DocumentCode :
3422625
Title :
Nonlinear kernel nuisance attribute projection for speaker verification
Author :
Zhao, Xianyu ; Dong, Yuan ; Yang, Hao ; Zhao, Jian ; Lu, Liang ; Wang, Haila
Author_Institution :
France Telecom R&D Center, Beijing
fYear :
2008
fDate :
March 31 2008-April 4 2008
Firstpage :
4125
Lastpage :
4128
Abstract :
Nuisance attribute projection (NAP) was successfully applied in SVM-based speaker verification systems to improve performance by doing projection to remove dimensions from the SVM feature space that cause unwanted variability in the kernel. Previous studies of NAP were focused mainly on linear and generalized linear kernel SVMs. In this paper, NAP in nonlinear kernel SVMs, e.g. polynomial or Gaussian kernels, are investigated. Instead of doing explicit feature expansion and projection in high-dimension feature space, kernel principal component analysis is employed to find nuisance dimensions; and, NAP is carried out implicitly by incorporating it into some compensated kernel functions. Experimental results on the 2006 NIST SRE corpus indicate the effectiveness of such nonlinear kernel NAP. Compared with linear NAP, nonlinear NAP with Gaussian kernel obtained about 11% relative improvement in equal error rate (EER).
Keywords :
Gaussian processes; polynomials; principal component analysis; speaker recognition; Gaussian kernels; SVM-based speaker verification systems; equal error rate; high-dimension feature space; kernel principal component analysis; nonlinear kernel; nonlinear kernel nuisance attribute projection; polynomial kernels; speaker verification; Input variables; Kernel; NIST; Polynomials; Principal component analysis; Research and development; Speaker recognition; Support vector machine classification; Support vector machines; Telecommunications; kernel principal component analysis; nuisance attribute projection; speaker location; speaker recognition; supporting vector machines;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Acoustics, Speech and Signal Processing, 2008. ICASSP 2008. IEEE International Conference on
Conference_Location :
Las Vegas, NV
ISSN :
1520-6149
Print_ISBN :
978-1-4244-1483-3
Electronic_ISBN :
1520-6149
Type :
conf
DOI :
10.1109/ICASSP.2008.4518562
Filename :
4518562
Link To Document :
بازگشت