Speaker Recognition System using Gaussian Mixture Model

In this paper,features for text-independent speaker recognition has been evaluated. Speaker identification from a set of templates and analyzing speaker recognition rate by extracting several key features like Mel Frequency Cepstral Coefficients [MFCC] from the speech signals of those persons by using the process of feature extraction using MATLAB2013 .These features are effectively captured using feature matching technique like Gaussian Mixture Model [GMM] , with varying mixture components of mixture model and the analyzing its effect on recognition rate . Improve the speaker recognition rate by varying the input parameters of the classifier. The experiments are evaluated on TIMIT Database effectively for a speech signal sampled at 16kHz.