Unsupervised learning of time-frequency patches as a noise-robust representation of speech

Speech Communication

Volumn 51, Issue 11, 2009, Pages 1124-1138

  Van Segbroeck, Maarten ^a   Van hamme, Hugo ^a  

^a UNIVERSITY OF LEUVEN   (Belgium)

Author keywords

Acoustic signal analysis; Automatic speech recognition; Language acquisition; Matrix factorization; Noise robustness

Indexed keywords

ACOUSTIC SIGNAL ANALYSIS; AUTOMATIC SPEECH RECOGNITION; LANGUAGE ACQUISITION; MATRIX FACTORIZATION; NOISE ROBUSTNESS;

ACOUSTIC NOISE; ACOUSTIC SIGNAL PROCESSING; ACOUSTICS; BLIND SOURCE SEPARATION; CHEMICAL ACTIVATION; EDUCATION; FACTORIZATION; LEARNING ALGORITHMS; LEARNING SYSTEMS; LIGHT MEASUREMENT; LINGUISTICS; SIGNAL ANALYSIS;

SPEECH RECOGNITION;

EID: 67651030071     PISSN: 01676393     EISSN: None     Source Type: Journal    
DOI: 10.1016/j.specom.2009.05.003     Document Type: Article

References (36)

1. Auger F., Flandrin, and P. Improving the readability of time-frequency and time-scale representations by the reassignment method. IEEE Trans. Signal Process. 43 5 (1995) 1068-1089
2. Aversano, G., Esposito, A., Marinaro, M., 2001. A new text-independent method for phoneme segmentation. In: Proc. IEEE Internat. Workshop Circuits and Systems 2, 516-519.
3. Baker, J.M., Deng, L., Khudanpur, S., Lee, C.-H., Glass, J., Morgan, N., 2006-2007. Minds historical development and future directions in speech recognition and understanding. Tech. Rep., Report of the Speech Understanding Working Group. .
4. Brugnara F., Falavigna D., and Omologo M. Automatic segmentation and labeling of speech based on hidden markov models. Speech Commun. 12 4 (1993) 357-370
5. Chomsky N. New Horizons in the Study of Language and Mind (2000), Cambridge University Press, Cambridge, UK
6. Ding, C., Li, T., Jordan, M., 2006. Convex and semi-nonnegative matrix factorizations for clustering and low-dimension representation. Tech. Rep. LBNL-60428, Lawrence Berkeley National Laboratory, US.
7. ETSI Standard Document, 2000a. Distributed Speech Recognition; Front End Feature Extraction Algorithm; Compression Algorithm. ETSI ES 201 108 v1.1.2, April.
8. ETSI Standard Document, 2000b. Speech Processing, Transmission and Quality Aspects (STQ); Distributed Speech Recognition; Advanced Front-end Feature Extraction Algorithm; Compression Algorithm. ETSI ES 202 050 v1.1.1 (2002-10), April.
9. Ezzat, T., Bouvrie, J., Poggio, T., 2007. Spectro-temporal Analysis of Speech Using 2-D Gabor Filters. In: Proc. International Conference on Spoken Language Processing, Antwerp, Belgium, August, pp. 506-509.
10. Hainsworth, S., Macleod, M., 2003. Time-frequency reassignment: a review and analysis. Tech. Rep. CUED/FINFENG/TR.459, Cambridge University Engineering Department.
11. Hermansky, H., Fousek, P., 2005. Multi-resolution RASTA Filtering for TANDEM-based ASR. In: Proc. International Conference on Spoken Language Processing, Lisbon, Portugal, October, pp. 361-364.
12. Hermansky, H., Sharma, S., 1997. Temporal Patterns (TRAPs) in ASR of Noisy Speech. In: Proc. International Conference on Acoustics, Speech and Signal Processing, vol. 1. Phoenix, Arizona, March, pp. 289-292.
13. Hermansky, H., Sharma, S., 1998. TRAPS Classifiers of Temporal Patterns. In: Proc. International Conference on Spoken Language Processing, Sydney, Australia, November, pp. 1003-1006.
14. Hirsch, H.G., Pearce, D., 2000. The AURORA experimental framework for the performance evaluation of speech recognition systems under noisy conditions. In: Proc. ISCA ITRW ASR2000 Workshop, Paris, France, September, pp. 18-20.
15. Hoyer P. Non-negative matrix factorization with sparseness constraints. J. Machine Learn. Res. 5 (2004) 1457-1469
16. Kingsbury B., Morgan N., and Greenberg S. Robust speech recognition using the modulation spectrogram. Speech Commun. 25 (1998) 117-132
17. Kleinschmidt, M., 2003. Localized spectro-temporal features for automatic speech recognition. In: Proc. Eurospeech, Geneva, Switzerland, September, pp. 2573-2576.
18. Kodera K., Gendrin R., and Villedary C. Analysis of time-varying signals with small bt values. IEEE Trans. Audio Speech Language Process. 26 1 (1978) 64-76
19. Lee D., and Seung H. Algorithms for non-negative matrix factorization. Adv. Neural Inform. Process. Systems 13 (2001) 556-562
20. Macho, D., Mauuary, L., Noé, B., Cheng, Y., Ealey, D., Jouvet, D., Kelleher, H., Pearce, D., Saadoun, F., 2002. Evaluation of a Noise-robust DSR Front-end on Aurora Databases. In: Proc. International Conference on Spoken Language Processing, Denver, Colorado, USA, September, pp. 17-20.
21. Martin, A., Doddington, G., Kamm, T., Ordowski, M., Przybocki, M., 1997. The DET curve in assessment of detection task performance. In: Proc. Eurospeech, Rhodes, Greece, September, pp. 1895-1898.
22. Meyer, B.T., Kollmeier, B., 2008. Optimization and Evaluation of Gabor Feature Sets for ASR. In: Proc. International Conference on Spoken Language Processing, Brisbane, Australia, September, pp. 906-909.
23. O'Grady P.D., and Pearlmutter B.A. Discovering speech phones using convolutive non-negative matrix factorisation with a sparseness constraint. Neurocomputing 72 (2008) 88-101
24. Park, A., Glass, J., 2005. Towards unsupervised pattern discovery in speech. In: Proc. ASRU, San Juan, Puerto Rico, December, pp. 53-58.
25. Plante F., Meyer G., and Ainsworth W. Improvement of speech spectrogram accuracy by the method of reassignment. IEEE Trans. Speech Audio Process. 6 3 (1998) 282-286
26. Qiao, Y., Shimomura, N. Minematsu, N., 2008. Unsupervised Phoneme Segmentation Using Transformed Cepstrum Features. In: Proc. Spring Meeting of Acoust. Soc. Japan, 287-2901-11-20.
27. Scharenborg O., Norris D., ten Bosch L., and McQueen J.M. How should a speech recognizer work?. Cognitive Sci. 29 6 (2005) 867-918
28. Scharenborg, O., Ernestus, M., Wan, V., 2007. Segmentation of Speech: Child's Play? In: Proc. International Conference on Spoken Language Processing, Antwerp, Belgium, August, pp. 1953-1956.
29. Siivola, V., Hirsimäki, T., Creutz, M., Kurimo, M., 2003. Unlimited vocabulary speech recognition based on morphs discovered in an unsupervised manner. In: Proc. Eurospeech, pp. 2293-2296.
30. Smaragdis P. Convolutive speech bases and their application to speech separation. IEEE Trans. Speech Audio Process. 15 1 (2007) 1-12
31. Stouten V., Demuynck K., and Van hamme H. Discovering phone patterns in spoken utterances by nonnegative matrix factorisation. IEEE Signal Process. Lett. 15 (2008) 131-134
32. Tyagi, V., McCowan, I., Misra, H., Bourlard, H., 2003. Mel-cepstrum modulation spectrum (MCMS) features for robust ASR. In: Proc. ASRU 2003 Workshop, St. Thomas, Virgin Islands, December, pp. 399-404.
33. Van hamme, H., 2008a. HAC-models: A Novel Approach to Continuous Speech Recognition. In: Proc. International Conference on Spoken Language Processing, Brisbane, Australia, pp. 2554-2557.
34. Van hamme, H., 2008b. Integration of asynchronous knowledge sources in a novel speech recognition framework. In: ISCA ITRW Workshop on Speech Analysis and Processing for Knowledge Discovery, Aalborg, Danmark, June.
35. Virtanen T. Monaural sound source separation by nonnegative matrix factorization with temporal continuity and sparseness criteria. IEEE Trans. Audio Speech Language Process. 15 3 (2007) 1066-1074
36. Young S., Kershaw D., Odell J., Ollason D., Valtchev V., and Woodland P. The HTK Book - Version2.2 (1999), Entropic