Speaker localization and separation using incremental distributed expectation-maximization

Yuval Dorfan, Dani Cherkassky, Sharon Gannot

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

9 Scopus citations

Abstract

A network of microphone pairs is utilized for the joint task of localizing and separating multiple concurrent speakers. The recently presented incremental distributed expectation-maximization (IDEM) is addressing the first task, namely detection and localization. Here we extend this algorithm to address the second task, namely blindly separating the speech sources. We show that the proposed algorithm, denoted distributed algorithm for localization and separation (DALAS), is capable of separating speakers in reverberant enclosure without a priori information on their number and locations. In the first stage of the proposed algorithm, the IDEM algorithm is applied for blindly detecting the active sources and to estimate their locations. In the second stage, the location estimates are utilized for selecting the most useful node of microphones for the subsequent separation stage. Separation is finally obtained by utilizing the hidden variables of the IDEM algorithm to construct masks for each source in the relevant node.

Original languageEnglish
Title of host publication2015 23rd European Signal Processing Conference, EUSIPCO 2015
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1256-1260
Number of pages5
ISBN (Electronic)9780992862633
DOIs
StatePublished - 22 Dec 2015
Event23rd European Signal Processing Conference, EUSIPCO 2015 - Nice, France
Duration: 31 Aug 20154 Sep 2015

Publication series

Name2015 23rd European Signal Processing Conference, EUSIPCO 2015

Conference

Conference23rd European Signal Processing Conference, EUSIPCO 2015
Country/TerritoryFrance
CityNice
Period31/08/154/09/15

Bibliographical note

Publisher Copyright:
© 2015 EURASIP.

Keywords

  • Wireless acoustic sensor network
  • blind source separation
  • incremental estimate-maximize

Fingerprint

Dive into the research topics of 'Speaker localization and separation using incremental distributed expectation-maximization'. Together they form a unique fingerprint.

Cite this