MUMBAI, India, Feb. 6 -- Intellectual Property India has published a patent application (202511133032 A) filed by Manipal University, Jaipur, Rajasthan, on Dec. 29, 2025, for 'a system and method for real-time speaker recognition using hybrid mfcc-lpc feature extraction.'
Inventor(s) include Mr. Ayush Saxena; and Dr. Peeyush Garg.
The application for the patent was published on Feb. 6, under issue no. 06/2026.
According to the abstract released by the Intellectual Property India: "The present invention relates to a system and method for real-time speaker recognition using hybrid MFCC-LPC feature extraction. The system comprises an audio input unit that receives configured to speech signals; a feature extraction module configured to combines the Mel Frequency Cepstral Coefficient (MFCC) and Linear Predictive Coding (LPC) parameters in order to create a hybrid feature vector; a classification module trained to classify or differentiate between the speaker profiles as authorised or unauthorised; and a decision-making module operating under real-time constraints using machine learning technology; wherein he decision-making module processes the extracted hybrid feature vectors and produces a binary classification result. Further, an embedded controller is set up to initiate a physical output in response to the classification outcome and is connected to the decision-making module through a processing and control unit. The system offers deterministic response times that are appropriate for embedded and low power applications and runs independently of external servers. For the purpose of security, automation and voice-controlled interface applications, the proposed architecture allows for a smaller, effective and lower latency speaker verification."
Disclaimer: Curated by HT Syndication.
