Multichannel audio modeling with elliptically stable tensor decomposition

  • Mathieu Fontaine
  • , Fabian Robert Stöter
  • , Antoine Liutkus
  • , Umut Şimşekli
  • , Romain Serizel
  • , Roland Badeau

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

This paper introduces a new method for multichannel speech enhancement based on a versatile modeling of the residual noise spectrogram. 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 spectrogram, 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.

Original languageEnglish
Title of host publicationLatent Variable Analysis and Signal Separation - 14th International Conference, LVA/ICA 2018, Proceedings
EditorsSharon Gannot, Yannick Deville, Russell Mason, Mark D. Plumbley, Dominic Ward
PublisherSpringer Verlag
Pages13-23
Number of pages11
ISBN (Print)9783319937632
DOIs
Publication statusPublished - 1 Jan 2018
Externally publishedYes
Event14th International Conference on Latent Variable Analysis and Signal Separation, LVA/ICA 2018 - Guildford, United Kingdom
Duration: 2 Jul 20185 Jul 2018

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume10891 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference14th International Conference on Latent Variable Analysis and Signal Separation, LVA/ICA 2018
Country/TerritoryUnited Kingdom
CityGuildford
Period2/07/185/07/18

Fingerprint

Dive into the research topics of 'Multichannel audio modeling with elliptically stable tensor decomposition'. Together they form a unique fingerprint.

Cite this