Browse > Article

Robust Feature Parameter for Implementation of Speech Recognizer Using Support Vector Machines  

김창근 (동아대학교 전자공학과)
박정원 (LG 이노텍 구미연구)
허강인 (동아대학교 전자공학과)
Publication Information
Abstract
In this paper we propose effective speech recognizer through two recognition experiments. In general, SVM is classification method which classify two class set by finding voluntary nonlinear boundary in vector space and possesses high classification performance under few training data number. In this paper we compare recognition performance of HMM and SVM at training data number and investigate recognition performance of each feature parameter while changing feature space of MFCC using Independent Component Analysis(ICA) and Principal Component Analysis(PCA). As a result of experiment, recognition performance of SVM is better than 1:.um under few training data number, and feature parameter by ICA showed the highest recognition performance because of superior linear classification.
Keywords
Support Vector Machines; Independent Component Analysis; Principal Component Analysis;
Citations & Related Records
연도 인용수 순위
  • Reference
1 A.Hyvarinen, J.Karhunen, E.Oja, Independent Component Analysis, John Wiley & Sons, 2001
2 S.Haykin, Neural Networks, Prentice hall, 1999
3 Scholkopf, Smola Learning with Kernels, The MIT Press, 2002
4 N.Cristianini, J.S.Taylor, An Introduction to Support Vector Machine and other kernel-based learning methods, Cambridge University Press, 2000
5 S.Choi, A.Cichocki, S.Amari, 'Flexible independent component analysis', IEICE Trans. Fundamentals, vol.E83-A, no.12, pp.2715- 2722, August, 2000
6 C.K.Kim, S.B.Kim, S.H.Kim, K.I.Hur, 'Performance Improvement of Speech Recognition Based on Independent Component Analysis', ICSP2001, Vol.2 of 2, pp.663-666, August, 2001
7 G.J.Jang, Y.H.Oh, 'Text-independent speaker identification using support vector machines', Proceeding of 11th KISS fall conference, pp.45 -48, November, 1999