Abstract

Usable speech criteria are proposed to extract minimally corrupted speech for speaker identification (SID) in co-channel speech. In co-channel speech, either speaker can randomly appear as the stronger speaker or the weaker one at a time. Hence, the extracted usable segments are separated in time and need to be organized into speaker streams for SID. In this paper, we focus to organize extracted usable speech segment into a single stream for the same speaker by speaker assignment system. For this, we develop model-based speaker assignment method based on posterior probability and exhaustive search algorithm. Evaluation of this method is performed on TIMIT database. The system is evaluated on co-channel speech and results show a significant improvement.

References

• Quatieri and Danisewicz, An approach to co-channel talker interference suppression using a sinusoidal model for speech, IEEE Transactions on Acoustics, Speech and Signal Processing – 1990 – Vol. 38 – P. 56-69.
• Morgan, D. P. , George, E. B. , Lee, L. T. , and Kay, S. M. Cochannel speaker separation by harmonic enhancement and suppression, IEEE Transactions on Speech and Audio Processing – 1997 – Vol. 5 – P. 407-424.
• Zissman, M. A. , and Seward, D. C. Two-talker pitch tracking for co-channel talker interference suppression, Technical Report, MIT Lincoln Laboratory –1992.
• Carlson, B. A. , and Clements, M. A. A computationally compact divergence measure for speech processing. IEEE Transactions on Pattern Analysis Machine Intelligenc,-1990 – Vol. 13 – P. 1-6.
• Kullback, S. Information theory and statistics. New York: Dover Publications 1968.
• Yu, G. , and Gish, H. Identification of speakers engaged in dialog. Proceedings of IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP) 1993 – P. 383-386.
• Dunn, R. B. , Reynolds, D. A. , and Quatieri, T. F. Approaches to speaker detection and tracking in conversational speech. Digital Signal Processin – 2000 Vol. 10 – P. 93-112.
• J. Lovekin, R. E. Yantorno, S. Benincasa, S. Wenndt, M. Huggins, Developing usable speech criteria for speaker identification Proc. ICASSP 2001– P. 421-424.
• Yantorno, R. E, Method for improving speaker identification by determining usable speech J. ACOUST. SOC. AM. , 2008 – Vol. 124, issue 5.
• J. M. Lovekin, K. R. Krishnamachari, and R. E. Yantorno, Adjacent pitch period comparison as a usability measure of speech segments under co-channel conditions, IEEE International Symposium on Intelligent Signal Processing and Communication Systems, 2001 – P. 139-142.
• Shao, Y. , and Wang, D. L. Model-based sequential organization in co-channel speech. IEEE Transactions on Audio, Speech, and Language Processing, 2006, – Vol. 14, – P. 289-298.
• A. Kizhanatham, R. E. Yantorno, Peak Difference Autocorrelation of Wavelet Transform Algorithm Based Usable Speech Measure, 7th World Multi-conference on Systemic, Cybernetics, and Informatics, 2003.
• W. Ghezaiel, A. Ben Slimane, E. Ben Braiek. Usable speech detection for speaker identification system under co-channel conditions, JTEA 2010 Tunisia.
• Wajdi GHEZAIEL, Amel BEN SLIMANE, Ezzedine BEN BRAIEK. Evaluation of a multi-resolution dyadic wavelet transform method for usable speech detection, waset journal. 2011 – Vol. 79 – P. 829-833.
• D. A. Reynolds, Speaker identification and verification using Gaussian mixture speaker models, Speech Commun. , 1995 – Vol 17 – P. 91–108,