Parameterization of Prosodic Feature Distributions for SVM Modeling in Speaker Recognition
Multiple recent studies have shown that speaker recognition performance using frame-based cepstral features is improved by adding higher-level information, including prosodic and lexical features. This paper explores the important question of finding a good kernel for a system that models syllable-b...
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Tagungsbericht |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext bestellen |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | IV-236 |
|---|---|
| container_issue | |
| container_start_page | IV-233 |
| container_title | |
| container_volume | 4 |
| creator | Ferrer, Luciana Shriberg, Elizabeth Kajarekar, Sachin Sonmez, Kemal |
| description | Multiple recent studies have shown that speaker recognition performance using frame-based cepstral features is improved by adding higher-level information, including prosodic and lexical features. This paper explores the important question of finding a good kernel for a system that models syllable-based prosodic features using support vector machines (SVMs). The system has been the best performing of our high-level systems in the last two NIST evaluations, and gives significant improvements when combined with cepstral-based systems. We introduce two new methods for transforming the syllable-level features into a single high-dimensional vector that can be well modeled by SVMs, resulting in significant gains in speaker recognition performance. |
| doi_str_mv | 10.1109/ICASSP.2007.366892 |
| format | Conference Proceeding |
| fullrecord | <record><control><sourceid>ieee_6IE</sourceid><recordid>TN_cdi_ieee_primary_4218080</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>4218080</ieee_id><sourcerecordid>4218080</sourcerecordid><originalsourceid>FETCH-LOGICAL-i175t-b23da31d89bded2587ee9987d74177861ad0d7c350dc356cd2c410c7a21fb31f3</originalsourceid><addsrcrecordid>eNpVj8tOwzAURM1LopT-AGz8Ayn32k5sL1GhgNSKiABCbConvqkMbVI56QK-nlawYTNnMUcjDWMXCGNEsFcPk-uiyMcCQI9llhkrDtjIaoNKKAVamOyQDYTUNkELb0f_Om2P2QBTAUmGyp6ys677AACjlRmw99xFt6aeYvh2fWgb3tY8j23X-lDxKbl-G4nfhK6PodzuhY7XbeTF65zPW0-r0Cx5aHixIfdJkT9R1S6bsBfP2UntVh2N_jhkL9Pb58l9Mnu82_2ZJQF12ielkN5J9MaWnrxIjSay1mivFWptMnQevK5kCn4XWeVFpRAq7QTWpcRaDtnl724gosUmhrWLXwsl0IAB-QPKCViq</addsrcrecordid><sourcetype>Publisher</sourcetype><iscdi>true</iscdi><recordtype>conference_proceeding</recordtype></control><display><type>conference_proceeding</type><title>Parameterization of Prosodic Feature Distributions for SVM Modeling in Speaker Recognition</title><source>IEEE Electronic Library (IEL) Conference Proceedings</source><creator>Ferrer, Luciana ; Shriberg, Elizabeth ; Kajarekar, Sachin ; Sonmez, Kemal</creator><creatorcontrib>Ferrer, Luciana ; Shriberg, Elizabeth ; Kajarekar, Sachin ; Sonmez, Kemal</creatorcontrib><description>Multiple recent studies have shown that speaker recognition performance using frame-based cepstral features is improved by adding higher-level information, including prosodic and lexical features. This paper explores the important question of finding a good kernel for a system that models syllable-based prosodic features using support vector machines (SVMs). The system has been the best performing of our high-level systems in the last two NIST evaluations, and gives significant improvements when combined with cepstral-based systems. We introduce two new methods for transforming the syllable-level features into a single high-dimensional vector that can be well modeled by SVMs, resulting in significant gains in speaker recognition performance.</description><identifier>ISSN: 1520-6149</identifier><identifier>ISBN: 9781424407279</identifier><identifier>ISBN: 1424407273</identifier><identifier>EISSN: 2379-190X</identifier><identifier>EISBN: 9781424407286</identifier><identifier>EISBN: 1424407281</identifier><identifier>DOI: 10.1109/ICASSP.2007.366892</identifier><language>eng</language><publisher>IEEE</publisher><subject>Cepstral analysis ; Feature extraction ; GMM ; Kernel ; Laboratories ; NIST ; Performance evaluation ; Performance gain ; Prosody ; Speaker recognition ; Speech ; Support vector machines ; SVM</subject><ispartof>2007 IEEE International Conference on Acoustics, Speech and Signal Processing - ICASSP '07, 2007, Vol.4, p.IV-233-IV-236</ispartof><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/4218080$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>309,310,780,784,789,790,2058,27925,54920</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/4218080$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc></links><search><creatorcontrib>Ferrer, Luciana</creatorcontrib><creatorcontrib>Shriberg, Elizabeth</creatorcontrib><creatorcontrib>Kajarekar, Sachin</creatorcontrib><creatorcontrib>Sonmez, Kemal</creatorcontrib><title>Parameterization of Prosodic Feature Distributions for SVM Modeling in Speaker Recognition</title><title>2007 IEEE International Conference on Acoustics, Speech and Signal Processing - ICASSP '07</title><addtitle>ICASSP</addtitle><description>Multiple recent studies have shown that speaker recognition performance using frame-based cepstral features is improved by adding higher-level information, including prosodic and lexical features. This paper explores the important question of finding a good kernel for a system that models syllable-based prosodic features using support vector machines (SVMs). The system has been the best performing of our high-level systems in the last two NIST evaluations, and gives significant improvements when combined with cepstral-based systems. We introduce two new methods for transforming the syllable-level features into a single high-dimensional vector that can be well modeled by SVMs, resulting in significant gains in speaker recognition performance.</description><subject>Cepstral analysis</subject><subject>Feature extraction</subject><subject>GMM</subject><subject>Kernel</subject><subject>Laboratories</subject><subject>NIST</subject><subject>Performance evaluation</subject><subject>Performance gain</subject><subject>Prosody</subject><subject>Speaker recognition</subject><subject>Speech</subject><subject>Support vector machines</subject><subject>SVM</subject><issn>1520-6149</issn><issn>2379-190X</issn><isbn>9781424407279</isbn><isbn>1424407273</isbn><isbn>9781424407286</isbn><isbn>1424407281</isbn><fulltext>true</fulltext><rsrctype>conference_proceeding</rsrctype><creationdate>2007</creationdate><recordtype>conference_proceeding</recordtype><sourceid>6IE</sourceid><sourceid>RIE</sourceid><recordid>eNpVj8tOwzAURM1LopT-AGz8Ayn32k5sL1GhgNSKiABCbConvqkMbVI56QK-nlawYTNnMUcjDWMXCGNEsFcPk-uiyMcCQI9llhkrDtjIaoNKKAVamOyQDYTUNkELb0f_Om2P2QBTAUmGyp6ys677AACjlRmw99xFt6aeYvh2fWgb3tY8j23X-lDxKbl-G4nfhK6PodzuhY7XbeTF65zPW0-r0Cx5aHixIfdJkT9R1S6bsBfP2UntVh2N_jhkL9Pb58l9Mnu82_2ZJQF12ielkN5J9MaWnrxIjSay1mivFWptMnQevK5kCn4XWeVFpRAq7QTWpcRaDtnl724gosUmhrWLXwsl0IAB-QPKCViq</recordid><startdate>200704</startdate><enddate>200704</enddate><creator>Ferrer, Luciana</creator><creator>Shriberg, Elizabeth</creator><creator>Kajarekar, Sachin</creator><creator>Sonmez, Kemal</creator><general>IEEE</general><scope>6IE</scope><scope>6IH</scope><scope>CBEJK</scope><scope>RIE</scope><scope>RIO</scope></search><sort><creationdate>200704</creationdate><title>Parameterization of Prosodic Feature Distributions for SVM Modeling in Speaker Recognition</title><author>Ferrer, Luciana ; Shriberg, Elizabeth ; Kajarekar, Sachin ; Sonmez, Kemal</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-i175t-b23da31d89bded2587ee9987d74177861ad0d7c350dc356cd2c410c7a21fb31f3</frbrgroupid><rsrctype>conference_proceedings</rsrctype><prefilter>conference_proceedings</prefilter><language>eng</language><creationdate>2007</creationdate><topic>Cepstral analysis</topic><topic>Feature extraction</topic><topic>GMM</topic><topic>Kernel</topic><topic>Laboratories</topic><topic>NIST</topic><topic>Performance evaluation</topic><topic>Performance gain</topic><topic>Prosody</topic><topic>Speaker recognition</topic><topic>Speech</topic><topic>Support vector machines</topic><topic>SVM</topic><toplevel>online_resources</toplevel><creatorcontrib>Ferrer, Luciana</creatorcontrib><creatorcontrib>Shriberg, Elizabeth</creatorcontrib><creatorcontrib>Kajarekar, Sachin</creatorcontrib><creatorcontrib>Sonmez, Kemal</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 Electronic Library (IEL)</collection><collection>IEEE Proceedings Order Plans (POP) 1998-present</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Ferrer, Luciana</au><au>Shriberg, Elizabeth</au><au>Kajarekar, Sachin</au><au>Sonmez, Kemal</au><format>book</format><genre>proceeding</genre><ristype>CONF</ristype><atitle>Parameterization of Prosodic Feature Distributions for SVM Modeling in Speaker Recognition</atitle><btitle>2007 IEEE International Conference on Acoustics, Speech and Signal Processing - ICASSP '07</btitle><stitle>ICASSP</stitle><date>2007-04</date><risdate>2007</risdate><volume>4</volume><spage>IV-233</spage><epage>IV-236</epage><pages>IV-233-IV-236</pages><issn>1520-6149</issn><eissn>2379-190X</eissn><isbn>9781424407279</isbn><isbn>1424407273</isbn><eisbn>9781424407286</eisbn><eisbn>1424407281</eisbn><abstract>Multiple recent studies have shown that speaker recognition performance using frame-based cepstral features is improved by adding higher-level information, including prosodic and lexical features. This paper explores the important question of finding a good kernel for a system that models syllable-based prosodic features using support vector machines (SVMs). The system has been the best performing of our high-level systems in the last two NIST evaluations, and gives significant improvements when combined with cepstral-based systems. We introduce two new methods for transforming the syllable-level features into a single high-dimensional vector that can be well modeled by SVMs, resulting in significant gains in speaker recognition performance.</abstract><pub>IEEE</pub><doi>10.1109/ICASSP.2007.366892</doi></addata></record> |
| fulltext | fulltext_linktorsrc |
| identifier | ISSN: 1520-6149 |
| ispartof | 2007 IEEE International Conference on Acoustics, Speech and Signal Processing - ICASSP '07, 2007, Vol.4, p.IV-233-IV-236 |
| issn | 1520-6149 2379-190X |
| language | eng |
| recordid | cdi_ieee_primary_4218080 |
| source | IEEE Electronic Library (IEL) Conference Proceedings |
| subjects | Cepstral analysis Feature extraction GMM Kernel Laboratories NIST Performance evaluation Performance gain Prosody Speaker recognition Speech Support vector machines SVM |
| title | Parameterization of Prosodic Feature Distributions for SVM Modeling in Speaker Recognition |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-04T18%3A37%3A14IST&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=Parameterization%20of%20Prosodic%20Feature%20Distributions%20for%20SVM%20Modeling%20in%20Speaker%20Recognition&rft.btitle=2007%20IEEE%20International%20Conference%20on%20Acoustics,%20Speech%20and%20Signal%20Processing%20-%20ICASSP%20'07&rft.au=Ferrer,%20Luciana&rft.date=2007-04&rft.volume=4&rft.spage=IV-233&rft.epage=IV-236&rft.pages=IV-233-IV-236&rft.issn=1520-6149&rft.eissn=2379-190X&rft.isbn=9781424407279&rft.isbn_list=1424407273&rft_id=info:doi/10.1109/ICASSP.2007.366892&rft_dat=%3Cieee_6IE%3E4218080%3C/ieee_6IE%3E%3Curl%3E%3C/url%3E&rft.eisbn=9781424407286&rft.eisbn_list=1424407281&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rft_ieee_id=4218080&rfr_iscdi=true |