GMM-based Bhattacharyya kernel Fisher discriminant analysis for speaker recognition

Clearly, the linear discriminant classifier is not robust enough to cope with most real-world data classification problems. Kernel Fisher discriminant analysis (KFDA) tries to increase the expressiveness of the discriminant based on the high order statistics of the data set. In this paper, we propos...

Full description

Saved in:
Bibliographic Details
Main Authors: Yi-Hsiang Chao, Hsin-Min Wang, Ruei-Chuan Chang
Format: Conference Proceeding
Language:English
Subjects:
Acoustic scattering
Chaos
Information science
Kernel
Linear discriminant analysis
Mel frequency cepstral coefficient
Principal component analysis
Speaker recognition
Speech recognition
Statistical analysis
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Clearly, the linear discriminant classifier is not robust enough to cope with most real-world data classification problems. Kernel Fisher discriminant analysis (KFDA) tries to increase the expressiveness of the discriminant based on the high order statistics of the data set. In this paper, we propose the GMM-based KFDA with the Bhattacharyya kernel to obtain a transformation, called a speaker eigenspace, based on which the transformed MFCC features are more discriminative for speaker recognition. In our approach, the eigenspace is directly constructed from the complete GMM parameter set, rather than the supervectors considering mean vectors only as in the eigenvoice approach. Moreover, FDA, which is believed to be more appropriate for classification accuracy than principal component analysis (PCA), is applied for eigenspace construction. The speaker identification experiments show that the new features outperform the MFCC features, in particular when the amount of enrollment data for each speaker is very small.
ISSN:1520-6149
2379-190X
DOI:10.1109/ICASSP.2005.1415197