Average computational time: 8.02 sec for test and 8.46 sec for test2 per excerpt, using Intel Core i7 965 (3.2GHz QuadCore), Matlab with Parallel Computing Toolbox

Needed prior information: the number of source signals

Reference: H. Sawada, S. Araki, S. Makino, "A Two-Stage Frequency-Domain Blind Source Separation Method for Underdetermined Convolutive Mixtures," IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (WASPAA 2007), pp. 139-142, Oct. 2007.  (An extended version has been submitted to IEEE Trans. Audio, Speech, and Language Processing.  It was conditionally accepted and its revised version is currently being reviewed.)