Variational EM for Binaural Sound-Source Separation and Localization

Antoine Deleforge 1, * Florence Forbes 2 Radu Horaud 1
* Corresponding author
1 PERCEPTION - Interpretation and Modelling of Images and Videos
Inria Grenoble - Rhône-Alpes, LJK - Laboratoire Jean Kuntzmann, INPG - Institut National Polytechnique de Grenoble
2 MISTIS - Modelling and Inference of Complex and Structured Stochastic Systems
Inria Grenoble - Rhône-Alpes, LJK - Laboratoire Jean Kuntzmann, INPG - Institut National Polytechnique de Grenoble
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).
Document type :
Conference papers
Complete list of metadatas

Cited literature [20 references]  Display  Hide  Download
https://hal.inria.fr/hal-00823453
Contributor : Florence Forbes <>
Submitted on : Friday, May 17, 2013 - 8:56:33 AM
Last modification on : Wednesday, April 11, 2018 - 1:58:30 AM
Long-term archiving on: Sunday, August 18, 2013 - 4:13:27 AM

File

deleforge_icassp_rev.pdf
Files produced by the author(s)

Identifiers

Collections

CNRS | INRIA | INSMI | LJK_GI | LJK_GI_PERCEPTION | LJK_PS_MISTIS | LJK | LJK_PS | UGA

Citation

Antoine Deleforge, Florence Forbes, Radu Horaud. Variational EM for Binaural Sound-Source Separation and Localization. ICASSP 2013 - 38th International Conference on Acoustics, Speech, and Signal Processing, May 2013, Vancouver, Canada. pp.76-80, ⟨10.1109/ICASSP.2013.6637612⟩. ⟨hal-00823453⟩

Export

BibTeX TEI DC DCterms EndNote

Share

Metrics

Record views

787

Files downloads

788

 

 

Contact

archive-ouverte@inria.fr