies directional audio processing to produce noise-suppressed audio signals. The hardware processor (112) performs offline speech decoding using acoustic model data and language model data stored on the storage medium (114), generating recognized text in one of the plurality of languages. The recognized text is output as a text record on a display (116) connected to the hardware processor (112). FIG. 1
Disclaimer: Curated by HT Syndication.
