MUMBAI, India, May 1 -- Intellectual Property India has published a patent application (202621024588 A) filed by Swati Atul Joshi; Vikrant Kishorsingh Thakur; Vira Amit Shah; Satyam Kumar; Utkarsh Trichal; Chinmay Mahesh Yadav; and Nihar Ravi Thalkar, Pune, Maharashtra, on March 2, for 'voice-biometric authentication system with multi-factor keypad security and adaptive digital filtering.'
Inventor(s) include Swati Atul Joshi; Vikrant Kishorsingh Thakur; Vira Amit Shah; Satyam Kumar; Utkarsh Trichal; Chinmay Mahesh Yadav; and Nihar Ravi Thalkar.
The application for the patent was published on May 1, under issue no. 18/2026.
According to the abstract released by the Intellectual Property India: "An innovative method and system successfully utilize biometrics as a means of authenticating individuals via voice recognition. Authentication systems typically only offer one form of authentication, which can result in unreliable or inconsistent results when used under adverse conditions (e.g., noise interference). As a result of this, there can be high rates of "false rejection," as well as instances of potential security breaches. The invention is designed to address these issues by utilizing a tiered security approach to implement the authentication process. First, users must enter a unique identifier and PIN via keypad before they can access their biometric profile. Next, an embedded database maintains access to the multiple biometric profiles of each user, allowing for the quick retrieval of voice template data for each verified ID. Finally, an ESP32 microprocessor analyzes real-time SNR and Frequency Spectra data in order to evaluate complex background sounds that could interfere with the recognition of individual voices. Based on this analysis, the system dynamically selects and applies appropriate software-based digital filters, such as High-Pass, Low-Pass, or Band-Pass, to isolate the user's voice. Finally, an interactive user feedback loop via an OLED display provides real-time instructions, such as "Too Noisy - Please Repeat," if the signal quality remains insufficient for reliable matching. This novel combination of multi-factor validation, adaptive filtering, and user guidance ensures that only clean, high-quality signals from authorized users are processed, significantly enhancing authentication reliability and security in real-world conditions."
Disclaimer: Curated by HT Syndication.
