A two-step approach to blind deconvolution of speech and sound sources in the time domain

• Autorzy: Okazaki F.A. , Kasprzak W.
• Języki publikacji: EN

Abstrakty

EN

In order to understand commands given through voice by an operator, user or any human, a robot needs to focus on a single source, to acquire a clear speech sample and to recognize it. A two-step approach to the deconvolution of speech and sound mixtures in the time-domain is proposed. At first, we apply a deconvolution procedure, constrained in the sense, that the de-mixing matrix has fixed diagonal values without non-zero delay parameters. We derive an adaptive rule for the modification of the de-convolution matrix. Hence, the individual outputs extracted in the first step are eventually still self-convolved. This corruption we try to eliminate by a de-correlation process independently for every individual output channel.

Słowa kluczowe

EN

blind signal analysis; convolved mixtures; independent component analysis; robotic sensors; speech reconstruction

• Wydawca: Polska Akademia Nauk, Wydział IV Nauk Technicznych
• Czasopismo: Bulletin of the Polish Academy of Sciences. Technical Sciences
• Rocznik: 2005
• Tom: Vol. 53, nr 1
• Strony: 49--55
• Opis fizyczny: Bibliogr. 19 poz., 10 rys.

Twórcy

autor

Okazaki F.A.

autor

Kasprzak W.

• Institute of Control and Computation Engineering, Warsaw University of Technology, 15/19 Nowowiejska St., 00-665 Warszawa, Poland.,

Bibliografia

• [1] K. Tchon (ed.), VIII State Conference of Robotics, (Polanica Zdrój, Poland, June 2004), WKiŁ, Warszawa, (2005), (in Polish).
• [2] C. Zielinski, “A unified formal description of behavioural and deliberative robotic multi-agent systems”, Proc. 7th IFAC International Symposium on Robot Control SYROCO 2003, Wrocław, Poland, vol. 2, 479-486 (2003).
• [3] A. Cichocki and S. Amari, Adaptive Blind Signal and Image Processing, John Wiley, Chichester, UK, (2002).
• [4] A. Hyvarinen, J. Karhunen and E. Oja, Independent Component Analysis, John Wiley & Sons, New York etc., (2001).
• [5] N. Murata, S. Ikeda and A. Ziehe, “An approach to blind source separation based on temporal structure of speech signal”, Neurocomputing 41(4), 1–24 (2001).
• [6] W. Kasprzak, “Blind deconvolution of timely correlated sources by gradient descent search”, in: Image Processing and Communications, An International Journal, edited by ATR Bydgoszcz, 9(1), 33–52 (2003).
• [7] P. Smaragdis, “Blind separation of convolved mixtures in frequency domain”, Neurocomputing 22 (1), 21–34 (1998).
• [8] N. Araki et al., “Fundamental limitation of frequency domain blind source separation for convolved mixture of speech”, Proceedings ICASSP2001, 5, 2737–2740 (2001).
• [9] W. Kasprzak and A. Okazaki, “Blind deconvolution of timely-correlated sources by homomorphic filtering in Fourier space”, Fourth Int. Symposium on Independent Component Analysis and Blind Signal Separation - ICA’2003, Nara, Japan, 2003, NTT Comm. Science Lab., pp. 1029–1034 (2003).
• [10] H. Saruwatari, S. Kurita and K. Takeda, “Blind source separation combining frequency-domain ICA and beamforming”, Proceedings ICASSP2001, 5, 2733–2736 (2001).
• [11] T. Nishikawa, H. Saruwatari, K. Shikano, S. Araki and S. Makino, “Multistage ICA for blind source separation of real acoustic convolutive mixture”, Fourth Int. Symposium on Independent Component Analysis and Blind Signal Separation – ICA’2003, Nara, Japan, NTT Comm. Science Lab., pp. 523–528 (2003).
• [12] D.C.B. Chan, Blind Signal Separation, Ph.D. thesis, University of Cambridge, Engineering Department, Cambridge, UK, (1997).
• [13] W. Kasprzak, A. Cichocki and S. Amari, “Blind source separation with convolutive noise cancellation”, Neural Computing and Applications, Springer-Verlag London Ltd., vol. 6, 127–141 (1997).
• [14] A. Okazaki and W. Kasprzak, “Deconvolution of speech signals from their mixture under constant mixing matrix diagonal”, VIII State Conference of Robotics, (Polanica Zdrój, czerwiec 2004), WKiL, Warszawa, (2004), (in Polish).
• [15] I. Sabala, Multichannel Deconvolution and Separation of Statistically Independent Signals for Unknown Dynamic Systems, Ph.D. dissertation, Warsaw University of Technology, Dep. of Electrical Engineering, Warsaw, (1998).
• [16] M-Audio Inc., “M-Audio Delta Series 44 User’s Manual”, M-Audio, http://www.m-audio.com, (2000).
• [17] Shure Corporation, “Shure C608 user’s manual”, Shure Company, (2003).
• [18] F. Antonioli, “WWW page of company FASoft with program n-Track Studio”, http://www.fasoft.com
• [19] M. Burbeck, J. Haberman and D. Mazzoni, “WWW page of project Audacity”, http://audacity.sourceforge.net/.