Chair of System Theory and Signal Processing, University of Stutgart
benedikt.loesch@lss.uni-stuttgart.de

My algorithm is based on TF sparseness and uses a blind beamformer approach.  
It is explained in a paper submitted to LVA/ICA 2010:
[1] B. Loesch, and B. Yang: "Blind Source Separation based on Time-Frequency 
Sparseness in the Presence of Spatial Aliasing", submitted to LVA/ICA 2010

The running time for each excerpt is about 1 second on a single core of 
Intel(R) Xeon(R) CPU E5440 @ 2.83GHz.