CPL - Chalmers Publication Library
| Utbildning | Forskning | Styrkeområden | Om Chalmers | In English In English Ej inloggad.

Sample iterative likelihood maximization for speaker verification systems

Guillermo Garcia (Institutionen för signaler och system, Kommunikationssystem) ; Thomas Eriksson (Institutionen för signaler och system, Kommunikationssystem)
18th European Signal Processing Conference, EUSIPCO 2010; Aalborg; Denmark; 23 August 2010 through 27 August 2010 (22195491). p. 596-600. (2010)
[Konferensbidrag, refereegranskat]

Gaussian Mixture Models (GMMs) have been the dominant technique used for modeling in speaker recognition systems. Traditionally, the GMMs are trained using the Expectation Maximization (EM) algorithm and a large set of training samples. However, the convergence of the EM algorithm to a global maximum is conditioned on proper parameter initialization, a large enough training sample set, and several iterations over this training set. In this work, a Sample Iterative Likelihood Maximization (SILM) algorithm based on a stochastic descent gradient method is proposed. Simulation results showed that our algorithm can attain high loglikelihoods with fewer iterations in comparison to the EMalgorithm. A maximum of eight times faster convergence rate can be achieved in comparison with the EM algorithm.

Denna post skapades 2015-05-06. Senast ändrad 2016-02-01.
CPL Pubid: 216542


Institutioner (Chalmers)

Institutionen för signaler och system, Kommunikationssystem (1900-2017)



Chalmers infrastruktur