Source number estimation in reverberant conditions via full-band weighted, adaptive fuzzy c-means clustering

We introduce a novel approach for source number estimation through an adaptive fuzzy c-means clustering. Spatial feature vectors are extracted from microphone observations, weighted for reliability and then clustered in a full-band manner using an adaptive variation on the fuzzy c-means. A number of...

Full description

Bibliographic Details
Main Authors: Hollick, J., Jafari, I., Togneri, R., Nordholm, Sven
Other Authors: ICASSP 2014 Publication Chair Maria Sabrina Greco
Format: Conference Paper
Published: IEEE 2014
Subjects:
Online Access:http://hdl.handle.net/20.500.11937/35254
_version_ 1848754446767489024
author Hollick, J.
Jafari, I.
Togneri, R.
Nordholm, Sven
author2 ICASSP 2014 Publication Chair Maria Sabrina Greco
author_facet ICASSP 2014 Publication Chair Maria Sabrina Greco
Hollick, J.
Jafari, I.
Togneri, R.
Nordholm, Sven
author_sort Hollick, J.
building Curtin Institutional Repository
collection Online Access
description We introduce a novel approach for source number estimation through an adaptive fuzzy c-means clustering. Spatial feature vectors are extracted from microphone observations, weighted for reliability and then clustered in a full-band manner using an adaptive variation on the fuzzy c-means. A number of quality measures are combined to produce a weighted sum which is used to find the optimal number of clusters at each iteration of the clustering algorithm. Experimental evaluations using real-world recordings from a reverberant room (RT60 = 390 ms) demonstrated encouraging performance in both even- and under-determined conditions.
first_indexed 2025-11-14T08:40:33Z
format Conference Paper
id curtin-20.500.11937-35254
institution Curtin University Malaysia
institution_category Local University
last_indexed 2025-11-14T08:40:33Z
publishDate 2014
publisher IEEE
recordtype eprints
repository_type Digital Repository
spelling curtin-20.500.11937-352542017-09-13T15:32:49Z Source number estimation in reverberant conditions via full-band weighted, adaptive fuzzy c-means clustering Hollick, J. Jafari, I. Togneri, R. Nordholm, Sven ICASSP 2014 Publication Chair Maria Sabrina Greco fuzzy c-means clustering source number estimation weights adaptive quality measure We introduce a novel approach for source number estimation through an adaptive fuzzy c-means clustering. Spatial feature vectors are extracted from microphone observations, weighted for reliability and then clustered in a full-band manner using an adaptive variation on the fuzzy c-means. A number of quality measures are combined to produce a weighted sum which is used to find the optimal number of clusters at each iteration of the clustering algorithm. Experimental evaluations using real-world recordings from a reverberant room (RT60 = 390 ms) demonstrated encouraging performance in both even- and under-determined conditions. 2014 Conference Paper http://hdl.handle.net/20.500.11937/35254 10.1109/ICASSP.2014.6855048 IEEE restricted
spellingShingle fuzzy c-means clustering
source number estimation
weights
adaptive
quality measure
Hollick, J.
Jafari, I.
Togneri, R.
Nordholm, Sven
Source number estimation in reverberant conditions via full-band weighted, adaptive fuzzy c-means clustering
title Source number estimation in reverberant conditions via full-band weighted, adaptive fuzzy c-means clustering
title_full Source number estimation in reverberant conditions via full-band weighted, adaptive fuzzy c-means clustering
title_fullStr Source number estimation in reverberant conditions via full-band weighted, adaptive fuzzy c-means clustering
title_full_unstemmed Source number estimation in reverberant conditions via full-band weighted, adaptive fuzzy c-means clustering
title_short Source number estimation in reverberant conditions via full-band weighted, adaptive fuzzy c-means clustering
title_sort source number estimation in reverberant conditions via full-band weighted, adaptive fuzzy c-means clustering
topic fuzzy c-means clustering
source number estimation
weights
adaptive
quality measure
url http://hdl.handle.net/20.500.11937/35254