GMM versus AR-Vector Models for Text Independent Speaker Verification
Charles B. de Lima, Abraham Alcaim, José A. Apolinário · Anais do 2002 International Telecommunications Symposium · 2002
This paper presents a performance evaluation of two classification systems for text independent speaker verification: the Gaussian Mixture Model (GMM) and the AR-Vector Model.For the GMM, , , and Gaussians are evaluated.On the other hand, an order model with the Itakura symmetric distance was used for the AR-Vector.Both classification systems presented no errors when training and testing times were not smaller than s and s, respectively.Using s as the test time, the most accurate classification systems errors were between and %.With s test, the errors presented by the GMM were around to % whereas those for the AR-Vector were above %.However, the best results using 10s as testing and training times were obtained with the AR-Vector, with errors around %.