Margin Based Learning: A Framework for Acoustic Model Parameter Estimation

Full Text (PDF, 419KB), PP.26-31

Views: 0 Downloads: 0

Author(s)

Syed Abbas Ali 1,* Najmi Ghani Haider 2 Mahmood K. Pathan 2

1. Computer & Information Systems Engineering, N.E.D University of Engineering & Technology

2. Computer Science & Information Technology, N.E.D University of Engineering & Technology

* Corresponding author.

DOI: https://doi.org/10.5815/ijisa.2012.12.04

Received: 3 Feb. 2012 / Revised: 11 Jun. 2012 / Accepted: 15 Sep. 2012 / Published: 8 Nov. 2012

Index Terms

Statistical Learning, Generalization Capability, Empirical Risk, Discriminative Training, Test Risk Bound

Abstract

Statistical learning theory has been introduced in the field of machine learning since last three decades. In speech recognition application, SLT combines generalization function and empirical risk in single margin based objective function for optimization. This paper incorporated separation (misclassification) measures conforming to conventional discriminative training criterion in loss function definition of margin based method to derive the mathematical framework for acoustic model parameter estimation and discuss some important issues related to hinge loss function of the derived model to enhance the performance of speech recognition system.

Cite This Paper

Syed Abbas Ali, Najmi Ghani Haider, Mahmood Khan Pathan, "Margin Based Learning: A Framework for Acoustic Model Parameter Estimation", International Journal of Intelligent Systems and Applications(IJISA), vol.4, no.12, pp.26-31, 2012. DOI:10.5815/ijisa.2012.12.04

Reference

[1]A.P. Dempster, N. M. Laird and D. B. Gopinath, “Maximum Likelihood from incomplete data via the EM algorithm,” J. Roy.Stat.Soc., 39(1), 1-38, 1977.

[2]B. -H. Juang, W. Chou, and C.-H. Lee, “Minimum Classification Error rate methods for speech recognition,” IEEE Trans. on Speech and Audio Proc., vol.5, no.3, pp.257-265, 1997.

[3]Y. Normandin, “Maximum Mutual Information Estimation of Hidden Markov Models,” In Automatic Speech and Speaker Recognition, Kluwer Academics Publishers, 1996.

[4]D. Povey and P. Woodland, “Minimum Phone error and I-smoothing for improved discriminative training,” Proc ICCASP, vol.1, pp. 105-108, 2002.

[5]V. Vapnik, “The nature of Statistical Learning Theory,” Springer-Verlag, New york, 1995.

[6]H. Jiang, X. Li and C. Liu, “Large Margin Hidden Markov models for speech recognition,” IEEE Trans. Audio, Speech, and Language Processing, vol.14, no.5, pp.1584-1595, 2006.

[7]J. Li and C. -H. Lee, “Soft margin feature extraction for automatic speech recognition,” Proc. Interspeech, 2007. 

[8]O. Bousquet, S. Bouchern and G. Lugosi, “Introduction to statistical learning theory. Advanced lectures on machine learning lecture notes in artificial intelligence 3176, 167-207. (Eds) Springer, Heidelberg, Germany (2004).

[9]J. Hui, “ Discriminative training of HMMs for automatic speech recognition: A survey,” Computer speech and language, Elsevier Ltd. 2010

[10]L. R. Bahl, P. F. Brown, P.V. De souza, R. L. Mercer, “ Maximum mutual information estimation of hidden markov model parameters for speech recognition,” in Proc IEEE, International conference on Acoustic, Speech and Signal processing (ICASSP 86), Tokyo, Japan, pp. 49-52.

[11]A. Nadas, D. Nahamoo, M. A. Picheny, “On a model-robust training method for speech recognition,” IEEE Transaction on Acoustic, Speech and Signal Processing 36(9), 1432-1436. 1988.

[12]B. -H. Juang, W. Chou, C.-H. Lee, “Minimum classification error rate methods for speech recognition,” IEEE Transaction on Speech and Audio Processing 5(3), 257-265. 1997.

[13]S. Katagiri, B. -H .Juang, C.-H. Lee, “Pattern recognition using a generalized probabilistic decent method,” In Proc IEEE 86(11), 2345-2373. 1998.

[14]X. He, L. Deng, W. Chou, “Discriminative learning in sequential pattern recognition: a unifying view for optimization-based speech recognition,” IEEE Signal Processing Magazine, 14-36. 2008.

[15]D. Povey, “Discriminative training for large vocabulary speech recognition,” Ph.D. Dissertation, Cambridge University. 2004.

[16]H. Jiang, X. Li and C. Liu, “Large margin hidden markov model for speech recognition,” IEEE Trans. On Audio, speech and Language Proc., vol.14, no.5, pp.1548-1595, 2006.

[17]X. Li, H. Jiang and C. Liu, “Large margin for speech recognition,” Proc. ICASSP, pp. V513-V516, 2005.

[18]J. Li, M. Yuan and C.-H. Lee, “Soft margin estimation of hidden markov model parameters,” Proc. Interspeech, pp.2422-2425, 2006. 

[19]C. Burges, “A tutorial on support vector machines for pattern recognition,” Data Mining and Knowledge Discovery, 2(2),121-167. 1998.