Generalization problem in ASR acoustic model training and adaptation

Proceedings of the 2009 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2009

Volumn , Issue , 2009, Pages 1-10

  Furui, Sadaoki ^a  

^a TOKYO INSTITUTE OF TECHNOLOGY   (Japan)

Author keywords

[No Author keywords available]

Indexed keywords

ACOUSTIC MODEL; APRIORI; AUTOMATIC SPEECH RECOGNITION SYSTEM; DATA SPARSENESS PROBLEM; DECISION BOUNDARY; DEGREE OF FREEDOM; LARGE-SCALE DATABASE; MODEL TRAINING; RECOGNITION ACCURACY; RESEARCH ISSUES; SPEECH DATA; SPEECH DATABASE; STATISTICAL MODELS; TRAINING SAMPLE;

FACE RECOGNITION; REMELTING;

SPEECH RECOGNITION;

EID: 77949431571     PISSN: None     EISSN: None     Source Type: Conference Proceeding    
DOI: 10.1109/ASRU.2009.5373493     Document Type: Conference Paper

References (78)

1. T. Anastasakos, J. McDonough, R. Schwarts, and J. Makhoul, "A compact model for speaker-adaptive training," Proc. ICSLP96, 2, FrP2L1.3, 1996.
2. M. Benzeghiba, et al., "Impact of variabilities on speech recognition," Proc. SPECOM, St Petersburg, pp. 3-16, 2006.
3. J. Bilmes, "Graphical models and automatic speech recognition," Technical Report UWEETR-2001-05, University of Washington, Dept. of EE, Seattle WA, 2001.
4. K. Chen and H. Wang, "Eigenspace-based maximum a posteriori linear regression for rapid speaker adaptation," Proc. ICASSP, p3.2, 2001.
5. C. Chesta, O. Siohan, and C. Lee, "Maximum a posteriori linear regression for hidden Markov model adaptation," Proc. Eurospeech, 1, pp. 211-214, 1999.
6. V. V. Digalakis, D. Rtischev, and L.G. Neumeyer, "Speaker adaptation using constrained estimation of Gaussian mixtures," IEEE Trans, Speech, Audio Processing, 3, pp. 357-366, 1995.
7. V. V. Digalakis and L.G. Neumeyer, "Speaker adaptation using combined transformation and Bayesian methods," IEEE Trans, Speech Audio Processing, 4, pp. 294-300, 1996.
8. J. Droppo and A. Acero, "Environmental robustness," Springer Handbook of Speech Processing, Springer, pp. 653-679, 2008.
9. nd Edition, John Wiley & Sons, Inc., 2001.
10. S. Furui, "A training procedure for isolated word recognition systems," IEEE Trans. Acoustics, Speech and Signal Processing, 28, 2, pp. 129-136, 1980.
11. S. Furui, "Unsupervised speaker adaptation based on hierarchical spectral clustering," IEEE Trans. Acoustics, Speech and Signal Processing, 37, 12, pp. 1923-1930, 1989.
12. S. Furui, "50 years of progress in speech and speaker recognition," Proc. SPECOM, Patras, Greece, pp. 1-9, 2005.
13. S. Furui, "Recent progress in corpus-based spontaneous speech recognition," IEICE Trans. Information and Systems, E88-D, 3, pp.366-375, 2005.
14. S. Furui, "Selected topics from 40 years of research on speech and speaker recognition," Proc. Interspeech, Brighton, UK, Mon-Ses1-K, 2009.
15. th Int. Conf. on Text, Speech and Dialogue (TSD 2005), Karlovy Vary, Czech Republic, pp. 9-22, 2005.
16. S. Furui, M. B. J. Hirschberg, S. Itahashi, T. Kawahara, S. Nakamura, and S. Nakayanan, "Introduction to the special issue on spontaneous speech processing," IEEE Trans. Speech, Audio Process, 12, pp. 349-350, 2004.
17. M. J. F. Gales, "The generation and use of regression class trees for MLLR adaptation," CUED/F-INFENG/TR263, Cambridge University Engineering Department, 1996.
18. M. J. F. Gales, "Maximum likelihood linear transformations for HMM-based speech recognition," Computer, Speech and Language, 12, pp. 75-98, 1998.
19. M. J. F. Gales, "Cluster adaptive training for speech recognition," Proc. ICSLP, pp. 1783-1786, 1998.
20. M. J. F. Gales, "Acoustic factorization," Proc. ASRU 2001, Madonna di Campiglio, Italy, 2001.
21. M. J. F. Gales and P. C. Woodland, "Mean and variance adaptation within the MLLR framework," Computer Speech and Language, 10, pp. 249-264, 1996.
22. M. J. F. Gales and S. J. Young, "The application of hidden Markov models in speech recognition," Foundations and Trends in Signal Processing, 1, 3, pp. 195-304, 2007.
23. J.-L. Gauvain and C.-H. Lee, "Maximum a posteriori estimation for multivariate Gaussian mixture observation of Markov chains," IEEE Trans. Speech, Audio Process, 2, 2, pp.291-298, 1994.
24. X.-D. He and L. Deng, "Discriminative learning in speech recognition," Technical Report, MSR-TR-2007-129, Microsoft, 2007.
25. G. Heigold, T.Deselaers, R. Schlueter and H. Ney, "Modified MMI/MPE: a direct evaluation of the margin in speech recognition," Proc. Int. Conf. Machine Learning (ICML), Helsinki, Finland, 2008
26. Q. Huo and C.-H. Lee, "On-line adaptive learning of the correlated continuous-density hidden Markov model for speech recognition," IEEE Trans. Speech Audio Processing, 6, pp. 386-397, 1998.
27. H. Jiang, "Confidence measures for speech recognition: A survey," Speech Communication, 45, pp. 455-470, 2005.
28. B.-H. Juang and S. Katagiri, "Discriminative learning for minimum error classification," IEEE Trans. Signal Processing, 40, 12, pp. 3043-3054, 1992.
29. F. Korkmazsky, M. Deviren, D. Fohr, and I. Illina, "Hidden factor dynamic Bayesian networks for speech recognition," Proc. ICSLP, Jeju Island, Korea, 2004.
30. T. Kosaka, S. Matsunaga and S. Sagayama, "Tree-structured speaker clustering for speaker-independent continuous speech recognition," Proc. ICSLP, Yokohama, pp.1375-1378, 1994.
31. R. Kuhn, P. Nguyen, J.-C. Junqua, L. Goldwasser, N. Niedzielski, S. Finke, K. Field, and M. Contoloni, "Eigenvoices for speaker adaptation," Proc. ICSLP, pp. 1771-1774,1998.
32. C.-H. Lee, "Adaptation and compensation for speech recognition - learning from extra data to improve robustness," Proc. Int. Workshop on Hands-Free Speech Communication (HSC 2001), Kyoto, Japan, pp. 27-30, 2001.
33. L. Lee and R. C. Rose, "Speaker normalization using efficient frequency warping procedures," Proc. ICASSP, 1, pp.353-356, 1996.
34. C. Leggetter and P. Woodland, "Maximum likelihood linear regression for speaker adaptation of continuous density Hidden Markov Models", Computer, Speech and Language, 9, pp. 171-185, 1995.
35. S. Matsuda, T. Jitsuhiro, K. Markov, and S. Nakamura, "Speech recognition system robust to noise and speaking styles," Proc. ICSLP, Jeju Island, Korea, 2004.
36. T. Matsui and S. Furui. "A study of speaker adaptation based on minimum classification error training," Proc. Eurospeech, Madrid, Spain, pp. 81-84, 1995.
37. T. Matsui and S. Furui, "N-best-based unsupervised speaker adaptation for speech recognition," Computer Speech and Language, 12, pp. 41-50, 1998.
38. E. McDermott and A. Nakamura, "Recent advances in discriminative training for speech recognition," Proc. Spring Meeting, Acoustical Society of Japan, Tokyo, Japan, 2-5-10, 2009.
39. S. Nakagawa, T. Watanabe, H. Nishizaki and T. Utsuro, "An unsupervised speaker adaptation method for lecture-style spontaneous speech recognition using multiple recognition systems," IEICE Trans. Information and Systems, E88-D, 3, pp.463-471, 2005.
40. H. Nanjo and T. Kawahara, "Speaking-rate dependent decoding and adaptation for spontaneous lecture speech recognition," Proc. ICASSP, pp. 725-728, 2002.
41. P. Nguyen, R. Kuhn, J.-C. Junqua, N. Niedzielski, and C. Wellekens, "Eigenvoices : a compact representation of speakers in a model space," Annales des Telecommunications, 55, 2000.
42. Y. Normandin, "Maximum mutual information estimation of hidden Markov models," in Automatic Speech and Speaker Recognition, C.-H. Lee, F. K. Soong and K. K. Paliwal eds., Kluwer Academic Publishers, Norwell, MA, 1996.
43. K. Ohkura, M. Sugiyama and S. Sagayama, "Speaker adaptation based on transfer-vector-field smoothing with continuous mixture density HMMs", Proc. ICSLP, Alberta, pp.369-372, 1992.
44. M. Padmanabhan, L. R. Bahl, D. Nahamoo and M. A. Picheny, "Speaker clustering and transformation for speaker adaptation in speech recognition systems," IEEE Trans. Speech Audio Processing, 6, pp. 71-77, 1998.
45. M. Picheny and D. Nahamoo, "Towards superhuman speech recognition," Springer Handbook of Speech Processing, Springer, pp. 597-616, 2008.
46. D. Povey, Discriminative training for large vocabulary speech recognition, Ph. D. thesis, Cambridge University, 2003.
47. D. Povey and P. C. Woodland, "Minimum phone error and i-smoothing for improved discriminative training," Proc. ICASSP, Orlando, pp. I-105-108, 2002.
48. D. Pye and P. C. Woodland, "Experiments in speaker normalization and adaptation for large vocabulary speech recognition," Proc. ICASSP, 2, pp. 1047-1050, 1997.
49. M. Rahim and B.-H. Juang, "Signal bias removal by maximum likelihood estimation for robust telephone speech recognition," IEEE Trans. Speech, Audio Process., 4, 1, pp.19-30, 1996.
50. C. K. Raut and M. J. Gales, "Bayesian discriminative adaptation for speech recognition," Proc. ICASSP, Taipei, Taiwan, pp. 4361-4364, 2009.
51. R. Rose, "Environmental robustness in automatic speech recognition," Proc. COST278 and ISCA Workshop on Robustness Issues in Conversational Interaction, Norwich, UK, 2004.
52. A-V. I. Rosti and M. J. F. Gales, "Factor analysed hidden Markov models," Proc. ICASSP, Orlando, pp. I-949-952, 2002.
53. S. Sagayama, K. Shinoda, M. Nakai and H. Shimodaira, "Analytic methods for acoustic model adaptation: a review," Proc. ISCA ITR-Workshop, Sophia-Antipolis, pp.67-76, 2001.
54. S. Sagayama, Y. Yamaguchi, S. Takahashi, and J. Takahashi, "Jacobian approach to fast acoustic model adaptation," Proc. ICASSP, pp.835-838, 1997.
55. G. Saon and D. Povey, "Penalty function maximization for large margin HMM training," Proc. Interspeech, Brisbane, Australia, pp. 920-923, 2008.
56. K. Shinoda, K. Iso, and T. Watanabe, "Speaker adaptation for demisyllable based continuous density HMM," Proc. ICASSP, S13.7, pp. 857-860, 1991.
57. K. Shinoda and C.-H. Lee, "A structural Bayes approach to speaker adaptation," IEEE Trans. Speech, Audio Process., 9, 3, pp. 276-287, 2001.
58. K. Shinoda and T. Watanabe, "Speaker adaptation with autonomous control using tree structure," Proc. Eurospeech, pp. 1143-1146, 1995.
59. K. Shinoda and T. Watanabe, "Speaker adaptation with autonomous model complexity control by MDL principle," Proc. ICASSP, pp.717-720, 1996.
60. T. Shinozaki and S. Furui, "Hidden mode HMM using bayesian network for modeling speaking rate fluctuation," Proc. of ASRU, US Virgin Islands, pp.417-422, 2003.
61. T. Shinozaki and S. Furui, "Time adjustable mixture weights for speaking rate fluctuation," Proc. Eurospeech, pp. 973-976, 2003.
62. T. Shinozaki and S. Furui, "Spontaneous speech recognition using a massively parallel decoder," Proc. ICSLP, Jeju Island, Korea, pp. 1705-1708, 2004.
63. T. Shinozaki, Y. Kubota and S. Furui, "Unsupervised cross-validation adaptation algorithms for improved adaptation performance," Proc. IEEE ICASSP, Taipei, Taiwan, pp. 4377-4380, 2009.
64. T. Shinozaki, Y. Kubota and S. Furui, "Unsupervised cross-validation and aggregated adaptations for improved speech recognition," IPSJ SIG Technical Report, 2009-SLP-75, pp. 1-6, 2009. (in Japanese)
65. Y. Shiraki and M. Honda, "Speaker adaptations algorithms for segment vocoder," Trans, Committee of Speech Res., Acoust. Soc. Japan, SP87-67, 1987. (in Japanese)
66. O. Siohan, C. Chesta and C.-H. Lee, "Hidden Markov model adaptation using maximum a posteriori liner regression," Proc. Workshop on Robust Methods for Speech Recognition in Adverse Conditions, pp.147-150, Tampere, Finland, 1999.
67. O. Siohan, T.-A. Myrvoll and C.-H. Lee, "Structural maximum a posteriori linear regression for fast HMM adaptation," Proc. ISCA ITRW Workshop on ASR, 2000.
68. A. C. Surendran and C.-H. Lee, "Transformation-based Bayesian prediction for adaptation of HMMs," Speech Communication, 34, pp. 159-174, 2001.
69. J. Takahashi and S. Sagayama, "Vector-field-smoothed Bayesian learning for incremental speaker adaptation," Proc. ICASSP, Detroit, pp. 696-699, 1995.
70. M. Tonomura, T. Kosaka, and S. Matsunaga, "Speaker adaptation based on transfer-vector-field-smoothing using maximum a posteriori probability estimation," Proc. ICASSP, Detroit, pp. 688-691, 1995.
71. H. Wakita, "Normalization of vowels by vocal tract length and its application to vowel identification," IEEE Trans. Acoustics, Speech and Signal Processing, 25, 2, pp. 183-192, 1977.
72. L. Wang and P. C. Woodland, "MPE-based discriminative linear transforms for speaker adaptation," Computer Speech and Language, 22, 3, pp. 256-272, 2008.
73. P. Woodland and D. Povey, "Large scale discriminative training for speech recognition," Proc. ITRW ASR, 2000.
74. D. Yu, L. Deng, X.-D. He and A. Acero, "Large-margin minimum classification error training: A theoretical risk minimization perspective," Computer Speech and Language, 22, pp. 415-429, 2008.
75. K. Yu and M. J. F. Gales, "Bayesian adaptive inference and adaptive training," IEEE Trans. Audio, Speech, Language Process., 15, 6, pp. 1932-1943, 2007.
76. K. Yu, M. J. F. Gales, and P. C. Woodland, "Unsupervised discriminative adaptation using discriminative mapping transforms," Proc. ICASSP, pp. 4273-4276, 2008.
77. G. Zavaliagkos, R. Schwartz, and J. McDonough, "Batch, incremental and instantaneous adaptation techniques for speech recognition" Proc. ICASSP, Detroit, pp. 676-679, 1995.
78. G. Zweig, "Speech recognition with dynamic Bayesian networks," Ph.D. Thesis, University of California, Berkeley, 1998.