Speaker Identification Using Gaussian Mixture Models

In this paper, the performance of Perceptual Linear Prediction (PLP) features has been compared with the performance of Linear Prediction Coefficient (LPC) features for speaker identification. Two classification techniques, Gaussian Mixture Models (GMM) and Vector Quantization (VQ) with Dynamic time wrapping (DTW) are used for classification of speakers based on their speech samples into respective classes. A database of fifty speakers, twenty one males and twenty nine females has been prepared in clean environment. The identification performance of PLP features is 3.6%better than LPC features with VQ and DTW classifier. PLP features have also shown 1.2% increment in identification performance over LPC features with GMM classifier. Keywords—Gaussian Mixture Models, Perceptual Linear Prediction, Linear Prediction Coefficient, Vector Quantization