Evaluation of a Multi-Resolution Dyadic Wavelet Transform Method for usable Speech Detection

Authors: Wajdi Ghezaiel, Amel Ben Slimane Rahmouni, Ezzedine Ben Braiek

Abstract:

Many applications of speech communication and speaker identification suffer from the problem of co-channel speech. This paper deals with a multi-resolution dyadic wavelet transform method for usable segments of co-channel speech detection that could be processed by a speaker identification system. Evaluation of this method is performed on TIMIT database referring to the Target to Interferer Ratio measure. Co-channel speech is constructed by mixing all possible gender speakers. Results do not show much difference for different mixtures. For the overall mixtures 95.76% of usable speech is correctly detected with false alarms of 29.65%.

Keywords: Co-channel speech, usable speech, multi-resolutionanalysis, speaker identification

Digital Object Identifier (DOI): doi.org/10.5281/zenodo.1074361

References:

[1] Morgan, P., George, E., Lee, T. and Kay, M., "Co-Channel Speaker Separation", ICASSP, part 1, vol.1, pages: 828-831, May, 1995.
[2] Benincasa, D. S. and Savic, M. I., "Co-channel speaker separation using constrained nonlinear optimization," Proc. IEEE ICASSP, pp: 1195- 1198, 1997.
[1] Meyer, G. F., Plante, F., and Bethommier, "Segregation of concurrent speech with the reassignment spectrum," Proc. IEEE ICASSP, pp: 1203- 1206, 1997.
[2] Savic, M., Gao, H. and Sorensen, J. S., "Co-channel speaker separation based on maximum-likelihood deconvolution,", IEEE ICASSP, pp:I-25- I-28, 1994.
[3] Iyer, A. N., Smolenski, B. Y., Yantorno, R. E., Cupples, J., Wenndth, S., "Speaker Identification Improvement Using Usable Speech Concept",EUSIPCO 2004
[4] S. Khanwalkar, Y. Smolenski, Robert E. Yantorno and S. J. Wenndt "Enhancement of speaker identification using SID usable speech", EUSIPCO 2005.
[5] Yantorno, R. E, "Method for improving speaker identification by determining usable speech" J.ACOUST.SOC.AM. Volume 124, issue 5, November 2008.
[6] R.Saeidi, P.Mowlaee, T.Kinnunen, Tan Zheng-Hua, M.G. Christensen, S.H Jensen, P.Fränti, "Signal to signal ratio independent speaker identification for co-channel speech signal" IEEE ICPR 2010.
[7] A. Kizhanatham and R. E. Yantorno, "Peak Difference Autocorrelation of Wavelet Transform Algorithm Based Usable Speech Measure", 7th World Multi-conference on Systemic, Cybernetics, and Informatics, Aug 2003
[8] J. M. Lovekin, K. R. Krishnamachari, and R. E. Yantorno, "Adjacent pitch period comparison (appc) as a usability measure of speech segments under co-channel conditions," IEEE International Symposium on Intelligent Signal Processing and Communication Systems, pp. 139- 142, Nov 2001.
[9] N. Chandra and R. E. Yantorno, "Usable speech detection using modified spectral autocorrelation peak to valley ration using the lpc residual," 4th IASTED International Conference Signal and Image Processing, pp. 146-150, 2002.
[10] N. Sundaram, A. N. Iyer, B. Y. Smolenski, and R. E. Yantorno, "Usable speech detection using linear predictive analysis - a model-based approach," IEEE International Symposium on Intelligent Signal Processing and Communication Systems, ISPACS, 2003.
[11] A. N. Iyer, M. Gleiter, B. Y. Smolenski, and R. E. Yantorno, "Structural usable speech measure using lpc residual," IEEE International Symposium on Intelligent Signal Processing and Communication Systems ,ISPACS, 2003.
[12] W.Ghezaiel, A.Ben Slimane, and E.Ben Braiek, "Usable speech detection for speaker identification system under co-channel conditions", JTEA 2010 Tunisia.
[13] Gonzalez, N. and Docampo, D., "Application of Singularity Detection with Wavelets for Pitch Estimation of Speech Signals", Proc. EUSIPCO, pp: 1657-1660, 1994.
[14] Janer, L. "New Pitch Detection Algorithm Based on Wavelet Transform", IEEE-Signal Processing, pp: 165-168, 1998.
[15] J.I. Agbinya "Discrete wavelet transforms techniques in speech processing" IEEE TENCON Digital Signal processing Application.
[16] Mallat SA, "Theory for MuItiresolution Signal Decomposition: The Wavelet Representation", IEEE Transactions on Pattern Analysis Machine Intelligence. Vol. 31, pp 674-693, 1989.
[17] Shi-huang Chen and Jhing-fa Wang "A pyramid structured wavelet algorithm for detecting pitch period of speech signal" 1998 international computer symposium.