SVMs을 이용한 효율적인 음소단위 분할 = Efficient Segmentation by Phoneme Unit using SVMs

다국어 초록 (Multilingual Abstract)

In this research, we used Support Vector Machines(SVMs) as the learning and recognition unit of the speech, one of artificial neural network, to segmented from the continuous speech into phonemes, an initial, medial, and final sound, and then, performed continuous speech recognition from it, A decision boundary of phoneme is determined by algorithm with maximum frequency in a short interval. speech recognition process is performed by Continuous Hidden Markov Model(CHMM), and we compared it with another phoneme segregated from the eye-measurement. From the simulation results, we confirmed that the method, SVMs, we proposed is more effective in an initial sound than Gaussian Mixture Models(GMMs). We plan to construct a optimum hybrid classifier of SVMs and GMMs, and apply to continuous speech recognition.

참고문헌 (Reference)

1 X. Peng, "TPMSVM: A novel twin parametric margin support vector machine for pattern recognition" 44 (44): 2678-2692, 2011

2 Edgar E. Osuna, "Robert Freund and Federico Girosi, Support vector machines: training and applications" C.B.C.L 2007

3 Philip Clarkson, "On the use of support vector machines for phonetic classification" 2005

4 J. Ghosh, "Multiclassifier systems: back to the future" 2364 : 1-15, 2002

5 J. Weston, "Multi-class support vector machines" University of London 2008

6 Yonas B. Dibike, "Model introduction with support vector machines:introduction and applications" 15 (15): 208-216, 2007

7 Bernhard Scholkopf, "Learning with Kernels" The MIT Press 2002

8 N. Cristianini, "An introduction to support vector machine and other kernel based learning methods" Cambridge University Press 2000

9 Christopher J.C. Burges, "A tutorial on support vector machines for pattern recognition" Bell Laboratories, Lucent Technologies 2008