Phonemic segmentation using the generalised Gamma distribution and small sample Bayesian information criterion

Speech Communication

Volumn 50, Issue 1, 2008, Pages 38-55

  Almpanidis, George ^a   Kotropoulos, Constantine ^a  

^a ARISTOTLE UNIVERSITY OF THESSALONIKI   (Greece)

Author keywords

Bayesian information criterion; Generalised Gamma distribution; Phonemic segmentation; Small sample

Indexed keywords

ALGORITHMS; BAYESIAN NETWORKS; INFORMATION ANALYSIS; PROBLEM SOLVING;

BAYESIAN INFORMATION CRITERION; GENERALISED GAMMA DISTRIBUTION; PHONEMIC SEGMENTATION;

SPEECH ANALYSIS;

EID: 35348882681     PISSN: 01676393     EISSN: None     Source Type: Journal    
DOI: 10.1016/j.specom.2007.06.005     Document Type: Article

References (81)

1. Adell, J., Bonafonte, A., 2004. Towards phone segmentation for concatenative speech synthesis. In: Proc. 5th ISCA Speech Synthesis Workshop (SSW5), June, pp. 139-144.
2. Akaike H. A new look at the statistical model identification. IEEE Trans. Automat. Contr. 19 24 (1974) 716-723
3. Anderson D., and Burnham K. Understanding information criteria for selection among capture-recapture or ring recovery models. Bird Study 46 (1999) 514-521
4. Anderson D., and White G. AIC model selection in overdispersed capture-recapture data. Ecology 75 (1994) 1780-1793
5. Aversano, G., Esposito, A., Marinaro, M., 2001. A new text-independent method for phoneme segmentation. In: Proc. 44th IEEE Midwest Symposium on Circuits and Systems, Vol. 2, pp. 516-519.
6. Bakis, R., Chen, S., Gopalakrishnan, P., Gopinath, R., Maes, S., Polymenakos, L., Franz, M., 1997. Transcription of broadcast news shows with the IBM large vocabulary speech recognition system. In: Proc. Speech Recognition Workshop, pp. 67-72.
7. Bernardo J. Algorithm AS 103: Psi (digamma) function. Appl. Statist. 25 (1976) 315-317
8. Bogdan M., Ghosh J., and Doerge R. Modifying the Schwarz Bayesian information criterion to locate multiple interacting quantitative trait loci. Genetics 167 (2004) 989-999
9. Bollen, K., Ray, O., Zavisca, J., 2005. A scaled unit information prior approximation to Bayes factor. SAMSI LVSSS Workshop Transition Workshop: Latent Variable Models in the Social Sciences, November.
10. Bozdogan H. Model selection and Akaike's information criterion (AIC): the general theory and its analytical extensions. Psychometrika 52 (1987) 345-370
11. Bozdogan, H., 1988. ICOMP: a new model selection criterion. In: Proc. Classification and Related Methods in of Data Analysis, April, pp. 599-608.
12. Brugnara, F., De Mori, R., Giuliani, D., Omologo, M., 1992. Improved connected digit recognition using spectral variation functions. In: Proc. Internat. Conf. on Spoken Language Processing (ICSLP92), October, Vol. 1, pp. 627-630.
13. Burnham K., and Anderson D. Multimodel inference understanding AIC and BIC in model selection. Sociol. Methods Res. 33 2 (2004) 261-304
14. Cettolo, M., Federico, M., 2000. Model selection criteria for acoustic segmentation. In: Proc. ISCA ITRW ASR2000 Automatic Speech Recognition, pp. 221-227.
15. Chang, J., Shin, J., Kim, N.S., 2003. Likelihood ratio test with complex Laplacian model for voice activity detection. In: Proc. European Conf. on Speech Communication Technology, pp. 1065-1068.
16. Chen, S., Gopalakrishnam, P., 1998. Speaker, environment and channel change detection and clustering via the Bayesian information criterion. In: DARPA Speech Rec. Workshop.
17. Chen S., Eide E., Gales M., Gopinath R., Kanvesky D., and Olsen P. Automatic transcription of broadcast news. Speech Comm. Archive - Special Issue on Automatic Transcription of Broadcast News Data 37 1-2 (2002) 69-87
18. Chickering D., and Heckerman D. Efficient approximations for the marginal likelihood of Bayesian networks with hidden variables. Machine Learning 29 2-3 (1997) 181-212
19. Cohen A., and Whitten B. Modified moment estimation for the three-parameter gamma distribution. J. Quality Technol. 17 (1986) 147-154
20. Dayton C. Model comparisons using information measures. Modern Appl. Statist. Methods 2 (2003) 281-292
21. Delacourt P., and Wellekens C.J. DISTBIC: a speaker-based segmentation for audio data indexing. Speech Comm. 32 1-2 (2000) 111-126
22. Esposito, A., Aversano, G., 2004. Text independent methods for speech segmentation. In: Proc. Summer School on Neural Networks, pp. 261-290.
23. Garofolo, J., Lamel, L., Fisher, W., Fiscus, J., Pallett, D., Dahlgren, N., 1990. The DARPA TIMIT acoustic-phonetic continuous speech corpus. National Institute of Standards and Technology Speech. Disc 1-1.1, NTIS Order No. PB91-505065.
24. Gazor S., and Zhang W. Speech probability distribution. IEEE Signal Process. Lett. 10 7 (2003) 204-207
25. Gelman A., Carlin J., Stern H., and Rubin D. Bayesian Data Analysis (1995), Chapman & Hall/CRC, Boca Raton
26. Gheissari, N., Bab-Hadiashar, A., 2003. Model selection criteria in computer vision are they different? In: Proc. Digital Image Computing Techniques and Applications, December, Vol. 1, pp. 185-204.
27. Glass J. A probabilistic framework for segment-based speech recognition. Comput. Speech Lang. 17 (2003) 137-152
28. Hager H., and Bain L. Inferential procedures for the generalized Gamma distribution. J. Amer. Statist. Assoc. 82 (1987) 528-550
29. Hannan E., and Quinn B. The determination of the order of an autoregression. J. Roy. Statist. Soci. Ser. B 41 (1978) 190-195
30. Harter H. Maximum likelihood estimation of the parameters of a four-parameter generalized gamma population from complete and censored samples. Technometrics 9 (1967) 159-165
31. Haughton D. On the choice of a model to fit the data from an exponential family. Ann. Statist. 16 (1988) 342-355
32. Huy Dat, T., Takeda, K., Itakura, F., 2006. Multichannel speech enhancement based on speech spectral magnitude estimation using generalized Gamma prior distribution. In: Proc. 2006 IEEE Internat. Conf. on Acoustics, Speech, and Signal Processing, May, Vol. 4, pp. 1149-1152.
33. Hwang T., and Huang P. On new moment estimation of parameters of the Gamma distribution using its characterization. Ann. Inst. Statist. Math. 54 4 (2002) 840-847
34. Jankowski, C., Kalyanswamy, A., Basson, S., Spitz, J., 1990. NTIMIT: a phonetically balanced, continuous speech, telephone bandwidth speech database. In: Proc. 1990 IEEE Internat. Conf. on Acoustics, Speech, and Signal Processing, pp. 109-112.
35. Kadri, H., Lachiri, Z., Ellouze, N., 2006. Hybrid approach for unsupervised audio speaker segmentation. In: Proc. European Conf. in Signal Processing.
36. Kashyap R. A Bayesian comparison of different classes of dynamic models using empirical data. IEEE Trans. Automat. Contr. AC-22 5 (1977) 715-727
37. Kass R., and Wasserman L. A reference Bayesian test for nested hypotheses and its relationship to the Schwarz criterion. J. Amer. Statist. Assoc. 90 (1995) 928-934
38. Kokkinakis, K., Nandi, A., 2006. Flexible score functions for blind separation of speech signals based on generalized gamma probability density functions. In: Proc. 2006 IEEE Internat. Conf. on Acoustics, Speech, and Signal Processing, May, Vol. 1, pp. 1217-1220.
39. Kominek, J., Bennet, C., Black, A., 2003. Evaluating and correcting phoneme segmentation for unit selection synthesis. In: Proc. 8th European Conf. on Speech Communication and Technology (EUROSPEECH2003), September, pp. 313-316.
40. Kuha J. AIC and BIC: comparisons of assumptions and performance. Sociol. Methods Res. 33 2 (2004) 188-229
41. Lawless J. Inference in the generalized gamma and log gamma distributions. Technometrics 33 (1980) 409-419
42. Lee M. On the complexity of additive clustering models. Math. Psychol. 45 1 (2002) 131-148
43. Martin, R., 2005. Speech enhancement using short time spectral estimation with Gamma distributed priors. In: Proc. 2005 IEEE Internat. Conf. on Acoustics, Speech, Signal Processing, Vol. 1, pp. 253-256.
44. Mitchell, C., Harper, M., Jamieson, L., 1995. Using explicit segmentation to improve HMM phone recognition. In: Proc. 1995 IEEE Internat. Conf. Acoustics, Speech, and Signal Processing, Vol. 1, pp. 229-232.
45. Nakamura A. Acoustic modeling for speech recognition based on a generalized Laplacian mixture distribution. IEICE Trans. Inform. Syst. J83-D-II 11 (2000) 2118-2127
46. Nemer E., Goubran R., and Mahmould S. Robust voice activity detection using higher-order statistics in the LPC residual domain. IEEE Trans. Speech Audio Process. 9 3 (2001) 217-231
47. Olsthoorn T. Effective parameter optimization for groundwater model calibration. Ground Water. 33 (1995) 42-48
48. Parr V., and Webster J. A method for discriminating between failure density functions used in reliability predictions. Technometrics 7 (1965) 1-10
49. Pellom B., and Hansen J. Automatic segmentation of speech recorded in unknown noisy channel characteristics. Speech Commun. 25 1-3 (1998) 97-116
50. Pickett J., and Morris S. The Acoustics of Speech Communication: Fundamentals, Speech Perception Theory, and Technology (1999), Allyn & Bacon, Boston
51. Pigeon S., and Vandendorpe L. The M2VTS multimodal face database. Lecture Notes in Computer Science: Audio-and Video-based Biometric Person Authentication 1206 (1997) 403-409
52. Raftery A. Bayes factors and BIC: comment on "A critique of the Bayesian information criterion for model selection". Sociol. Methods Res. 27 3 (1999) 411-427
53. Raftery A., and Richardson S. Model selection for generalized linear models via GLIB, with application to epidemiology. Bayesian Biostatist. (1996) 321-354
54. Rissanen J. Modeling by the shortest data description. Automatica 14 (1978) 465-471
55. Schneider B. Algorithm AS 121: trigamma function. Appl. Statist. 27 (1978) 97-99
56. Schwarz G. Estimating the dimension of a model. Ann. Statist. 6 (1978) 461-464
57. Schwarz, P., Matejka, P., Cernocky, J., 2006. Hierarchical structures of neural networks for phoneme recognition. In: Proc. 2006 IEEE Internat. Conf. on Acoustics, Speech, and Signal Processing, May, Vol. 1, pp. 325-328.
58. Shewchuk, J., 1994. An introduction to the conjugate gradient method without the agonizing pain. Technical Report CMU-CS-94-125, School of Computer Science, Carnegie Mellon University, Pittsburgh, Philadelphia, March.
59. Shin J.W., and Chang J.-H. Statistical modeling of speech signals based on generalized gamma distribution. IEEE Signal Process. Lett. 12 3 (2005) 258-261
60. Shin, J.W., Chang, J.-H., Yun, H.S., Kim, N.S., 2005. Voice activity detection based on generalized gamma distribution. In: Proc. 2005 IEEE Internat. Conf. on Acoustics, Speech, and Signal Processing, Vol. 1, pp. 781-784.
61. Sohn J., Kim N.S., and Sung W. A statistical model based voice activity detection. IEEE Signal Process. Lett. 6 1 (1999) 1-3
62. Spiegelhalter D., Best N., Carlin B., and van der Linde A. Bayesian measures of model complexity and fit (with discussion). J. Roy. Statist. Soc. Ser. B (Statist. Methodol.) 64 4 (2002) 583-639
63. Stacy E. A generalization of the gamma distribution. Ann. Math. Statist. 33 (1962) 1187-1192
64. Sugiura N. Further analysis of the data by Akaike's information criterion and the finite corrections. Comm. Statist.: Theory Methods 7 1 (1978) 13-26
65. Sugiyama, M., Ogawa, H., 2001. Model selection with small samples. In: Proc. Artificial Neural Nets and Genetic Algorithms, April, pp. 418-421.
66. Takeuchi K. Distribution of informational statistics and a criterion of model fitting. Suri-Kagaku (Math. Sci.) 153 (1976) 12-18
67. Toledano D., and Gomez A. Automatic phonetic segmentation. IEEE Trans. Speech Audio Process. 11 6 (2003) 617-625
68. Tremblay M., and Wallach D. Comparison of parameter estimation methods for crop models. Agronomie 24 (2004) 351-365
69. Tritschler, A., Gopinath, R., 1999. Improved speaker segmentation and segments clustering using the Bayesian information criterion. In: Proc. 1999 European Speech Processing, Vol. 2, pp. 679-682.
70. Tsionas E. Exact inference in four-parameter generalized gamma distributions. Comm. Statist.: Theory Methods 30 4 (2001) 747-756
71. Van Allen, T., 2000. Handling uncertainty when you're handling uncertainty: model selection and error bars for belief networks. Master of Science thesis, Department of Computer Science, University of Alberta. Available from: .
72. Varga, A., Steeneken, H., Tomlinson, M., Jones, D., 1992. The NOISEX-92 study on the affect of additive noise on automatic speech recognition. Technical Report, DRA Speech Research Unit, Malvern, England.
73. Villiers, E., Preez, J.A., 2001. The advantage of using higher order HMM's for segmenting acoustic files. In: Proc. 12th Annual Symposium of the Pattern Recognition Association of South Africa, November, pp. 120-122.
74. Visser, I., Ray, S., Jang, W., Berger, J., 2005. Effective sample size and the Bayes factor. SAMSI LVSSS Workshop Transition Workshop: Latent Variable Models in the Social Sciences, November.
75. Volinsky C., and Raftery A. Bayesian information criterion for censored survival models. Biometrics 56 (2000) 256-262
76. Wang, H., Cheng, S., 2004. METRIC-SEQDAC: a hybrid approach for audio segmentation. In: Proc. Internat. Conf. on Spoken Language Processing (ICSLP2004), October, pp. 1617-1620.
77. Wang L., Zhao Y., Chu M., Soong F., Zhou J., and Cao Z. Context-dependent boundary model for refining boundaries segmentation of TTS units. IEICE Trans. Inform. Syst. E89-D 3 (2006) 1082-1091
78. Weakliem D. A critique of the Bayesian information criterion for model selection. Sociol. Methods Res. 27 3 (1999) 359-397
79. Woodland, P., Gales, M., Pye, D., Young, S., 1997. The development of the 1996 HTK broadcast news transcription system. In: Proc. Speech Recognition Workshop, pp. 73-78.
80. Zhang, S., Zhang, S., Xu, B., 2006. A two-level method for unsupervised speaker-based audio segmentation. In: Proc. 18th Internat. Conf. on Pattern Recognition (ICPR2006), pp. 298-301.
81. Zhao, Y., Wang, L., Chu, M., Soong, F., Cao, Z., 2005. Refining phoneme segmentations using speaker-adaptive context dependent boundary models. In: Proc. Interspeech, pp. 2557-2560.