SPEAKER IDENTIFICATION IN NOISY ENVIRONMENT

Abstract

The human ear is a marvelous organ . Beyond our uniquely human abilitylto receive and decode spoken language , the ear supplies us with the ability to perform many diverse functions . These include , for example , localization of objects, I enjoyment of music , and the identification of people by their voices . Currently,

along with efforts to develop computer procedures that understand spoken messages, there is also considerable interest in developing procedures that identify people frem their voices .

The proposed system introduced in this thesis is an application of speaker recognition to identify people by their voices in both clean and noisy environment . In addition to identifying people by their voices , the system provides simulation, of human auditory nerve , human ear and their application in speaker identification .

The proposed system used acquisition subsystem , preprocessing and feature extraction subsystem , and a classifier is designed on the basis of dynamic time warping method .The system is implemented using a personal computer, and the developed programs works in close association with the MATLAP soft - ware package.

From the system investigation and performance tests that have been conducted in this work , we can draw out the following conclusions : 1-The PLP Features shows good performance over LPC , and cepstrum analysis. 2-The PLP , as a noise immune feature , gives considerable improvement in ' recognition accuracy over the commonly used features