Confidence measure improvement using useful predictor features and support vector machines
In traditional keyword spotting (KWS) systems, confidence measure (CM) of each keyword is computed from normalized acoustic likelihoods. In addition to likelihood based scores, some keyword dependent features named predictor features such as duration and prosodic features could be defined to improve...
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 | 1171 |
|---|---|
| container_issue | |
| container_start_page | 1168 |
| container_title | |
| container_volume | |
| creator | Shekofteh, Y. Kabudian, J. Goodarzi, M. M. Rezaei, I. S. |
| description | In traditional keyword spotting (KWS) systems, confidence measure (CM) of each keyword is computed from normalized acoustic likelihoods. In addition to likelihood based scores, some keyword dependent features named predictor features such as duration and prosodic features could be defined to improve the performance of CM. In this paper a discriminative and probabilistic computation of CM based upon some useful predictor features and support vector machines (SVM) is presented for Persian conversational telephone speech KWS. Our experimental results show that higher performance will be achieved by appending utilized predictor features. The proposed CM with linear kernel function of SVM is obtained an improvement about 8.6% in Figure-of-Merit (FOM) of KWS system. |
| doi_str_mv | 10.1109/IranianCEE.2012.6292531 |
| format | Conference Proceeding |
| fullrecord | <record><control><sourceid>ieee_6IE</sourceid><recordid>TN_cdi_ieee_primary_6292531</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>6292531</ieee_id><sourcerecordid>6292531</sourcerecordid><originalsourceid>FETCH-LOGICAL-c144t-89e0f009e02764ede1f6cc6088784d837219dbee54bff46c281bb1658f4433263</originalsourceid><addsrcrecordid>eNo1kLFOwzAYhI0AiVL6BAz4BVL8245jjygqUKkSCywsleP8BqPGieykEm9PBGX5Tqc73XCE3AFbAzBzv002BhvrzWbNGfC14oaXAs7IylQapKoEgKz4Obn-N8ZckAUHJYuKlfKKrHL-YowJ0FqbckHe6z760GJ0SDu0eUpIQzek_ogdxpFOOcSPmeinAx0StsGNfaIe7ThXM7WxpXkahj6N9Ii_WWfdZ4iYb8ilt4eMq5Muydvj5rV-LnYvT9v6YVc4kHIstEHmGZvJKyWxRfDKOcW0rrRstag4mLZBLGXjvVSOa2gaUKX2UgrBlViS27_dgIj7IYXOpu_96RrxAy5bWRI</addsrcrecordid><sourcetype>Publisher</sourcetype><iscdi>true</iscdi><recordtype>conference_proceeding</recordtype></control><display><type>conference_proceeding</type><title>Confidence measure improvement using useful predictor features and support vector machines</title><source>IEEE Electronic Library (IEL) Conference Proceedings</source><creator>Shekofteh, Y. ; Kabudian, J. ; Goodarzi, M. M. ; Rezaei, I. S.</creator><creatorcontrib>Shekofteh, Y. ; Kabudian, J. ; Goodarzi, M. M. ; Rezaei, I. S.</creatorcontrib><description>In traditional keyword spotting (KWS) systems, confidence measure (CM) of each keyword is computed from normalized acoustic likelihoods. In addition to likelihood based scores, some keyword dependent features named predictor features such as duration and prosodic features could be defined to improve the performance of CM. In this paper a discriminative and probabilistic computation of CM based upon some useful predictor features and support vector machines (SVM) is presented for Persian conversational telephone speech KWS. Our experimental results show that higher performance will be achieved by appending utilized predictor features. The proposed CM with linear kernel function of SVM is obtained an improvement about 8.6% in Figure-of-Merit (FOM) of KWS system.</description><identifier>ISSN: 2164-7054</identifier><identifier>ISBN: 1467311499</identifier><identifier>ISBN: 9781467311496</identifier><identifier>EISBN: 9781467311472</identifier><identifier>EISBN: 1467311480</identifier><identifier>EISBN: 9781467311489</identifier><identifier>EISBN: 1467311472</identifier><identifier>DOI: 10.1109/IranianCEE.2012.6292531</identifier><language>eng</language><publisher>IEEE</publisher><subject>Acoustics ; confidence measure ; Hidden Markov models ; Ions ; Kernel ; keyword spotting ; predictor feature ; speech recognition ; Support vector machines ; SVM</subject><ispartof>20th Iranian Conference on Electrical Engineering (ICEE2012), 2012, p.1168-1171</ispartof><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c144t-89e0f009e02764ede1f6cc6088784d837219dbee54bff46c281bb1658f4433263</citedby></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/6292531$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>310,311,781,785,790,791,2059,27929,54924</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/6292531$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc></links><search><creatorcontrib>Shekofteh, Y.</creatorcontrib><creatorcontrib>Kabudian, J.</creatorcontrib><creatorcontrib>Goodarzi, M. M.</creatorcontrib><creatorcontrib>Rezaei, I. S.</creatorcontrib><title>Confidence measure improvement using useful predictor features and support vector machines</title><title>20th Iranian Conference on Electrical Engineering (ICEE2012)</title><addtitle>IranianCEE</addtitle><description>In traditional keyword spotting (KWS) systems, confidence measure (CM) of each keyword is computed from normalized acoustic likelihoods. In addition to likelihood based scores, some keyword dependent features named predictor features such as duration and prosodic features could be defined to improve the performance of CM. In this paper a discriminative and probabilistic computation of CM based upon some useful predictor features and support vector machines (SVM) is presented for Persian conversational telephone speech KWS. Our experimental results show that higher performance will be achieved by appending utilized predictor features. The proposed CM with linear kernel function of SVM is obtained an improvement about 8.6% in Figure-of-Merit (FOM) of KWS system.</description><subject>Acoustics</subject><subject>confidence measure</subject><subject>Hidden Markov models</subject><subject>Ions</subject><subject>Kernel</subject><subject>keyword spotting</subject><subject>predictor feature</subject><subject>speech recognition</subject><subject>Support vector machines</subject><subject>SVM</subject><issn>2164-7054</issn><isbn>1467311499</isbn><isbn>9781467311496</isbn><isbn>9781467311472</isbn><isbn>1467311480</isbn><isbn>9781467311489</isbn><isbn>1467311472</isbn><fulltext>true</fulltext><rsrctype>conference_proceeding</rsrctype><creationdate>2012</creationdate><recordtype>conference_proceeding</recordtype><sourceid>6IE</sourceid><sourceid>RIE</sourceid><recordid>eNo1kLFOwzAYhI0AiVL6BAz4BVL8245jjygqUKkSCywsleP8BqPGieykEm9PBGX5Tqc73XCE3AFbAzBzv002BhvrzWbNGfC14oaXAs7IylQapKoEgKz4Obn-N8ZckAUHJYuKlfKKrHL-YowJ0FqbckHe6z760GJ0SDu0eUpIQzek_ogdxpFOOcSPmeinAx0StsGNfaIe7ThXM7WxpXkahj6N9Ii_WWfdZ4iYb8ilt4eMq5Muydvj5rV-LnYvT9v6YVc4kHIstEHmGZvJKyWxRfDKOcW0rrRstag4mLZBLGXjvVSOa2gaUKX2UgrBlViS27_dgIj7IYXOpu_96RrxAy5bWRI</recordid><startdate>201205</startdate><enddate>201205</enddate><creator>Shekofteh, Y.</creator><creator>Kabudian, J.</creator><creator>Goodarzi, M. M.</creator><creator>Rezaei, I. S.</creator><general>IEEE</general><scope>6IE</scope><scope>6IL</scope><scope>CBEJK</scope><scope>RIE</scope><scope>RIL</scope></search><sort><creationdate>201205</creationdate><title>Confidence measure improvement using useful predictor features and support vector machines</title><author>Shekofteh, Y. ; Kabudian, J. ; Goodarzi, M. M. ; Rezaei, I. S.</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c144t-89e0f009e02764ede1f6cc6088784d837219dbee54bff46c281bb1658f4433263</frbrgroupid><rsrctype>conference_proceedings</rsrctype><prefilter>conference_proceedings</prefilter><language>eng</language><creationdate>2012</creationdate><topic>Acoustics</topic><topic>confidence measure</topic><topic>Hidden Markov models</topic><topic>Ions</topic><topic>Kernel</topic><topic>keyword spotting</topic><topic>predictor feature</topic><topic>speech recognition</topic><topic>Support vector machines</topic><topic>SVM</topic><toplevel>online_resources</toplevel><creatorcontrib>Shekofteh, Y.</creatorcontrib><creatorcontrib>Kabudian, J.</creatorcontrib><creatorcontrib>Goodarzi, M. M.</creatorcontrib><creatorcontrib>Rezaei, I. S.</creatorcontrib><collection>IEEE Electronic Library (IEL) Conference Proceedings</collection><collection>IEEE Proceedings Order Plan All Online (POP All Online) 1998-present by volume</collection><collection>IEEE Xplore All Conference Proceedings</collection><collection>IEEE Electronic Library (IEL)</collection><collection>IEEE Proceedings Order Plans (POP All) 1998-Present</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Shekofteh, Y.</au><au>Kabudian, J.</au><au>Goodarzi, M. M.</au><au>Rezaei, I. S.</au><format>book</format><genre>proceeding</genre><ristype>CONF</ristype><atitle>Confidence measure improvement using useful predictor features and support vector machines</atitle><btitle>20th Iranian Conference on Electrical Engineering (ICEE2012)</btitle><stitle>IranianCEE</stitle><date>2012-05</date><risdate>2012</risdate><spage>1168</spage><epage>1171</epage><pages>1168-1171</pages><issn>2164-7054</issn><isbn>1467311499</isbn><isbn>9781467311496</isbn><eisbn>9781467311472</eisbn><eisbn>1467311480</eisbn><eisbn>9781467311489</eisbn><eisbn>1467311472</eisbn><abstract>In traditional keyword spotting (KWS) systems, confidence measure (CM) of each keyword is computed from normalized acoustic likelihoods. In addition to likelihood based scores, some keyword dependent features named predictor features such as duration and prosodic features could be defined to improve the performance of CM. In this paper a discriminative and probabilistic computation of CM based upon some useful predictor features and support vector machines (SVM) is presented for Persian conversational telephone speech KWS. Our experimental results show that higher performance will be achieved by appending utilized predictor features. The proposed CM with linear kernel function of SVM is obtained an improvement about 8.6% in Figure-of-Merit (FOM) of KWS system.</abstract><pub>IEEE</pub><doi>10.1109/IranianCEE.2012.6292531</doi><tpages>4</tpages></addata></record> |
| fulltext | fulltext_linktorsrc |
| identifier | ISSN: 2164-7054 |
| ispartof | 20th Iranian Conference on Electrical Engineering (ICEE2012), 2012, p.1168-1171 |
| issn | 2164-7054 |
| language | eng |
| recordid | cdi_ieee_primary_6292531 |
| source | IEEE Electronic Library (IEL) Conference Proceedings |
| subjects | Acoustics confidence measure Hidden Markov models Ions Kernel keyword spotting predictor feature speech recognition Support vector machines SVM |
| title | Confidence measure improvement using useful predictor features and support vector machines |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-16T17%3A34%3A43IST&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=Confidence%20measure%20improvement%20using%20useful%20predictor%20features%20and%20support%20vector%20machines&rft.btitle=20th%20Iranian%20Conference%20on%20Electrical%20Engineering%20(ICEE2012)&rft.au=Shekofteh,%20Y.&rft.date=2012-05&rft.spage=1168&rft.epage=1171&rft.pages=1168-1171&rft.issn=2164-7054&rft.isbn=1467311499&rft.isbn_list=9781467311496&rft_id=info:doi/10.1109/IranianCEE.2012.6292531&rft_dat=%3Cieee_6IE%3E6292531%3C/ieee_6IE%3E%3Curl%3E%3C/url%3E&rft.eisbn=9781467311472&rft.eisbn_list=1467311480&rft.eisbn_list=9781467311489&rft.eisbn_list=1467311472&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rft_ieee_id=6292531&rfr_iscdi=true |