Maximum penalized likelihood Kernel regression for fast adaptation

IEEE Transactions on Audio, Speech and Language Processing

Volumn 17, Issue 7, 2009, Pages 1372-1381

  Mak, Brian Kan Wing ^a   Lai, Tsz Chung ^a   Tsang, Ivor W ^b   Kwok, James Tin Yau ^a  

^a HONG KONG UNIVERSITY OF SCIENCE AND TECHNOLOGY   (Hong Kong)

^b NANYANG TECHNOLOGICAL UNIVERSITY   (Singapore)

Author keywords

Kernel regression; Maximum likelihood linear regression (MLLR); Reference speaker weighting; Speaker adaptation

Indexed keywords

ADAPTATION ALGORITHMS; ADAPTATION METHODS; AFFINE MODEL; BAYESIAN PERSPECTIVE; EIGENVOICES; FAST ADAPTATIONS; GAUSSIANS; HIGH-DIMENSIONAL FEATURE SPACE; KERNEL METHODS; KERNEL REGRESSION; MAXIMUM A POSTERIORI; MAXIMUM PENALIZED LIKELIHOOD; MAXIMUM-LIKELIHOOD LINEAR REGRESSION (MLLR); NONLINEAR GENERALIZATIONS; REFERENCE SPEAKER WEIGHTING; RESOURCE MANAGEMENT; SPEAKER ADAPTATION; SYSTEM OF LINEAR EQUATIONS; UNSUPERVISED SPEAKER ADAPTATION; WALL STREET JOURNAL; WORD ERROR RATE REDUCTIONS;

CONVEX OPTIMIZATION; LINEAR REGRESSION; RESOURCE ALLOCATION; SPEECH RECOGNITION;

MAXIMUM LIKELIHOOD;

EID: 68549133255     PISSN: 15587916     EISSN: None     Source Type: Journal    
DOI: 10.1109/TASL.2009.2019920     Document Type: Article

References (30)

1. J. R. Bellegarda, P. V. de Souza, A. J. Nadas, D. Nahamoo, M. A. Picheny, and L. R. Bahl, "Robust speaker adaptation using a piecewise linear acoustic mapping," in Proc. IEEE Int. Conf. Acoust, Speech, Signal Process., 1992, vol. 1, pp. 445-448.
2. M. J. F. Gales, "Maximum likelihood linear transformations for HMMbased speech recognition," Comput. Speech Lang., vol. 12, pp. 75-98, Apr. 1998.
3. V. V. Digalakis, D. Rtischev, and L. G. Neumeyer, "Speaker adaptation using constrained estimation of Gaussian mixtures," IEEE Trans. Speech Audio Process., vol. 3, no. 5, pp. 357-366, Sep. 1995.
4. T. Kosaka, S. Matsunaga, and S. Sagayama, "Speaker-independent speech recognition based on tree-structured speaker clustering," Comput. Speech Lang., vol. 10, pp. 55-74, 1996.
5. M. F. J. Gales, "Cluster adaptive training of hidden Markov models," IEEE Trans. Speech Audio Process., vol. 8, no. 4, pp. 417-428, Jul. 2000.
6. T. J. Hazen, "A comparison of novel techniques for rapid speaker adaptation," Speech Commun., vol. 31, pp. 15-33, May 2000.
7. R. Kuhn, J.-C. Junqua, P. Nguyen, and N. Niedzielski, "Rapid speaker adaptation in eigenvoice space," IEEE Trans. Speech Audio Process., vol. 8, no. 6, pp. 695-707, Nov. 2000.
8. K. T. Chen, W. W. Liau, H. M. Wang, and L. S. Lee, "Fast speaker adaptation using eigenspace-based maximum likelihood linear regression," in Proc. Int. Conf. Spoken Lang. Process., 2000, vol. 3, pp. 742-745.
9. J. L. Gauvain and C. H. Lee, "Maximum a posteriori estimation for multivariate Gaussian mixture observations of Markov chains," IEEE Trans. Speech Audio Process., vol. 2, no. 2, pp. 291-298, Apr. 1994.
10. C. J. Leggetter and P. C. Woodland, "Maximum likelihood linear regression for speaker adaptation of continuous density hidden Markov models," Comput. Speech Lang., vol. 9, pp. 171-185, 1995.
11. B. Mak, T.-C. Lai, and R. Hsiao, "Improving reference speaker weighting adaptation by the use of maximum-likelihood reference speakers," in Proc. IEEE Int. Conf. Acoust., Speech, Signal Process., Toulouse, France, May 14-19, 2006, vol. 1, pp. 229-232.
12. C. Chesta, O. Siohan, and C. H. Lee, "Maximum a posteriori linear regression for hidden Markov model adaptation," in Proc. Eur. Conf. Speech Commun. Technol., 1999, vol. 1, pp. 211-214.
13. O. Siohan, T. A. Myrvoll, and C. H. Lee, "Structural maximum a posteriori linear regression for fast HMM adaptation," Comput. Speech Lang., vol. 16, pp. 5-24, Jan. 2002.
14. K. Shinoda and C. H. Lee, "Unsupervised adaptation using structural Bayes approach," in Proc. IEEE Int. Conf. Acoust., Speech, Signal Process., 1998, pp. 793-796.
15. H. Erdogan, Y. Q. Gao, and M. Picheny, "Rapid adaptation using pe-nalized-likelihood methods," in Proc. IEEE Int. Conf. Acoust., Speech, Signal Process., 2001, vol. 1, pp. 333-336.
16. A. Gunawardana and W. Byrne, "Discounted likelihood linear regression for rapid speaker adaptation," Comput. Speech Lang., vol. 15, pp. 15-38, Jan. 2001.
17. I. W. Tsang, J. T. Kwok, B. Mak, K. Zhang, and J. J. Pan, "Fast speaker adaptation via maximum penalized likelihood kernel regression," in Proc. IEEE Int. Conf. Acoust., Speech, Signal Process., Toulouse, France, May 14-19, 2006, vol. 1, pp. 997-1000.
18. N. Cristianini and J. Shawe-Taylor, An Introduction to Support Vector Machines. Cambridge, U.K.: Cambridge Univ. Press, 2000.
19. B. Schölkopf and A. J. Smola, Learning With Kernels: Support Vector Machines, Regularization, Optimization, and Beyond. Cambridge, MA: MIT Press, 2002.
20. V. Vapnik, Statistical Learning Theory. New York: Wiley, 1998.
21. G. Saon, "A non-linear speaker adaptation technique using kernel ridge regression," in Proc. IEEE Int. Conf. Acoust., Speech, Signal Process., 2006, vol. I, pp. 225-228.
22. M. J. F. Gales, "Maximum likelihood linear transformations for HMM-based speech recognition," Comput. Speech Lang., vol. 12, pp. 75-98, Apr. 1998.
23. A. J. Hewett, "Training and Speaker Adaptation in Template-Based Speech Recognition," Ph.D. dissertation, Cambridge Univ., Cambridge, U.K., 1989.
24. B. Schölkopf and A. J. Smola, Learning With Kernels. Cambridge, MA: MIT Press, 2002.
25. P. J. Green, "Penalized likelihood," in Encyclopedia of Statistical Sciences, Update Volume 2. New York: Wiley, 1998, pp. 578-586.
26. B. Mak, J. T. Kwok, and S.Ho, "Kernel eigenvoice speaker adaptation," IEEE Trans. Speech Audio Process., vol. 13, no. 5, pp. 984-992, Sep. 2005.
27. B. Mak, R. Hsiao,S. Ho, andJ.T. Kwok, "Embedded kernel eigenvoice speaker adaptation and its implication to reference speaker weighting," IEEE Trans. Speech Audio Process., vol. 14, no. 4, pp. 1267-1280, Jul. 2006.
28. B. Mak and R. Hsiao, "Kernel eigenspace-based MLLR adaptation," IEEE Trans. Audio, Speech, Lang. Process., vol. 15, no. 3, pp. 784-795, Mar. 2007.
29. P. Price, W. M. Fisher, J. Bernstein, and D. S. Pallett, "The DARPA 1000-word resource management database for continuous speech recognition," in Proc. IEEE Int. Conf. Acoust., Speech, Signal Process., 1988, vol. 1, pp. 651-654.
30. D. B. Paul and J. M. Baker, "The design of the Wall Street Journalbased CSR corpus," in Proc. DARPA Speech Natural Lang. Workshop, Feb. 1992.