Hierarchical Speaker Verification Based on PCA and Kernel Fisher Discriminant

Abstract

In this paper, a novel hierarchical speaker verification method based on PCA classifier and kernel fisher discriminant (KFD) classifier was proposed. Firstly, we gota coarse decision by a fast scan all registered speakers using PCA classifier to find R possible target speakers, and then KFD classifier was used to make final decision. PCA also has another advantage: reduction of the feature vectors dimensions, and the noise is removed from speech simultaneity. So, it can reduce the computational complexity and improve the performance of speaker verification. KFD classifier achieved high verification accuracy since it utilized all training samples. The experiment results showed that the proposed method could improve recognition accuracy of system remarkably and the system has better robustness by comparing with the traditional speaker verification method.