Variational EM for binaural sound-source separation and localization

Deleforge A, Forbes F, Horaud R (2013)


Publication Language: English

Publication Type: Conference contribution

Publication year: 2013

Pages Range: 76-80

Event location: Vancouver CA

ISBN: 978-1-4799-0356-6

DOI: 10.1109/ICASSP.2013.6637612

Abstract

The sound-source separation and localization (SSL) problems are addressed within a unified formulation. Firstly, a mapping between white-noise source locations and binaural cues is estimated. Secondly, SSL is solved via Bayesian inversion of this mapping in the presence of multiple sparse-spectrum emitters (such as speech), noise and reverberations. We propose a variational EM algorithm which is described in detail together with initialization and convergence issues. Extensive real-data experiments show that the method outperforms the state-of-the-art both in separation and localization (azimuth and elevation).

Authors with CRIS profile

How to cite

APA:

Deleforge, A., Forbes, F., & Horaud, R. (2013). Variational EM for binaural sound-source separation and localization. In Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing (pp. 76-80). Vancouver, CA.

MLA:

Deleforge, Antoine, Florence Forbes, and Radu Horaud. "Variational EM for binaural sound-source separation and localization." Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing, Vancouver 2013. 76-80.

BibTeX: Download