Return to Research

EM Algorithms for Joint Source Separation and Diarisation of Speech

IEEE International Conference on Acoustics, Speech and Signal Processing, 2017
IEEE Workshop on Applications of Signal Processing to Audio Acoustics, 2017

D. Kounades-Bastian, L. Girin, X. Alameda-Pineda, S. Gannot, R. Horaud

In this page you can find two EM algorithms for simultaneous separation and diarision of multichannel convolutive audio mixtures. The algorithm on Icassp 2017 paper
[IEEXplore] [pdf] uses the Narrow-Band model to represent the multichannel mixture in the STFT domain. The algorithm on Waspaa 2017 paper [IEEXplore] [pdf] uses the Spatial Covariance model of Duong et. al. to represent the multichannel mixture.

Slides & Poster

slides from Icassp 2017 [pdf]
poster from Waspaa 2017 [pdf]


MATLAB for the Icassp algorithm [code].
MATLAB for the Waspaa algorithm [code].


EM notes on the Waspaa algorithm [pdf]