Comparison of Selected Classification Methods in Automatic Speaker Identification

Abstract

This paper presents performance comparison of three different classifiers applied in Automatic SpeakeR Identification: Gaussian Mixture Model (GMM), k Nearest Neighbor algorithm (kNN) and Support Vector Machines (SVM). Each classifier represents different approach to the classification procedure. Mel Frequency Cepstral Coefficients (MFCC) were used as feature vectors in the experiment. Classification precision for each classifier was evaluated on frame and recording level. Experiments were conducted over dataset MobilDat-SK, which was recorded in mobile telecommunication network. Experiment shows promising results for SVM classifier.