Skip to Main content Skip to Navigation
Journal articles

Spatial location priors for Gaussian model based reverberant audio source separation

Ngoc Duong 1 Emmanuel Vincent 2, * Rémi Gribonval 3
* Corresponding author
2 PAROLE - Analysis, perception and recognition of speech
Inria Nancy - Grand Est, LORIA - NLPKD - Department of Natural Language Processing & Knowledge Discovery
3 PANAMA - Parcimonie et Nouveaux Algorithmes pour le Signal et la Modélisation Audio
Inria Rennes – Bretagne Atlantique , IRISA-D5 - SIGNAUX ET IMAGES NUMÉRIQUES, ROBOTIQUE
Abstract : We consider the Gaussian framework for reverberant audio source separation, where the sources are modeled in the time-frequency domain by their short-term power spectra and their spatial covariance matrices. We propose two alternative probabilistic priors over the spatial covariance matrices which are consistent with the theory of statistical room acoustics and we derive expectation-maximization algorithms for maximum a posteriori (MAP) estimation. We argue that these algorithms provide a statistically principled solution to the permutation problem and to the risk of overfitting resulting from conventional maximum likelihood (ML) estimation. We show experimentally that in a semi-informed scenario where the source positions and certain room characteristics are known, the MAP algorithms outperform their ML counterparts. This opens the way to rigorous statistical treatment of this family of models in other scenarios in the future.
Complete list of metadatas

Cited literature [36 references]  Display  Hide  Download

https://hal.inria.fr/hal-00870191
Contributor : Ed. Bmc <>
Submitted on : Sunday, October 6, 2013 - 5:10:31 AM
Last modification on : Friday, March 6, 2020 - 1:34:48 AM
Document(s) archivé(s) le : Friday, April 7, 2017 - 6:58:31 AM

Files

Identifiers

Citation

Ngoc Duong, Emmanuel Vincent, Rémi Gribonval. Spatial location priors for Gaussian model based reverberant audio source separation. EURASIP Journal on Advances in Signal Processing, SpringerOpen, 2013, 2013 (1), pp.149. ⟨10.1186/1687-6180-2013-149⟩. ⟨hal-00870191⟩

Share

Metrics

Record views

1368

Files downloads

812