Front-end compensation methods for LVCSR under Lombard effect

Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH

Volumn , Issue , 2011, Pages 1257-1260

  Bořil, Hynek ^a   Grézl, František ^b   Hansen, John H L ^a  

^a The University of Texas at Dallas   (United States)

^b BRNO UNIVERSITY OF TECHNOLOGY   (Czech Republic)

Author keywords

Bottleneck features; Histogram equalization; Lombard effect; Quantile based cepstral distribution normalization; Speech recognition; UT Scope database

Indexed keywords

BACKGROUND NOISE; BACKGROUND VARIATION; BAND PASS FILTERING; BOTTLENECK FEATURES; CEPSTRAL; CEPSTRAL MEAN SUBTRACTION; COMPENSATION METHOD; CRITICAL BANDS; FEATURE DISTRIBUTION; HISTOGRAM EQUALIZATIONS; LARGE VOCABULARY CONTINUOUS SPEECH RECOGNITION; LOMBARD EFFECT; TRAINING FEATURES; TRANSIENT EFFECT;

CONTINUOUS SPEECH RECOGNITION; FEATURE EXTRACTION; GRAPHIC METHODS; SIGNAL TO NOISE RATIO; SPEECH RECOGNITION;

NORMAL DISTRIBUTION;

EID: 84865772156     PISSN: None     EISSN: 19909772     Source Type: Conference Proceeding    
DOI: None     Document Type: Conference Paper

References (26)

1. J.-C. Junqua, "The Lombard reflex and its role on human listeners and automatic speech recognizers," JASA, vol. 93, no. 1, pp. 510-524, 1993.
2. J. H. L. Hansen, "Analysis and compensation of speech under stress and noise for environmental robustness in speech recognition," Speech Comm., vol. 20, no. 1-2, pp. 151-173, 1996.
3. H. Bořil, "Robust speech recognition: Analysis and equalization of Lombard effect in Czech corpora," Ph.D. dissertation, CTU in Prague, Czech Rep., http://www.utdallas.edu/~hynek, 2008.
4. Y. Lu and M. Cooke, "Speech production modifications produced by competing talkers, babble and stationary noise," JASA, vol. 124, no. 5, pp. 3261-3275, 2008.
5. M. Garnier, "Communication in noisy environments: From adaptation to vocal straining," Ph.D. dissertation, Univ. of Paris VI, France, 2007.
6. H. Bořil and J. H. L. Hansen, "Unsupervised equalization of Lombard effect for speech recognition in noisy adverse environments," IEEE Trans. on ASLP, vol. 18, no. 6, pp. 1379-1393, August 2010.
7. S. E. Bou-Ghazale and J. H. L. Hansen, "A comparative study of traditional and newly proposed features for recognition of speech under stress," IEEE Trans. on SAP, vol. 8, no. 4, pp. 429-442, 2000.
8. H. Bořil, P. Fousek, and P. Pollák, "Data-driven design of front-end filter bank for Lombard speech recognition," in Proc. ICSLP'06, Pittsburgh, Pennsylvania, 2006, pp. 381-384.
9. H. Bořil and J. H. L. Hansen, "UT-Scope: Towards LVCSR under Lombard effect induced by varying types and levels of noisy background," in Proc. IEEE ICASSP'11, Prague, Czech, 2011, pp. 4472-4475.
10. H. Hermansky and N. Morgan, "RASTA processing of speech," IEEE Transactions on SAP, vol. 2, no. 4, pp. 578 -589, Oct. 1994.
11. S. B. Davis and P. Mermelstein, "Comparison of parametric representations for monosyllabic word recognition in continuously spoken sentences," IEEE Trans. ASSP, vol. 28, no. 4, pp. 357-366, 1980.
12. H. Hermansky, "Perceptual linear predictive (PLP) analysis of speech," JASA, vol. 87, no. 4, pp. 1738-1752, 1990.
13. U. H. Yapanel and J. H. L. Hansen, "A new perceptually motivated MVDR-based acoustic front-end (PMVDR) for robust automatic speech recognition," Speech Commun., vol. 50, no. 2, pp. 142-152, 2008.
14. F. Grézl et al., "Probabilistic and bottle-neck features for LVCSR of meetings," in Proc. IEEE ICASSP'07, Apr 2007, pp. 757-760.
15. S. Yoshizawa, N. Hayasaka, N. Wada, and Y. Miyanaga, "Cepstral gain normalization for noise robust speech recognition," in Proc. IEEE ICASSP'04, vol. 1, May 2004, pp. 209-212.
16. J. Pelecanos and S. Sridharan, "Feature warping for robust speaker verification," in In ODYSSEY-2001, Crete, Greece, 2001, pp. 213-218.
17. S. Dharanipragada and M. Padmanabha, "A nonlinear unsupervised adaptation technique for speech recognition," in ICSLP, 2000, pp. 556-559.
18. H. Hermansky et al., "Tandem connectionist feature extraction for conventional HMM systems," in Proc. IEEE ICASSP'00, 2000.
19. B. Chen, Q. Zhu, and N. Morgan, "Learning long-term temporal features in LVCSR using neural networks," in Proc. ICSLP'04, 2004.
20. C. Benitz et al., "Robust ASR front-end using spectral-based and discriminant features: experiments on the AURORA task," in Proc. Eurospeech' 01, Aalborg, Denmark, Sept. 2001.
21. F. Grézl and J. Černocký, "Trap-based techniques for recognition of noisy speech," Lecture Notes in Comp. Science, vol. 2007, no. 9, pp. 270-277.
22. F. Grézl and P. Fousek, "Optimizing bottle-neck features for LVCSR," in Proc. IEEE ICASSP'08, Las Vegas, NV, April 2008, pp. 4729-4732.
23. R. A. Gopinath, "Maximum likelihood modeling with gaussian distributions for classification," in Proc. IEEE ICASSP'98, 1998, pp. 661-664.
24. J. H. L. Hansen and V. Varadarajan, "Analysis and compensation of Lombard speech across noise type and levels with application to in-set/out-of-set speaker recognition," IEEE Trans. ASLP, vol. 17, no. 2, pp. 366 -378, Feb. 2009.
25. V. Zue, S. Seneff, and J. Glass, "Speech database development at MIT: TIMIT and beyond," Speech Comm., vol. 9, no. 4, pp. 351 - 356, 1990.
26. J. Volín, R. Skarnitzl, and P. Pollák, "Confronting HMM-based phone labelling with human evaluation of speech production," in Proc. of INTERSPEECH' 05, Lisbon, Portugal, Sept. 2005, pp. 1541-1544.