MUMBAI, India, Nov. 2 -- Intellectual Property India has published a patent application (202611109056 A) filed by Manipal University Jaipur on September 11, 2026, for An Embedded Multi-Layer Voice Biometric Authentication System With Real-Time Anti-Spoof Harmonic Analysis To Prevent Proxy.
Inventors include Dr. Mohit Kushwaha; Shreyanshu Shekhar Shukla; and Puneet Sharma.
The application for the patent was published on October 30, 2026, under issue no. 44/2026.
Abstract: The present invention relates to a biometric identification system based on a voice for identifying a person. The system comprises a voice sensor module configured with MEMS microphone to capture the user’s voice with a signal to noise ratio greater than 60dB; an analog Front-End Circuit having an operational amplifier type of pre-amplifier band-pass filter (Active type) 300 Hz–3.4 kHz; a processing unit microcontroller comprising an ESP32, STM32 or embedded computing board configured to execute voice signal processing and artificial-intelligence-based voice authentication; a flash memory to store voice templates; an encrypted local database; a communication module comprising Wi-Fi, GSM and Bluetooth configured to transmit attendance information; and a software module comprising a signal preprocessing algorithm, a feature extraction engine and a pattern matching engine configured to preprocess the captured voice signal, extract voice features, compare the extracted voice features with a stored voice template, authenticate a user based on the comparison, and record attendance upon successful authentication. The integration of liveness detection in attendance devices improves real-time authentication, anti-proxy robustness, and security reliability improvements.
Disclaimer: Curated by HT Syndication.
