Loading…
Speaker diarization of meetings based on large TDOA feature vectors
This paper investigates the use of large TDOA feature vectors together with acoustic information in speaker diarization of meetings. TDOAs are obtained by considering all possible microphones pairs and this approach is compared with conventional TDOA features extracted w.r.t. a reference channel. Th...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Conference Proceeding |
| Language: | English |
| Subjects: | |
| Online Access: | Request full text |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| cited_by | |
|---|---|
| cites | |
| container_end_page | 4176 |
| container_issue | |
| container_start_page | 4173 |
| container_title | |
| container_volume | |
| creator | Vijayasenan, D. Valente, F. |
| description | This paper investigates the use of large TDOA feature vectors together with acoustic information in speaker diarization of meetings. TDOAs are obtained by considering all possible microphones pairs and this approach is compared with conventional TDOA features extracted w.r.t. a reference channel. The study is carried using two systems, the first based on Gaussian Mixture Modeling and the second based on the Information Bottleneck approach. Results on NIST RT06/RT07/RT09 evaluation datasets show a large speaker error reduction of 30% relative going from 14.3% to 10.8% for the first and from 12.3% to 8.2% for the second whenever the feature weighting is properly handled. Furthermore results reveal that the IB system is more robust to different number of microphones even when all pairs large TDOA vectors are used thus outperforming the HMM/GMM by 25% relative (8.2% error compared to 10.8%). |
| doi_str_mv | 10.1109/ICASSP.2012.6288838 |
| format | conference_proceeding |
| fullrecord | <record><control><sourceid>ieee_6IE</sourceid><recordid>TN_cdi_ieee_primary_6288838</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>6288838</ieee_id><sourcerecordid>6288838</sourcerecordid><originalsourceid>FETCH-LOGICAL-i220t-edc818a254d5b1655626ee252501c96de677fab35c75f8b8f6bbf96ccbe6f56a3</originalsourceid><addsrcrecordid>eNo1UG1LwzAYjG_gnP0F-5I_0JqX5mnycVSnwmBCJ_htJO2TEd3WkVRBf70F5305uOOO4wiZcVZwzszdcz1vmpdCMC4KEFprqc9IZirNS6gkYyWYczIRsjI5N-ztgtz8G6q8JBOuBMuBl-aaZCm9sxFjlEmYkLo5ov3ASLtgY_ixQ-gPtPd0jziEwzZRZxN2dBR3Nm6Rru9Xc-rRDp8R6Re2Qx_TLbnydpcwO_GUvC4e1vVTvlw9jsuXeRCCDTl2rebaClV2ynFQCgQgCiUU462BDqGqvHVStZXy2mkPznkDbesQvAIrp2T21xsQcXOMYW_j9-b0h_wFv71QJg</addsrcrecordid><sourcetype>Publisher</sourcetype><iscdi>true</iscdi><recordtype>conference_proceeding</recordtype></control><display><type>conference_proceeding</type><title>Speaker diarization of meetings based on large TDOA feature vectors</title><source>IEEE Electronic Library (IEL) Conference Proceedings</source><creator>Vijayasenan, D. ; Valente, F.</creator><creatorcontrib>Vijayasenan, D. ; Valente, F.</creatorcontrib><description>This paper investigates the use of large TDOA feature vectors together with acoustic information in speaker diarization of meetings. TDOAs are obtained by considering all possible microphones pairs and this approach is compared with conventional TDOA features extracted w.r.t. a reference channel. The study is carried using two systems, the first based on Gaussian Mixture Modeling and the second based on the Information Bottleneck approach. Results on NIST RT06/RT07/RT09 evaluation datasets show a large speaker error reduction of 30% relative going from 14.3% to 10.8% for the first and from 12.3% to 8.2% for the second whenever the feature weighting is properly handled. Furthermore results reveal that the IB system is more robust to different number of microphones even when all pairs large TDOA vectors are used thus outperforming the HMM/GMM by 25% relative (8.2% error compared to 10.8%).</description><identifier>ISSN: 1520-6149</identifier><identifier>ISBN: 1467300454</identifier><identifier>ISBN: 9781467300452</identifier><identifier>EISSN: 2379-190X</identifier><identifier>EISBN: 9781467300469</identifier><identifier>EISBN: 1467300446</identifier><identifier>EISBN: 9781467300445</identifier><identifier>EISBN: 1467300462</identifier><identifier>DOI: 10.1109/ICASSP.2012.6288838</identifier><language>eng</language><publisher>IEEE</publisher><subject>Acoustics ; Delay ; Hidden Markov models ; Meetings Recordings ; Microphones ; Model combination ; NIST ; Speaker diarization ; Speech ; Time Delay Of Arrival features ; Vectors</subject><ispartof>2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2012, p.4173-4176</ispartof><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/6288838$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>309,310,780,784,789,790,2058,27925,54555,54920,54932</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/6288838$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc></links><search><creatorcontrib>Vijayasenan, D.</creatorcontrib><creatorcontrib>Valente, F.</creatorcontrib><title>Speaker diarization of meetings based on large TDOA feature vectors</title><title>2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)</title><addtitle>ICASSP</addtitle><description>This paper investigates the use of large TDOA feature vectors together with acoustic information in speaker diarization of meetings. TDOAs are obtained by considering all possible microphones pairs and this approach is compared with conventional TDOA features extracted w.r.t. a reference channel. The study is carried using two systems, the first based on Gaussian Mixture Modeling and the second based on the Information Bottleneck approach. Results on NIST RT06/RT07/RT09 evaluation datasets show a large speaker error reduction of 30% relative going from 14.3% to 10.8% for the first and from 12.3% to 8.2% for the second whenever the feature weighting is properly handled. Furthermore results reveal that the IB system is more robust to different number of microphones even when all pairs large TDOA vectors are used thus outperforming the HMM/GMM by 25% relative (8.2% error compared to 10.8%).</description><subject>Acoustics</subject><subject>Delay</subject><subject>Hidden Markov models</subject><subject>Meetings Recordings</subject><subject>Microphones</subject><subject>Model combination</subject><subject>NIST</subject><subject>Speaker diarization</subject><subject>Speech</subject><subject>Time Delay Of Arrival features</subject><subject>Vectors</subject><issn>1520-6149</issn><issn>2379-190X</issn><isbn>1467300454</isbn><isbn>9781467300452</isbn><isbn>9781467300469</isbn><isbn>1467300446</isbn><isbn>9781467300445</isbn><isbn>1467300462</isbn><fulltext>true</fulltext><rsrctype>conference_proceeding</rsrctype><creationdate>2012</creationdate><recordtype>conference_proceeding</recordtype><sourceid>6IE</sourceid><recordid>eNo1UG1LwzAYjG_gnP0F-5I_0JqX5mnycVSnwmBCJ_htJO2TEd3WkVRBf70F5305uOOO4wiZcVZwzszdcz1vmpdCMC4KEFprqc9IZirNS6gkYyWYczIRsjI5N-ztgtz8G6q8JBOuBMuBl-aaZCm9sxFjlEmYkLo5ov3ASLtgY_ixQ-gPtPd0jziEwzZRZxN2dBR3Nm6Rru9Xc-rRDp8R6Re2Qx_TLbnydpcwO_GUvC4e1vVTvlw9jsuXeRCCDTl2rebaClV2ynFQCgQgCiUU462BDqGqvHVStZXy2mkPznkDbesQvAIrp2T21xsQcXOMYW_j9-b0h_wFv71QJg</recordid><startdate>20120101</startdate><enddate>20120101</enddate><creator>Vijayasenan, D.</creator><creator>Valente, F.</creator><general>IEEE</general><scope>6IE</scope><scope>6IH</scope><scope>CBEJK</scope><scope>RIE</scope><scope>RIO</scope></search><sort><creationdate>20120101</creationdate><title>Speaker diarization of meetings based on large TDOA feature vectors</title><author>Vijayasenan, D. ; Valente, F.</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-i220t-edc818a254d5b1655626ee252501c96de677fab35c75f8b8f6bbf96ccbe6f56a3</frbrgroupid><rsrctype>conference_proceedings</rsrctype><prefilter>conference_proceedings</prefilter><language>eng</language><creationdate>2012</creationdate><topic>Acoustics</topic><topic>Delay</topic><topic>Hidden Markov models</topic><topic>Meetings Recordings</topic><topic>Microphones</topic><topic>Model combination</topic><topic>NIST</topic><topic>Speaker diarization</topic><topic>Speech</topic><topic>Time Delay Of Arrival features</topic><topic>Vectors</topic><toplevel>online_resources</toplevel><creatorcontrib>Vijayasenan, D.</creatorcontrib><creatorcontrib>Valente, F.</creatorcontrib><collection>IEEE Electronic Library (IEL) Conference Proceedings</collection><collection>IEEE Proceedings Order Plan (POP) 1998-present by volume</collection><collection>IEEE Xplore All Conference Proceedings</collection><collection>IEEE Xplore</collection><collection>IEEE Proceedings Order Plans (POP) 1998-present</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Vijayasenan, D.</au><au>Valente, F.</au><format>book</format><genre>proceeding</genre><ristype>CONF</ristype><atitle>Speaker diarization of meetings based on large TDOA feature vectors</atitle><btitle>2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)</btitle><stitle>ICASSP</stitle><date>2012-01-01</date><risdate>2012</risdate><spage>4173</spage><epage>4176</epage><pages>4173-4176</pages><issn>1520-6149</issn><eissn>2379-190X</eissn><isbn>1467300454</isbn><isbn>9781467300452</isbn><eisbn>9781467300469</eisbn><eisbn>1467300446</eisbn><eisbn>9781467300445</eisbn><eisbn>1467300462</eisbn><abstract>This paper investigates the use of large TDOA feature vectors together with acoustic information in speaker diarization of meetings. TDOAs are obtained by considering all possible microphones pairs and this approach is compared with conventional TDOA features extracted w.r.t. a reference channel. The study is carried using two systems, the first based on Gaussian Mixture Modeling and the second based on the Information Bottleneck approach. Results on NIST RT06/RT07/RT09 evaluation datasets show a large speaker error reduction of 30% relative going from 14.3% to 10.8% for the first and from 12.3% to 8.2% for the second whenever the feature weighting is properly handled. Furthermore results reveal that the IB system is more robust to different number of microphones even when all pairs large TDOA vectors are used thus outperforming the HMM/GMM by 25% relative (8.2% error compared to 10.8%).</abstract><pub>IEEE</pub><doi>10.1109/ICASSP.2012.6288838</doi><tpages>4</tpages><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext_linktorsrc |
| identifier | ISSN: 1520-6149 |
| ispartof | 2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2012, p.4173-4176 |
| issn | 1520-6149 2379-190X |
| language | eng |
| recordid | cdi_ieee_primary_6288838 |
| source | IEEE Electronic Library (IEL) Conference Proceedings |
| subjects | Acoustics Delay Hidden Markov models Meetings Recordings Microphones Model combination NIST Speaker diarization Speech Time Delay Of Arrival features Vectors |
| title | Speaker diarization of meetings based on large TDOA feature vectors |
| url | http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-05T14%3A59%3A12IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-ieee_6IE&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=proceeding&rft.atitle=Speaker%20diarization%20of%20meetings%20based%20on%20large%20TDOA%20feature%20vectors&rft.btitle=2012%20IEEE%20International%20Conference%20on%20Acoustics,%20Speech%20and%20Signal%20Processing%20(ICASSP)&rft.au=Vijayasenan,%20D.&rft.date=2012-01-01&rft.spage=4173&rft.epage=4176&rft.pages=4173-4176&rft.issn=1520-6149&rft.eissn=2379-190X&rft.isbn=1467300454&rft.isbn_list=9781467300452&rft_id=info:doi/10.1109/ICASSP.2012.6288838&rft.eisbn=9781467300469&rft.eisbn_list=1467300446&rft.eisbn_list=9781467300445&rft.eisbn_list=1467300462&rft_dat=%3Cieee_6IE%3E6288838%3C/ieee_6IE%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-i220t-edc818a254d5b1655626ee252501c96de677fab35c75f8b8f6bbf96ccbe6f56a3%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_id=info:pmid/&rft_ieee_id=6288838&rfr_iscdi=true |