Title: Neuro-fuzzy-based biometric system using speech features

Authors: Anupam Shukla, Ritu Tiwari, Chandra Prakash Rathore

Addresses: Indian Institute of Information Technology and Management (IIITM), National Highway, Morena Link Road, Gwalior 474010, India. ' Indian Institute of Information Technology and Management (IIITM), National Highway, Morena Link Road, Gwalior 474010, India. ' Indian Institute of Information Technology and Management (IIITM), National Highway, Morena Link Road, Gwalior 474010, India

Abstract: Biometric identification is one of the most developing areas. In this paper, a biometric system is simulated using speech features, which identifies the speaker along with their gender and mental status. Work here is broadly classified into two parts, i.e., extraction of the speech features, namely Pitch, Amplitude, Number of Zero-Crossing (NZC), Average Power Spectral Density (PSD) content in the speech of informant and in the second part an adaptive neuro-fuzzy based simulation model has been developed for speaker identification along with their gender and mental status. The recognition score varies depending on different input and output Membership Functions (MFs).

Keywords: neuro-fuzzy biometrics; speech features; speaker identification; gender identification; mental status determination; pitch; amplitude; NZC; number of zero-crossings; PSD; power spectral density; average power spectral density; speaker recognition; speech feature extraction; simulation; neural networks; fuzzy logic.

DOI: 10.1504/IJBM.2010.035452

International Journal of Biometrics, 2010 Vol.2 No.4, pp.391 - 406

Published online: 30 Sep 2010 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article