Investigation of speech separation as a front-end for noise robust speech recognition

IEEE Transactions on Audio, Speech and Language Processing

Volumn 22, Issue 4, 2014, Pages 826-835

  Narayanan, Arun ^a   Wang, DeLiang ^a  

^a The Ohio State University   (United States)

Author keywords

Aurora 4; Deep neural networks; Feature mapping; Robust ASR; Time frequency masking

Indexed keywords

AURORA-4; DEEP NEURAL NETWORKS; FEATURE MAPPING; ROBUST ASR; TIME-FREQUENCY MASKING;

FUNCTIONS; NEURAL NETWORKS; SPEECH RECOGNITION; ADDITIVE NOISE; SEPARATION; SPEECH; SPEECH ANALYSIS;

SPEECH ANALYSIS; SPEECH RECOGNITION;

EID: 84898075497     PISSN: 15587916     EISSN: None     Source Type: Journal    
DOI: 10.1109/TASLP.2014.2305833     Document Type: Article

References (48)

1. R. Astudillo and R. Orglmeister, "Computing MMSE estimates and residual uncertainty directly in the feature domain of ASR using STFT domain speech distortion models," IEEE Trans. Audio, Speech, Lang. Process., vol. 21, no. 5, pp. 1023-1034, May 2013.
2. L. Bottou and O. Bousquet, "The tradeoffs of large-scale learning," Adv. Neural Inf. Process. Syst. 20, pp. 161-168, 2008.
3. C.-P. Chen and J. A. Bilmes, "MVA processing of speech features," IEEE Trans. Audio, Speech, Lang. Process., vol. 15, no. 1, pp. 257-270, Jan. 2007.
4. G. E. Dahl, T. N. Sainath, and G. Hinton, "Improving deep neural networks for LVCSR using rectified linear units and dropout," in Proc. IEEE ICASSP, 2013, pp. 8609-8613.
5. G. Dahl, D. Yu, L. Deng, and A. Acero, "Context-dependent pre-trained deep neural networks for large-vocabulary speech recognition," IEEE Trans. Audio, Speech, Lang. Process., vol. 20, no. 1, pp. 30-42, Mar. 2012.
6. M. Delcroix, Y. Kubo, T. Nakatani, and A. Nakamura, "Is speech enhancement pre-processing still relevant when using deep neural networks for acoustic modeling?," in Proc. Interspeech, 2013, pp. 2992-2996.
7. L. Deng, A. Acero, L. Jiang, J. Droppo, and X. Huang, "High-performance robust speech recognition using stereo training data," in Proc. IEEE ICASSP, 2001, pp. 301-304.
8. L. Deng, J. Droppo, and A. Acero, "Dynamic compensation of HMM variances using the feature enhancement uncertainty computed from a parametric model of speech distortion," IEEE Trans. Speech Audio Process., vol. 13, no. 3, pp. 412-421, May 2005.
9. J. Droppo, "Feature compensation," in Techniques for Noise Robustness in Automatic Speech Recognition, T. Virtanen, B. Raj, and R. Singh, Eds. West Sussex, U.K.: Wiley, 2012, ch. 9, pp. 229-250.
10. J. Du, Y. Hu, L.-R. Dai, and R.-H. Wang, "HMM-based pseudo-clean speech synthesis for splice algorithm," in Proc. IEEE ICASSP, 2010, pp. 4570-4573.
11. J. Duchi, E. Hazan, and Y. Singer, "Adaptive subgradient methods for online learning and stochastic optimization," J. Mach. Learn. Res., vol. 12, pp. 2121-2159, 2010.
12. D. P. W. Ellis, "PLP and RASTA (and MFCC, and inversion) in Matlab," 2005 [Online]. Available: http://www.ee.columbia.edu/dpwe/ resources/matlab/rastamat/
13. Speech processing transmission and quality aspects (STQ); Distributed speech recognition; Advanced front-end feature extraction algorithm; Compression algorithms, ES 202 050 V1.1.4, ETSI, 2005.
14. F. Flego and M. J. F. Gales, "Discriminative adaptive training with VTS and JUD," in Proc. IEEE ASRU, 2009, pp. 170-175.
15. M. J. F. Gales, "Maximum likelihood linear transformations for HMM-based speech recognition," Comput. Speech Lang., vol. 12, no. 2, pp. 75-98, 1998.
16. I. J. Goodfellow, D. Warde-Farley, M. Mirza, A. Courville, and Y. Bengio, "Maxout networks," J. Mach. Learn. Res. Workshop Conf. Proc., vol. 28, no. 3, pp. 1319-1327, 2013.
17. K. Han and D. L. Wang, "A classification based approach to speech segregation," J. Acoust. Soc. Amer., vol. 132, no. 5, pp. 3475-3483, 2012.
18. W. Hartmann, A. Narayanan, E. Fosler-Lussier, and D. L. Wang, "A direct masking approach to robust ASR," IEEE Trans. Audio, Speech, Lang. Process., vol. 21, no. 10, pp. 1993-2005, Oct. 2013.
19. H. Hermansky and N. Morgan, "RASTA processing of speech," IEEE Trans. Speech Audio Process., vol. 2, no. 4, pp. 578-589, Oct. 1994.
20. G. E. Hinton, N. Srivastava, A. Krizhevsky, I. Sutskever, and R. Salakhutdinov, "Improving neural networks by preventing co-adaptation of feature detectors," arXiv preprint arXiv:1207.0580, 2012.
21. G. Hinton, S. Osindero, and Y. Teh, "A fast learning algorithm for deep belief nets," Neural Comput., vol. 18, no. 7, pp. 1527-1554, 2006.
22. G. Kim, Y. Lu, Y. Hu, and P. Loizou, "An algorithm that improves speech intelligibility in noise for normal-hearing listeners," J. Acoust. Soc. Amer., vol. 126, no. 3, pp. 1486-1494, 2009.
23. W. Kim and R. Stern, "Mask classifcation for missing-feature reconstruction for robust speech recognition in unknown background noise," Speech Commun., vol. 53, pp. 1-11, 2011.
24. A. Krizhevsky, I. Sutskever, and G. E. Hinton, "Imagenet classification with deep convolutional neural networks," Adv. Neural Inf. Process. Syst., vol. 25, pp. 1106-1114, 2012.
25. A. L. Maas, Q. V. Le, T. M. O'Neil, O. Vinyals, P. Nguyen, and A. Y. Ng, "Recurrent neural networks for noise reduction in robust ASR," in Proc. Interspeech, 2012.
26. A. Mohamed, G. Dahl, and G. Hinton, "Acoustic modeling using deep belief networks," IEEE Trans. Audio, Speech, Lang. Process., vol. 20, no. 1, pp. 14-22, Jan. 2012.
27. V. Nair and G. E. Hinton, "Rectified linear units improve restricted Boltzmann machines," in Proc. ICML 27, 2010, pp. 807-814.
28. A. Narayanan and D. Wang, "Coupling binary masking and robust ASR," in Proc. IEEE ICASSP, 2013, pp. 6817-6821.
29. A. Narayanan and D. Wang, "Ideal ratio mask estimation using deep neural networks for robust speech recognition," in Proc. IEEE ICASSP, 2013, pp. 7092-7096.
30. N. Parihar and J. Picone, "Analysis of the Aurora large vocabulary evaluations," in Proc. Eurospeech, 2003, pp. 337-340.
31. R. Prabhavalkar, T. N. Sainath, D. Nahamoo, B. Ramabhadran, and D. Kanevsky, "An evaluation of posterior modeling techniques for phonetic recognition," in Proc. IEEE ICASSP, 2013, pp. 7165-7169.
32. B. Raj and R. Stern, "Missing-feature approaches in speech recognition," IEEE Signal Process. Mag., vol. 22, no. 5, pp. 101-116, 2005.
33. N. Roman, D. L. Wang, and G. J. Brown, "Speech segregation based on sound localization," J. Acoust. Soc. Amer., vol. 114, no. 4, pp. 2236-2252, 2003.
34. N. Roman and J. Woodruff, "Intelligibility of reverberant noisy speech with ideal binary masking," J. Acoust. Soc. Amer., vol. 130, no. 4, pp. 2153-2161, 2011.
35. F. Seide, G. Li, X. Chen, and D. Yu, "Feature engineering in context-dependent deep neural networks for conversational speech transcription," in Proc. IEEE ASRU, 2011, pp. 24-29.
36. M. L. Seltzer, B. Raj, and R. M. Stern, "A Bayesian classifier for spectrographic mask estimation for missing feature speech recognition," Speech Commun., vol. 43, no. 4, pp. 379-393, 2004.
37. M. L. Seltzer, D. Yu, and Y.-Q. Wang, "An investigation of deep neural networks for noise robust speech recognition," in Proc. IEEE ICASSP, 2013, pp. 7398-7402.
38. S. Srinivasan, N. Roman, and D. L. Wang, "Binary and ratio time-frequency masks for robust speech recognition," Speech Commun., vol. 48, pp. 1486-1501, 2006.
39. K. Vertanen, "HTKWall Street Journal training recipe," 2005 [Online]. Available: http://www.keithv.com/software/htk/
40. D. L. Wang, "On ideal binary masks as the computational goal of auditory scene analysis," in Speech Separation by Humans and Machines, P. Divenyi, Ed. Boston, MA, USA: Kluwer, 2005, pp. 181-197.
41. D. L. Wang, U. Kjems, M. S. Pedersen, J. B. Boldt, and T. Lunner, "Speech intelligibility in background noise with ideal binary time-frequency masking," J. Acoust. Soc. Amer., vol. 125, pp. 2336-2347, 2009.
42. Y. Wang, K. Han, and D. L. Wang, "Exploring monaural features for classification-based speech segregation," IEEE Trans. Audio, Speech, Lang. Process., vol. 21, pp. 270-279, 2013.
43. Y. Wang and D. L. Wang, "Towards scaling up classification-based speech separation," IEEE Trans. Audio, Speech, Lang. Process., vol. 21, no. 7, pp. 1381-1390, Jul. 2013.
44. Y. Wang and D. L. Wang, "Feature denoising for speech separation in unknown noisy environments," in Proc. IEEE ICASSP, 2013, pp. 7472-7476.
45. Y.-Q. Wang and M. J. F. Gales, "Speaker and noise factorization for robust speech recognition," IEEE Trans. Audio, Speech, Lang. Process., vol. 20, no. 7, pp. 2149-2158, 2012.
46. F. Weninger, J. Geiger, M. Wöllmer, B. Schuller, and G. Rigoll, "The Munich feature enhancement approach to the 2nd CHiME challenge using BLSTM recurrent neural networks," in Proc. 2nd CHiME Mach. Listen. Multisource Environ. Workshop, 2013, pp. 86-90.
47. S. Young, G. Evermann, T. Hain, D. Kershaw, G. Moore, J. Odell, D. Ollason, D. Povey, V. Valtchev, and P. Woodland, The HTK Book. Cambridge, U.K.: Cambridge Univ. Press, 2002 [Online]. Available: http://htk.eng.cam.ac.uk.
48. D. Yu, M. L. Seltzer, J. Li, J.-T. Huang, and F. Seide, "Feature learning in deep neural networks - studies on speech recognition tasks," in Proc. ICLR, 2013.