Speaker recognition using Radial Basis Function neural networks

A text-dependent speaker recognition system base on Radial Basis Function (RBF) neural network is presented. A two-stage recognition approach is proposed, in which the speaker-cohort model and the gender model are integrated to give the decision. The speaker recognition system has been evaluated in terms of both speaker verification and closed-set speaker identification, The results clearly indicate that the two-stage procedure is able to improve the overall performance of the speaker recognition system.