Multiple speaker tracking with the GLMB filter

© 2017 IEEE. In this paper we propose a new solution to the problem of tracking multiple speakers from multiple microphone arrays in a reverberant acoustic environment. The acoustic environment with its complex reflection patterns with its underlying data association uncertainty pose the two most si...

Full description

Bibliographic Details
Main Authors: Kim, Du Yong, Vo, B., Nordholm, Sven
Format: Conference Paper
Published: 2017
Online Access:http://purl.org/au-research/grants/arc/DP170104854
http://hdl.handle.net/20.500.11937/67212
_version_ 1848761505690943488
author Kim, Du Yong
Vo, B.
Nordholm, Sven
author_facet Kim, Du Yong
Vo, B.
Nordholm, Sven
author_sort Kim, Du Yong
building Curtin Institutional Repository
collection Online Access
description © 2017 IEEE. In this paper we propose a new solution to the problem of tracking multiple speakers from multiple microphone arrays in a reverberant acoustic environment. The acoustic environment with its complex reflection patterns with its underlying data association uncertainty pose the two most significant challenges in the multi-speaker tracking problem. We provide an approach that employs individual Time Difference of Arrival measurements collected by pairs of microphones in using multiple distributed pairs in conjunction with the Generalized Labeled Multi-Bernoulli (GLMB) tracker. The distributed measurements together with the GLMB tracking filter exploits the spatiotemporal correlation of the true sources from data frame to data frame, whereas the spurious measurements arising from reverberations exhibit no temporal consistency as the speakers move in the room.
first_indexed 2025-11-14T10:32:45Z
format Conference Paper
id curtin-20.500.11937-67212
institution Curtin University Malaysia
institution_category Local University
last_indexed 2025-11-14T10:32:45Z
publishDate 2017
recordtype eprints
repository_type Digital Repository
spelling curtin-20.500.11937-672122022-10-27T07:13:04Z Multiple speaker tracking with the GLMB filter Kim, Du Yong Vo, B. Nordholm, Sven © 2017 IEEE. In this paper we propose a new solution to the problem of tracking multiple speakers from multiple microphone arrays in a reverberant acoustic environment. The acoustic environment with its complex reflection patterns with its underlying data association uncertainty pose the two most significant challenges in the multi-speaker tracking problem. We provide an approach that employs individual Time Difference of Arrival measurements collected by pairs of microphones in using multiple distributed pairs in conjunction with the Generalized Labeled Multi-Bernoulli (GLMB) tracker. The distributed measurements together with the GLMB tracking filter exploits the spatiotemporal correlation of the true sources from data frame to data frame, whereas the spurious measurements arising from reverberations exhibit no temporal consistency as the speakers move in the room. 2017 Conference Paper http://hdl.handle.net/20.500.11937/67212 10.1109/ICCAIS.2017.8217590 http://purl.org/au-research/grants/arc/DP170104854 restricted
spellingShingle Kim, Du Yong
Vo, B.
Nordholm, Sven
Multiple speaker tracking with the GLMB filter
title Multiple speaker tracking with the GLMB filter
title_full Multiple speaker tracking with the GLMB filter
title_fullStr Multiple speaker tracking with the GLMB filter
title_full_unstemmed Multiple speaker tracking with the GLMB filter
title_short Multiple speaker tracking with the GLMB filter
title_sort multiple speaker tracking with the glmb filter
url http://purl.org/au-research/grants/arc/DP170104854
http://hdl.handle.net/20.500.11937/67212