ERROR CORRECTIVE CLASSIFIER FUSION FOR SPOKEN LANGUAGE RECOGNITION

被引：0

作者：

Dehzangi, Omid ^{[1
]}

Ma, Bin ^{[2
]}

Chng, Eng Siong ^{[1
]}

Li, Haizhou ^{[1
,2
,3
]}

机构：

[1] Nanyang Technol Univ, Sch Comp Engn, Singapore, Singapore

[2] ASTAR, Inst Infocomm Res, Singapore 138632, Singapore

[3] Univ Eastern Finland, Dept Comp Sci & Stat, FI-80101 Joensuu, Finland

来源：

2010 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING | 2010年

关键词：

Error Corrective Training; ROC Analysis; Classifier Fusion; Spoken Language Recognition;

D O I：

10.1109/ICASSP.2010.5495235

中图分类号：

O42 [声学];

学科分类号：

070206 ; 082403 ;

摘要：

A number of effective classification algorithms have been developed for spoken language recognition, and it has been a common practice in the NIST Language Recognition Evaluations (LREs) that an information fusion is applied to boost the performance of the recognition system. This paper investigates the fusion of multiple output scores generated using different classifiers that complement to further reduce the classification error rate in spoken language recognition. We introduce a local performance metric to optimize the performance of the classifier fusion. The experiments are conducted on the 2009 NIST LRE corpus. The experimental results show that the proposed fusion effectively improves the performance over individual classifiers.

引用

页码：1994 / 1997

页数：4

共 11 条

[1] [Anonymous], P ICASSP
[2] Bezdek J., 1999, FUZZY MODELS ALGORIT
[3] Brummer N., FOCAL MULTICLASS TOO
[4] FAWCETT T, HPL2004
[5] An introduction to ROC analysis
Fawcett, Tom
[J]. PATTERN RECOGNITION LETTERS, 2006, 27 (08) : 861 - 874
[6] KITTLER J, 1998, IEEE T PATT ANAL MAC, V20
[7] Lachiche Nicolas., 2003, P 20 INT C MACHINE L, P416
[8] LI H, 2009, 2009 NIST LRE WORKSH
[9] TONG R, 2006, P ICASSP
[10] TORRESCARRASQUI.PA, 2007, P INT

← 1 2 →