Online Speaker Clustering Using Incremental Learning of an Ergodic Hidden Markov Model
A novel online speaker clustering method based on a generative model is proposed. It employs an incremental variant of variational Bayesian learning and provides probabilistic (non-deterministic) decisions for each input utterance, on the basis of the history of preceding utterances. It can be expec...
Gespeichert in:
| Veröffentlicht in: | IEICE Transactions on Information and Systems 2012/10/01, Vol.E95.D(10), pp.2469-2478 |
|---|---|
| Hauptverfasser: | , , |
| Format: | Artikel |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | 2478 |
|---|---|
| container_issue | 10 |
| container_start_page | 2469 |
| container_title | IEICE Transactions on Information and Systems |
| container_volume | E95.D |
| creator | KOSHINAKA, Takafumi NAGATOMO, Kentaro SHINODA, Koichi |
| description | A novel online speaker clustering method based on a generative model is proposed. It employs an incremental variant of variational Bayesian learning and provides probabilistic (non-deterministic) decisions for each input utterance, on the basis of the history of preceding utterances. It can be expected to be robust against errors in cluster estimation and the classification of utterances, and hence to be applicable to many real-time applications. Experimental results show that it produces 50% fewer classification errors than does a conventional online method. They also show that it is possible to reduce the number of speech recognition errors by combining the method with unsupervised speaker adaptation. |
| doi_str_mv | 10.1587/transinf.E95.D.2469 |
| format | Article |
| fullrecord | <record><control><sourceid>proquest_cross</sourceid><recordid>TN_cdi_proquest_miscellaneous_1221893669</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>1221893669</sourcerecordid><originalsourceid>FETCH-LOGICAL-c541t-2d4c190563793bfb207fd5b599c3fbe924f03635ab76c1f0696dbb48988f987a3</originalsourceid><addsrcrecordid>eNpdkMtKAzEUhoMoWC9P4CYbwc3UXCaZyVJqvVFx4WUbMpmTGk0zNZkKvr1TqhXcnAOH7z8_fAidUDKmoq7O-2Ri9tGNp0qML8eslGoHjWhVioJySXfRiCgqi1pwto8Ocn4jhNaMihF6eYjBR8CPSzDvkPAkrHIPycc5fs7reRttggXE3gQ8A5Pi-tg5bCKepnnXeotvfNtCxPcmvXef-L5rIRyhPWdChuOffYier6ZPk5ti9nB9O7mYFVaUtC9YW1qqiJC8UrxxDSOVa0UjlLLcNaBY6QiXXJimkpY6IpVsm6asVV07VVeGH6Kzzd9l6j5WkHu98NlCCCZCt8qaMkZrxaVUA8o3qE1dzgmcXia_MOlLU6LXFvWvRT1Y1Jd6bXFInf4UmGxNcANifd5GmSzpgJGBu9twb7k3c9gCJvXeBvj_e6j8K9lC9tUkDZF_A4Sqj9M</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>1221893669</pqid></control><display><type>article</type><title>Online Speaker Clustering Using Incremental Learning of an Ergodic Hidden Markov Model</title><source>Elektronische Zeitschriftenbibliothek - Frei zugängliche E-Journals</source><source>J-STAGE (Japan Science & Technology Information Aggregator, Electronic) Freely Available Titles - Japanese</source><creator>KOSHINAKA, Takafumi ; NAGATOMO, Kentaro ; SHINODA, Koichi</creator><creatorcontrib>KOSHINAKA, Takafumi ; NAGATOMO, Kentaro ; SHINODA, Koichi</creatorcontrib><description>A novel online speaker clustering method based on a generative model is proposed. It employs an incremental variant of variational Bayesian learning and provides probabilistic (non-deterministic) decisions for each input utterance, on the basis of the history of preceding utterances. It can be expected to be robust against errors in cluster estimation and the classification of utterances, and hence to be applicable to many real-time applications. Experimental results show that it produces 50% fewer classification errors than does a conventional online method. They also show that it is possible to reduce the number of speech recognition errors by combining the method with unsupervised speaker adaptation.</description><identifier>ISSN: 0916-8532</identifier><identifier>EISSN: 1745-1361</identifier><identifier>DOI: 10.1587/transinf.E95.D.2469</identifier><language>eng</language><publisher>Oxford: The Institute of Electronics, Information and Communication Engineers</publisher><subject>Applied sciences ; Artificial intelligence ; Classification ; Clustering ; Clusters ; Computer science; control theory; systems ; Errors ; Exact sciences and technology ; HMM ; Information, signal and communications theory ; Learning ; meeting recognition ; model selection ; On-line systems ; Online ; Real time ; Signal and communications theory ; Signal processing ; Signal representation. Spectral analysis ; Signal, noise ; Speech and sound recognition and synthesis. Linguistics ; Speech processing ; Telecommunications and information theory ; variational Bayesian learning</subject><ispartof>IEICE Transactions on Information and Systems, 2012/10/01, Vol.E95.D(10), pp.2469-2478</ispartof><rights>2012 The Institute of Electronics, Information and Communication Engineers</rights><rights>2015 INIST-CNRS</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><cites>FETCH-LOGICAL-c541t-2d4c190563793bfb207fd5b599c3fbe924f03635ab76c1f0696dbb48988f987a3</cites></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>314,780,784,1882,4023,27922,27923,27924</link.rule.ids><backlink>$$Uhttp://pascal-francis.inist.fr/vibad/index.php?action=getRecordDetail&idt=26414690$$DView record in Pascal Francis$$Hfree_for_read</backlink></links><search><creatorcontrib>KOSHINAKA, Takafumi</creatorcontrib><creatorcontrib>NAGATOMO, Kentaro</creatorcontrib><creatorcontrib>SHINODA, Koichi</creatorcontrib><title>Online Speaker Clustering Using Incremental Learning of an Ergodic Hidden Markov Model</title><title>IEICE Transactions on Information and Systems</title><addtitle>IEICE Trans. Inf. & Syst.</addtitle><description>A novel online speaker clustering method based on a generative model is proposed. It employs an incremental variant of variational Bayesian learning and provides probabilistic (non-deterministic) decisions for each input utterance, on the basis of the history of preceding utterances. It can be expected to be robust against errors in cluster estimation and the classification of utterances, and hence to be applicable to many real-time applications. Experimental results show that it produces 50% fewer classification errors than does a conventional online method. They also show that it is possible to reduce the number of speech recognition errors by combining the method with unsupervised speaker adaptation.</description><subject>Applied sciences</subject><subject>Artificial intelligence</subject><subject>Classification</subject><subject>Clustering</subject><subject>Clusters</subject><subject>Computer science; control theory; systems</subject><subject>Errors</subject><subject>Exact sciences and technology</subject><subject>HMM</subject><subject>Information, signal and communications theory</subject><subject>Learning</subject><subject>meeting recognition</subject><subject>model selection</subject><subject>On-line systems</subject><subject>Online</subject><subject>Real time</subject><subject>Signal and communications theory</subject><subject>Signal processing</subject><subject>Signal representation. Spectral analysis</subject><subject>Signal, noise</subject><subject>Speech and sound recognition and synthesis. Linguistics</subject><subject>Speech processing</subject><subject>Telecommunications and information theory</subject><subject>variational Bayesian learning</subject><issn>0916-8532</issn><issn>1745-1361</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2012</creationdate><recordtype>article</recordtype><recordid>eNpdkMtKAzEUhoMoWC9P4CYbwc3UXCaZyVJqvVFx4WUbMpmTGk0zNZkKvr1TqhXcnAOH7z8_fAidUDKmoq7O-2Ri9tGNp0qML8eslGoHjWhVioJySXfRiCgqi1pwto8Ocn4jhNaMihF6eYjBR8CPSzDvkPAkrHIPycc5fs7reRttggXE3gQ8A5Pi-tg5bCKepnnXeotvfNtCxPcmvXef-L5rIRyhPWdChuOffYier6ZPk5ti9nB9O7mYFVaUtC9YW1qqiJC8UrxxDSOVa0UjlLLcNaBY6QiXXJimkpY6IpVsm6asVV07VVeGH6Kzzd9l6j5WkHu98NlCCCZCt8qaMkZrxaVUA8o3qE1dzgmcXia_MOlLU6LXFvWvRT1Y1Jd6bXFInf4UmGxNcANifd5GmSzpgJGBu9twb7k3c9gCJvXeBvj_e6j8K9lC9tUkDZF_A4Sqj9M</recordid><startdate>2012</startdate><enddate>2012</enddate><creator>KOSHINAKA, Takafumi</creator><creator>NAGATOMO, Kentaro</creator><creator>SHINODA, Koichi</creator><general>The Institute of Electronics, Information and Communication Engineers</general><general>Oxford University Press</general><scope>IQODW</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7SC</scope><scope>8FD</scope><scope>JQ2</scope><scope>L7M</scope><scope>L~C</scope><scope>L~D</scope></search><sort><creationdate>2012</creationdate><title>Online Speaker Clustering Using Incremental Learning of an Ergodic Hidden Markov Model</title><author>KOSHINAKA, Takafumi ; NAGATOMO, Kentaro ; SHINODA, Koichi</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c541t-2d4c190563793bfb207fd5b599c3fbe924f03635ab76c1f0696dbb48988f987a3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2012</creationdate><topic>Applied sciences</topic><topic>Artificial intelligence</topic><topic>Classification</topic><topic>Clustering</topic><topic>Clusters</topic><topic>Computer science; control theory; systems</topic><topic>Errors</topic><topic>Exact sciences and technology</topic><topic>HMM</topic><topic>Information, signal and communications theory</topic><topic>Learning</topic><topic>meeting recognition</topic><topic>model selection</topic><topic>On-line systems</topic><topic>Online</topic><topic>Real time</topic><topic>Signal and communications theory</topic><topic>Signal processing</topic><topic>Signal representation. Spectral analysis</topic><topic>Signal, noise</topic><topic>Speech and sound recognition and synthesis. Linguistics</topic><topic>Speech processing</topic><topic>Telecommunications and information theory</topic><topic>variational Bayesian learning</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>KOSHINAKA, Takafumi</creatorcontrib><creatorcontrib>NAGATOMO, Kentaro</creatorcontrib><creatorcontrib>SHINODA, Koichi</creatorcontrib><collection>Pascal-Francis</collection><collection>CrossRef</collection><collection>Computer and Information Systems Abstracts</collection><collection>Technology Research Database</collection><collection>ProQuest Computer Science Collection</collection><collection>Advanced Technologies Database with Aerospace</collection><collection>Computer and Information Systems Abstracts Academic</collection><collection>Computer and Information Systems Abstracts Professional</collection><jtitle>IEICE Transactions on Information and Systems</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>KOSHINAKA, Takafumi</au><au>NAGATOMO, Kentaro</au><au>SHINODA, Koichi</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Online Speaker Clustering Using Incremental Learning of an Ergodic Hidden Markov Model</atitle><jtitle>IEICE Transactions on Information and Systems</jtitle><addtitle>IEICE Trans. Inf. & Syst.</addtitle><date>2012</date><risdate>2012</risdate><volume>E95.D</volume><issue>10</issue><spage>2469</spage><epage>2478</epage><pages>2469-2478</pages><issn>0916-8532</issn><eissn>1745-1361</eissn><abstract>A novel online speaker clustering method based on a generative model is proposed. It employs an incremental variant of variational Bayesian learning and provides probabilistic (non-deterministic) decisions for each input utterance, on the basis of the history of preceding utterances. It can be expected to be robust against errors in cluster estimation and the classification of utterances, and hence to be applicable to many real-time applications. Experimental results show that it produces 50% fewer classification errors than does a conventional online method. They also show that it is possible to reduce the number of speech recognition errors by combining the method with unsupervised speaker adaptation.</abstract><cop>Oxford</cop><pub>The Institute of Electronics, Information and Communication Engineers</pub><doi>10.1587/transinf.E95.D.2469</doi><tpages>10</tpages><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext |
| identifier | ISSN: 0916-8532 |
| ispartof | IEICE Transactions on Information and Systems, 2012/10/01, Vol.E95.D(10), pp.2469-2478 |
| issn | 0916-8532 1745-1361 |
| language | eng |
| recordid | cdi_proquest_miscellaneous_1221893669 |
| source | Elektronische Zeitschriftenbibliothek - Frei zugängliche E-Journals; J-STAGE (Japan Science & Technology Information Aggregator, Electronic) Freely Available Titles - Japanese |
| subjects | Applied sciences Artificial intelligence Classification Clustering Clusters Computer science control theory systems Errors Exact sciences and technology HMM Information, signal and communications theory Learning meeting recognition model selection On-line systems Online Real time Signal and communications theory Signal processing Signal representation. Spectral analysis Signal, noise Speech and sound recognition and synthesis. Linguistics Speech processing Telecommunications and information theory variational Bayesian learning |
| title | Online Speaker Clustering Using Incremental Learning of an Ergodic Hidden Markov Model |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-08T17%3A29%3A44IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_cross&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Online%20Speaker%20Clustering%20Using%20Incremental%20Learning%20of%20an%20Ergodic%20Hidden%20Markov%20Model&rft.jtitle=IEICE%20Transactions%20on%20Information%20and%20Systems&rft.au=KOSHINAKA,%20Takafumi&rft.date=2012&rft.volume=E95.D&rft.issue=10&rft.spage=2469&rft.epage=2478&rft.pages=2469-2478&rft.issn=0916-8532&rft.eissn=1745-1361&rft_id=info:doi/10.1587/transinf.E95.D.2469&rft_dat=%3Cproquest_cross%3E1221893669%3C/proquest_cross%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=1221893669&rft_id=info:pmid/&rfr_iscdi=true |