Text dependent speaker recognition using discrete stationary wavelet transform and PCA

摘要

In this paper we present an effective and robust method for speaker identification based on discrete stationary wavelet transform and principal component analysis techniques. The time invariant characteristic of SWT is particularly used in this paper for Speaker recognition. We have selected PCA as it is a core of modern data analysis. For classification purpose we have used Artificial Neural Network. One hundred and fifty samples are collected from 10 different people at various time intervals for the initial study. Data base is created from a spoken language Malayalam.