Time-domain Blind Audio Source Separation Method Producing Separating Filters of Generalized Feedforward Structure

0348111 - ÚTIA 2011 RIV DE eng C - Conference Paper (international conference)
Koldovský, Zbyněk - Tichavský, Petr - Málek, J.
Time-domain Blind Audio Source Separation Method Producing Separating Filters of Generalized Feedforward Structure.
Lecture Notes in Computer Science. Vol. 6365. Heidelberg: Springer-Verlag, 2010 - (Gavrilova, M.; Kumar, V.; Mun, Y.; Tan, C.; Gervasi, O.), s. 17-24. ISBN 978-3-642-15994-7.
[Latent Variable Analysis and Signal Separation. St. Malo (FR), 27.09.2010-30.09.2010]
R&D Projects: GA MŠMT 1M0572; GA ČR GA102/09/1278
Institutional research plan: CEZ:AV0Z10750506
Keywords : blind source separation * audio * convolutive mixture
Subject RIV: BB - Applied Statistics, Operational Research
http://library.utia.cas.cz/separaty/2010/SI/tichavsky-time-domain blind audio source separation method producing separating filters of generalized feedforward structure.pdf

Time-domain methods for blind separation of audio signals are preferred due to their lower demand for available data and the avoidance of the permutation problem. However, their computational demands increase rapidly with the length of separating filters due to the simultaneous growth of the dimension of an observation space. We propose, in this paper, a general framework that allows the time-domain methods to compute separating filters of theoretically infinite length without increasing the dimension. Based on this framework, we derive a generalized version of the time-domain method of Koldovsky and Tichavsky (2008). For instance, it is demonstrated that its performance might be improved by 4dB of SIR using the Laguerre filter bank.
Permanent Link: http://hdl.handle.net/11104/0188723