Zbynek Koldovsky (1), Petr Tichavsky (2), and Jiri Malek (1)

(1)
Faculty of Mechatronic and Interdisciplinary Studies
Technical University of Liberec, Studentska 2, 461 17 Liberec,
Czech Republic
{jiri . malek, zbynek . koldovsky} (at) tul . cz

(2)
Institute of Information Theory and Automation,
Pod vodarenskou vezi 4,
P.O.Box 18, 182 08 Praha 8, Czech Republic
tichavsk (at) utia . cas . cz

The separation was performed by the subband version of the T-ABCD algorithm described 
in [1]. The method utilizes the whole second of recordings and processes signals in four 
subbands of the same bandwidth in parallel. The other parameters were L=10 (the 
length of separating filters within subbands) and alpha=2 (the weighting parameter).

The processing is linear, therefore, Task 1 is considered only.

The method was running in Matlab.

Average running time per test: 0.45s
Total running time per GHz of CPU: 31.0s


[1] Zbynek Koldovsky, Petr Tichavsky, and Jiri Malek: Subband Blind 
Audio Source Separation Using a Time-Domain Algorithm and Tree-Structured 
QMF Filter Bank, submitted to LVA/ICA 2010, 2010.