New background modeling for speaker verification
Dat Thanh Tran · 2004
A new background speaker modelling method is presented in this paper for text-independent speaker verification using Gaussian mixture models. This method does not require speech databases of other speakers to build background speaker models. A background model can be built directly from the same claimed speaker's database and has a smaller number of Gaussian mixtures compared to the claimed speaker model. Experiments performed on the YOHO database showed a better result for speaker verification using the 64-mixture claimed speaker model and 16-mixture background model compared to current background model set methods using five closest background models.