Ron
Weiss
LabROSA, Columbia University
ronw@ee.columbia.edu
These results are joint work with Michael Mandel.

Results were obtained using the binaural source localization algorithm
described in the references below.  We construct a generative model of the
interaural phase difference and learn the parameters corresponding to each
source using an EM algorithm.  This algorithm does not as well on the signals
recorded using microphones with 5cm spacing because the IPDs for different
sources are not very pronounced.

The EM procedure.is used to generate a spectrographic mask for each source
where each cell contains the probability that a given source dominates the
mixture in that cell.  Stereo separation was performed simply by multiplying
the STFT of each channel of the mixture by the \"soft mask\" for each source and
inverting.  Monaural separation was performed by choosing the better ear for
the particular source (i.e. the one which contained the most energy after
applying the mask).

Our Matlab implementation of the algorithm took about 5 minutes per signal on
a 1.8 GHz Intel Xeon.  The only prior information needed was the number of
sources in the mixture.

References:

M. Mandel, D. Ellis, and T. Jebara. An EM algorithm for localizing
multiple sound sources in reverberant environments. NIPS, December
2006.