Multichannel Audio Modeling with Elliptically Stable Tensor Decomposition

Mathieu Fontaine 1 Fabian Robert-Stöter 2 Antoine Liutkus 2 Umut Simsekli 3 Romain Serizel 1 Roland Badeau 3
1 MULTISPEECH - Speech Modeling for Facilitating Oral-Based Communication
Inria Nancy - Grand Est, LORIA - NLPKD - Department of Natural Language Processing & Knowledge Discovery
2 ZENITH - Scientific Data Management
LIRMM - Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier, CRISAM - Inria Sophia Antipolis - Méditerranée
3 LTCI - Laboratoire Traitement et Communication de l'Information
Abstract : This paper introduces a new method for multichannel speech enhancement based on a versatile modeling of the residual noise spec-trogram. Such a model has already been presented before in the single channel case where the noise component is assumed to follow an alpha-stable distribution for each time-frequency bin, whereas the speech spec-trogram, supposed to be more regular, is modeled as Gaussian. In this paper, we describe a multichannel extension of this model, as well as a Monte Carlo Expectation-Maximisation algorithm for parameter estimation. In particular, a multichannel extension of the Itakura-Saito nonnegative matrix factorization is exploited to estimate the spectral parameters for speech, and a Metropolis-Hastings algorithm is proposed to estimate the noise contribution. We evaluate the proposed method in a challenging multichannel denoising application and compare it to other state-of-the-art algorithms.
Document type :
Conference papers
Complete list of metadatas

Cited literature [21 references]  Display  Hide  Download
https://hal-lirmm.ccsd.cnrs.fr/lirmm-01766795
Contributor : Antoine Liutkus <>
Submitted on : Saturday, April 14, 2018 - 9:58:31 AM
Last modification on : Saturday, November 16, 2019 - 7:04:01 PM

File

LVA-ICA2018_046_original_v5.pd...
Files produced by the author(s)

Identifiers

Collections

LIRMM | CNRS | INSTITUT-TELECOM | ZENITH | UNIV-LORRAINE | LORIA | LORIA-NLPKD | MIPS | INRIA | UNIV-MONTPELLIER | PARISTECH | LTCI | UNIV-PARIS-SACLAY

Citation

Mathieu Fontaine, Fabian Robert-Stöter, Antoine Liutkus, Umut Simsekli, Romain Serizel, et al.. Multichannel Audio Modeling with Elliptically Stable Tensor Decomposition. LVA/ICA: Latent Variable Analysis and Signal Separation, Jul 2018, Surrey, United Kingdom. pp.13-23, ⟨10.1007/978-3-319-93764-9_2⟩. ⟨lirmm-01766795⟩

Export

BibTeX TEI DC DCterms EndNote

Share

Metrics

Record views

883

Files downloads

439

Contact

Informations

CCSD