A novel alternative hypothesis characterization using kernel classifiers for LLR-Based speaker verification

Authors:
Yi-Hsiang Chao;Hsin-Min Wang;Ruei-Chuan Chang
Affiliations:
Institute of Information Science, Academia Sinica, Taipei;Institute of Information Science, Academia Sinica, Taipei;Institute of Information Science, Academia Sinica, Taipei
Venue:
ISCSLP'06 Proceedings of the 5th international conference on Chinese Spoken Language Processing
Year:
2006

Citing 3
Cited 0

Speaker identification and verification using Gaussian mixture speaker models

Speech Communication
A Tutorial on Support Vector Machines for Pattern Recognition

Data Mining and Knowledge Discovery
Pattern Classification (2nd Edition)

Pattern Classification (2nd Edition)

Quantified Score

Hi-index	0.00

Visualization

Abstract

In a log-likelihood ratio (LLR)-based speaker verification system, the alternative hypothesis is usually ill-defined and hard to characterize a priori, since it should cover the space of all possible impostors. In this paper, we propose a new LLR measure in an attempt to characterize the alternative hypothesis in a more effective and robust way than conventional methods. This LLR measure can be further formulated as a non-linear discriminant classifier and solved by kernel-based techniques, such as the Kernel Fisher Discriminant (KFD) and Support Vector Machine (SVM). The results of experiments on two speaker verification tasks show that the proposed methods outperform classical LLR-based approaches.