-----------------------------------------------------------------------

 [Contact information]
  Shoko Araki, NTT Communication Science Labs., NTT Corporation
  shoko@cslab.kecl.ntt.co.jp

 [Algorithm]
 - Method:  Dereverberation + Maximum SNR beamformer 

 - Running time: 
     2 cm: 370 secs
     6 cm: 350 secs
     10cm: 385 secs
   The algorithm was implemented in MATLAB and run on a Linux workstation
   equipped with a 2.4GHz processor.

 - References: 
    Dereverberation part of 
       [1] T. Yoshioka, T. Nakatani, M. Miyoshi, and H. G. Okuno, 
       "Blind separation and dereverberation of speech mixtures by
       joint optimization," accepted for publication in IEEE Transactions on
       Audio, Speech, and Language Processing, Jan. 2010. 
       (available on IEEE Xplore 
       http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=5428853&tag=1).
   and the Maximum SNR beamformer in 
       [2] S. Araki, H. Sawada and S. Makino, "Blind Speech Separation in a 
       Meeting Situation," ICASSP2007, vol. I, pp. 41--45, 2007.

 -----------------------------------------------------------------------