MUMBAI, India, June 26 -- Intellectual Property India has published a patent application (202441099042 A) filed by Manipal Academy Of Higher Education on December 14, 2024, for A System And Method For Forensic Voice Recognition Using Acoustic Analysis Techniques.
Inventor includes Arjun M S.
The application for the patent was published on June 19, 2026, under issue no. 25/2026.
Abstract: The invention provides an advanced automated voice recognition system for forensic investigations, designed to accurately identify speakers based on language, dialect, age, emotional state, and environmental conditions. By integrating Mel Frequency Cepstral Coefficients (MFCC), Linear Predictive Coding (LPC), and a Gaussian model, the system enables reliable identification in noisy or noiseless environments. The system automatically records voice samples from crime scenes, analyzes them for speaker characteristics, and compares them with a forensic database for suspect matching. It incorporates advanced emotional state detection and acoustic analysis to ensure accuracy across various real-world conditions. This innovation offers an efficient, scalable solution for forensic investigations, providing automated, reliable voice identification for legal purposes and facilitating future criminal inquiries. (FIG. 2)
Disclaimer: Curated by HT Syndication.
