Application of independent component analysis for speech–music separation using an efficient score function estimation

Arash Pishravian; Masoud Reza Aghabozorgi Sahaf

doi:10.2478/v10187-012-0056-0

Abstract

In this paper speech-music separation using Blind Source Separation is discussed. The separating algorithm is based on the mutual information minimization where the natural gradient algorithm is used for minimization. In order to do that, score function estimation from observation signals (combination of speech and music) samples is needed. The accuracy and the speed of the mentioned estimation will affect on the quality of the separated signals and the processing time of the algorithm. The score function estimation in the presented algorithm is based on Gaussian mixture based kernel density estimation method. The experimental results of the presented algorithm on the speech-music separation and comparing to the separating algorithm which is based on the Minimum Mean Square Error estimator, indicate that it can cause better performance and less processing time

References

[1] PANAGIOTAKIS, C.-TZIRITAS, G. : Speech/Music Discriminator based on RMS and Zero-Crossing, IEEE Transactions on Multimedia 7 No. 2 (2005), 155-166.10.1109/TMM.2004.840604
Search in Google Scholar
[2] PINQUIER, J.-S´ENAC, C.-ORBRECHT, R. : Speech and Music Classification in Audio Documents, In Proceedings of ICASSP’2002,, vol. 4, USA, May 2002.10.1109/ICASSP.2002.1004854
Search in Google Scholar
[3] SCHEIRER, E.-SLANEY, M. : Construction and Evaluation of a Robust Multifeature Speech/Music Discriminator, In Proceedings of ICASSP’97, Munich, Germany,, Apr 1997, pp. 1331-1334.
Search in Google Scholar
[4] CAREY, M. J.-PARRIS, E. S.-LLOYD-THOMAS, H. : A Comparison of Features for Speech- Music Discrimination, In Proceedings of ICASSP’99, vol. 1, USA,March 1999, pp.149-152.10.1109/ICASSP.1999.758084
Search in Google Scholar
[5] JUTTEN, C.-HERAULT, J. : Blind Separation of Sources I. an Adaptive Algorithm based on Neuromimetric Architecture, Signal Processing 24 No. 1 (1991), 1-10.
Search in Google Scholar
[6] COMON, P. : Independent Component Analysis, a New Concept?, Signal Processing 36 No. 3 (1994), 287-314.10.1016/0165-1684(94)90029-9
Search in Google Scholar
[7] CARDOSO, J. F. : Blind Signal Separation: Statistical Principles, Proceedings of the IEEE 9 No. 10 (1998), 2009-2025.10.1109/5.720250
Search in Google Scholar
[8] HYV¨ARINEN, A.-OJA, E. : Independent Component Analysis: Algorithms and Applications, Neural Networks 13 No. 4 (2000), 411-430.10.1016/S0893-6080(00)00026-5
Search in Google Scholar
[9] CHOI, S.-CICHOCKI, A.-PARK, H.-M.-LEE, S.-Y. : Blind Source Separation and Independent Component Analysis: a Review, Neural Information Processing - Letters and Reviews 6 No. 1 (2005), 1-57.
Search in Google Scholar
[10] PEDERSEN, M. S.-LARSEN, J.-KJEMS, U.-PARRA, L. C. : A Survey of Convolutive Blind Source Separation Methods, Springer Handbook on Speech Processing and Speech Communication, Springer, 2007, pp. 1-34.10.1007/978-3-540-49127-9_52
Search in Google Scholar
[11] CARDOSO, J.-F. : Source Separation using Higher Order Moments, Proceeding ICASSP, 1989, pp. 2109-2112.
Search in Google Scholar
[12] MANSOUR, A.-JUTTEN, C. : Fourth Order Criteria for Blind Separation of Sources, IEEE Trans. on Signal Processing 43 No. 8 (1995), 2022-2025.10.1109/78.403370
Search in Google Scholar
[13] BABAIE-ZADE, M.-JUTTEN, C.-NAYEBI, K. : Differential of Mutual Information Function, IEEE Signal Processing Letters 11 No. 1 (2004), 48-51.10.1109/LSP.2003.819344
Search in Google Scholar
[14] SAMADI, S.-BABAIE-ZADE, M.-JUTTEN, C.-NAYEBI, K. : Blind Source Separation by Adaptive Estimation of Score Function Difference, Proceedings of ICA 2004, Granada, Spain, 2004, pp. 9-17.10.1007/978-3-540-30110-3_2
Search in Google Scholar
[15] VLASSIS, N.-MOTOMURA, Y. : Efficient Source Adaptivity in Independent Component Analysis, IEEE Transaction on Neural Networks 12 No. 3 (2001), 559-566.10.1109/72.92555818249888
Search in Google Scholar
[16] BABAIE-ZADE, M.-JUTTEN, C. : A General Approach for Mutual Information Minimization and its Application to Blind Source Separation, Signal Processing 85 No. 5 (2005), 975-995.10.1016/j.sigpro.2004.11.021
Search in Google Scholar
[17] AMARI, S.-CICHOCKI, A.-YANG, H. H. : A New Learning Algorithm for Blind Signal Separation, Advances in Neural Information Processing Systems (1996,), 757-763, The MIT Press.
Search in Google Scholar

Application of independent component analysis for speech–music separation using an efficient score function estimation

Abstract

Paradigm

My account