MUMBAI, India, May 29 -- Intellectual Property India has published a patent application (202621042666 A) filed by Sage University, Indore, Madhya Pradesh, on April 2, for 'apparatus and method for enhanced recognition of speech in indian regional languages.'
Inventor(s) include Dr. Prashant Jain; Ms. Sakshi Agrawal; Dr. Ati Jain; Dr. Piyush Moghe; Akash Dwivedi; and Anshuman Mishra.
The application for the patent was published on May 29, under issue no. 22/2026.
According to the abstract released by the Intellectual Property India: "The present disclosure relates to speech recognition in Indian regional languages. The apparatus (102) includes a linear microphone array (104) that detects acoustic pressure waves and generates analogue electrical signals. An acoustic pre-processing subsystem (106) performs analogue-domain spectral conditioning of the analogue electrical signals prior to analogue-to-digital conversion. A microcontroller unit (MCU) (108) receives the spectrally conditioned signals, extracts phoneme feature parameters for successive audio frames, and determines a phoneme category for each audio frame by traversing a binary decision tree using the extracted phoneme feature parameters. The MCU (108) outputs a sequence of phoneme identifiers corresponding to the determined phoneme categories to a language decoder, which generates recognized speech text. The apparatus (102) addresses challenges in accurately recognizing phonetically rich and diverse Indian regional languages through early analogue-domain spectral conditioning combined with binary decision tree-based phoneme classification."
Disclaimer: Curated by HT Syndication.
