Adaptive Speaker Recognition Based on Hidden Markov Model Parameter Optimization
The Hidden Markov Model (HMM) is a widely used method for speaker recognition. During its training, the composite order of the measurement probability matrix and the number of re-evaluations of the initial model affect the speed and accuracy of a recognition system. However, theoretical analysis and...
Gespeichert in:
| Veröffentlicht in: | IEEE access 2020, Vol.8, p.34942-34948 |
|---|---|
| 1. Verfasser: | |
| Format: | Artikel |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | 34948 |
|---|---|
| container_issue | |
| container_start_page | 34942 |
| container_title | IEEE access |
| container_volume | 8 |
| creator | Wei, Yangjie |
| description | The Hidden Markov Model (HMM) is a widely used method for speaker recognition. During its training, the composite order of the measurement probability matrix and the number of re-evaluations of the initial model affect the speed and accuracy of a recognition system. However, theoretical analysis and related quantitative methods are rarely used for adaptively acquiring them. In this paper, a quantitative method for adaptively selecting the optimal composite order and the optimal number of re-evaluations is proposed based on theoretical analysis and experimental results. First, the standard deviation (SD) is introduced to calculate the recognition rate considering its relationship with Mel frequency cepstrum coefficients (MFCCs) dimension, then the composite order is optimized according to its relationship curve with the SD. Second, the composited measurement probability with different number of re-evaluations is calculated and the number of re-evaluations is optimized when a convergence condition is satisfied. Experiments show that the recognition rate with the optimal composite order obtained in this paper is 97.02%, and the recognition rate with the optimal number of re-evaluations is 98.9%. |
| doi_str_mv | 10.1109/ACCESS.2020.2972511 |
| format | Article |
| fullrecord | <record><control><sourceid>proquest_cross</sourceid><recordid>TN_cdi_proquest_journals_2454872378</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>8995577</ieee_id><doaj_id>oai_doaj_org_article_aaaff5dbd8df498ea6cefde427e47c23</doaj_id><sourcerecordid>2454872378</sourcerecordid><originalsourceid>FETCH-LOGICAL-c408t-79e51149c8db0bccf2571f85d3664006620839fbb8dc220b3ce0b6b29192ac3b3</originalsourceid><addsrcrecordid>eNpNkVtLAzEQhRdRUNRf4MuCz625bDbJYy3eQGmx-hwmyURS26ZmV0F_vakr4rzMcJjvzMCpqjNKxpQSfTGZTq8WizEjjIyZlkxQulcdMdrqERe83f83H1anXbckpVSRhDyq5hMP2z5-YL3YIrxirh_RpZdN7GPa1JfQoa_LcBu9x039APk1fdQPyeOqnkOGNfYFmRWHdfyCHXNSHQRYdXj624-r5-urp-nt6H52czed3I9cQ1Q_khrLn412yltinQtMSBqU8LxtG0LalhHFdbBWeccYsdwhsa1lmmoGjlt-XN0Nvj7B0mxzXEP-NAmi-RFSfjGQ--hWaAAgBOGtVz40WiG0DoPHhklspGO8eJ0PXtuc3t6x680yvedNed-wRjRKMi5V2eLDlsup6zKGv6uUmF0SZkjC7JIwv0kU6mygIiL-EUprIaTk35G1hNI</addsrcrecordid><sourcetype>Open Website</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2454872378</pqid></control><display><type>article</type><title>Adaptive Speaker Recognition Based on Hidden Markov Model Parameter Optimization</title><source>IEEE Open Access Journals</source><source>DOAJ Directory of Open Access Journals</source><source>Elektronische Zeitschriftenbibliothek - Frei zugängliche E-Journals</source><creator>Wei, Yangjie</creator><creatorcontrib>Wei, Yangjie</creatorcontrib><description>The Hidden Markov Model (HMM) is a widely used method for speaker recognition. During its training, the composite order of the measurement probability matrix and the number of re-evaluations of the initial model affect the speed and accuracy of a recognition system. However, theoretical analysis and related quantitative methods are rarely used for adaptively acquiring them. In this paper, a quantitative method for adaptively selecting the optimal composite order and the optimal number of re-evaluations is proposed based on theoretical analysis and experimental results. First, the standard deviation (SD) is introduced to calculate the recognition rate considering its relationship with Mel frequency cepstrum coefficients (MFCCs) dimension, then the composite order is optimized according to its relationship curve with the SD. Second, the composited measurement probability with different number of re-evaluations is calculated and the number of re-evaluations is optimized when a convergence condition is satisfied. Experiments show that the recognition rate with the optimal composite order obtained in this paper is 97.02%, and the recognition rate with the optimal number of re-evaluations is 98.9%.</description><identifier>ISSN: 2169-3536</identifier><identifier>EISSN: 2169-3536</identifier><identifier>DOI: 10.1109/ACCESS.2020.2972511</identifier><identifier>CODEN: IAECCG</identifier><language>eng</language><publisher>Piscataway: IEEE</publisher><subject>Adaptation models ; Evaluation ; Feature extraction ; Gaussian composite order ; Hidden Markov models ; Markov chains ; Model accuracy ; Optimization ; parameter optimization ; Quantitative analysis ; Radio spectrum management ; re-evaluation ; Speaker recognition ; Speech recognition ; Training</subject><ispartof>IEEE access, 2020, Vol.8, p.34942-34948</ispartof><rights>Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2020</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c408t-79e51149c8db0bccf2571f85d3664006620839fbb8dc220b3ce0b6b29192ac3b3</citedby><cites>FETCH-LOGICAL-c408t-79e51149c8db0bccf2571f85d3664006620839fbb8dc220b3ce0b6b29192ac3b3</cites><orcidid>0000-0001-6615-5484</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/8995577$$EHTML$$P50$$Gieee$$Hfree_for_read</linktohtml><link.rule.ids>314,776,780,860,2096,4010,27610,27900,27901,27902,54908</link.rule.ids></links><search><creatorcontrib>Wei, Yangjie</creatorcontrib><title>Adaptive Speaker Recognition Based on Hidden Markov Model Parameter Optimization</title><title>IEEE access</title><addtitle>Access</addtitle><description>The Hidden Markov Model (HMM) is a widely used method for speaker recognition. During its training, the composite order of the measurement probability matrix and the number of re-evaluations of the initial model affect the speed and accuracy of a recognition system. However, theoretical analysis and related quantitative methods are rarely used for adaptively acquiring them. In this paper, a quantitative method for adaptively selecting the optimal composite order and the optimal number of re-evaluations is proposed based on theoretical analysis and experimental results. First, the standard deviation (SD) is introduced to calculate the recognition rate considering its relationship with Mel frequency cepstrum coefficients (MFCCs) dimension, then the composite order is optimized according to its relationship curve with the SD. Second, the composited measurement probability with different number of re-evaluations is calculated and the number of re-evaluations is optimized when a convergence condition is satisfied. Experiments show that the recognition rate with the optimal composite order obtained in this paper is 97.02%, and the recognition rate with the optimal number of re-evaluations is 98.9%.</description><subject>Adaptation models</subject><subject>Evaluation</subject><subject>Feature extraction</subject><subject>Gaussian composite order</subject><subject>Hidden Markov models</subject><subject>Markov chains</subject><subject>Model accuracy</subject><subject>Optimization</subject><subject>parameter optimization</subject><subject>Quantitative analysis</subject><subject>Radio spectrum management</subject><subject>re-evaluation</subject><subject>Speaker recognition</subject><subject>Speech recognition</subject><subject>Training</subject><issn>2169-3536</issn><issn>2169-3536</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2020</creationdate><recordtype>article</recordtype><sourceid>ESBDL</sourceid><sourceid>RIE</sourceid><sourceid>DOA</sourceid><recordid>eNpNkVtLAzEQhRdRUNRf4MuCz625bDbJYy3eQGmx-hwmyURS26ZmV0F_vakr4rzMcJjvzMCpqjNKxpQSfTGZTq8WizEjjIyZlkxQulcdMdrqERe83f83H1anXbckpVSRhDyq5hMP2z5-YL3YIrxirh_RpZdN7GPa1JfQoa_LcBu9x039APk1fdQPyeOqnkOGNfYFmRWHdfyCHXNSHQRYdXj624-r5-urp-nt6H52czed3I9cQ1Q_khrLn412yltinQtMSBqU8LxtG0LalhHFdbBWeccYsdwhsa1lmmoGjlt-XN0Nvj7B0mxzXEP-NAmi-RFSfjGQ--hWaAAgBOGtVz40WiG0DoPHhklspGO8eJ0PXtuc3t6x680yvedNed-wRjRKMi5V2eLDlsup6zKGv6uUmF0SZkjC7JIwv0kU6mygIiL-EUprIaTk35G1hNI</recordid><startdate>2020</startdate><enddate>2020</enddate><creator>Wei, Yangjie</creator><general>IEEE</general><general>The Institute of Electrical and Electronics Engineers, Inc. (IEEE)</general><scope>97E</scope><scope>ESBDL</scope><scope>RIA</scope><scope>RIE</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7SC</scope><scope>7SP</scope><scope>7SR</scope><scope>8BQ</scope><scope>8FD</scope><scope>JG9</scope><scope>JQ2</scope><scope>L7M</scope><scope>L~C</scope><scope>L~D</scope><scope>DOA</scope><orcidid>https://orcid.org/0000-0001-6615-5484</orcidid></search><sort><creationdate>2020</creationdate><title>Adaptive Speaker Recognition Based on Hidden Markov Model Parameter Optimization</title><author>Wei, Yangjie</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c408t-79e51149c8db0bccf2571f85d3664006620839fbb8dc220b3ce0b6b29192ac3b3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2020</creationdate><topic>Adaptation models</topic><topic>Evaluation</topic><topic>Feature extraction</topic><topic>Gaussian composite order</topic><topic>Hidden Markov models</topic><topic>Markov chains</topic><topic>Model accuracy</topic><topic>Optimization</topic><topic>parameter optimization</topic><topic>Quantitative analysis</topic><topic>Radio spectrum management</topic><topic>re-evaluation</topic><topic>Speaker recognition</topic><topic>Speech recognition</topic><topic>Training</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Wei, Yangjie</creatorcontrib><collection>IEEE All-Society Periodicals Package (ASPP) 2005-present</collection><collection>IEEE Open Access Journals</collection><collection>IEEE All-Society Periodicals Package (ASPP) 1998-Present</collection><collection>IEEE Electronic Library (IEL)</collection><collection>CrossRef</collection><collection>Computer and Information Systems Abstracts</collection><collection>Electronics & Communications Abstracts</collection><collection>Engineered Materials Abstracts</collection><collection>METADEX</collection><collection>Technology Research Database</collection><collection>Materials 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><collection>DOAJ Directory of Open Access Journals</collection><jtitle>IEEE access</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Wei, Yangjie</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Adaptive Speaker Recognition Based on Hidden Markov Model Parameter Optimization</atitle><jtitle>IEEE access</jtitle><stitle>Access</stitle><date>2020</date><risdate>2020</risdate><volume>8</volume><spage>34942</spage><epage>34948</epage><pages>34942-34948</pages><issn>2169-3536</issn><eissn>2169-3536</eissn><coden>IAECCG</coden><abstract>The Hidden Markov Model (HMM) is a widely used method for speaker recognition. During its training, the composite order of the measurement probability matrix and the number of re-evaluations of the initial model affect the speed and accuracy of a recognition system. However, theoretical analysis and related quantitative methods are rarely used for adaptively acquiring them. In this paper, a quantitative method for adaptively selecting the optimal composite order and the optimal number of re-evaluations is proposed based on theoretical analysis and experimental results. First, the standard deviation (SD) is introduced to calculate the recognition rate considering its relationship with Mel frequency cepstrum coefficients (MFCCs) dimension, then the composite order is optimized according to its relationship curve with the SD. Second, the composited measurement probability with different number of re-evaluations is calculated and the number of re-evaluations is optimized when a convergence condition is satisfied. Experiments show that the recognition rate with the optimal composite order obtained in this paper is 97.02%, and the recognition rate with the optimal number of re-evaluations is 98.9%.</abstract><cop>Piscataway</cop><pub>IEEE</pub><doi>10.1109/ACCESS.2020.2972511</doi><tpages>7</tpages><orcidid>https://orcid.org/0000-0001-6615-5484</orcidid><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext |
| identifier | ISSN: 2169-3536 |
| ispartof | IEEE access, 2020, Vol.8, p.34942-34948 |
| issn | 2169-3536 2169-3536 |
| language | eng |
| recordid | cdi_proquest_journals_2454872378 |
| source | IEEE Open Access Journals; DOAJ Directory of Open Access Journals; Elektronische Zeitschriftenbibliothek - Frei zugängliche E-Journals |
| subjects | Adaptation models Evaluation Feature extraction Gaussian composite order Hidden Markov models Markov chains Model accuracy Optimization parameter optimization Quantitative analysis Radio spectrum management re-evaluation Speaker recognition Speech recognition Training |
| title | Adaptive Speaker Recognition Based on Hidden Markov Model Parameter Optimization |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-13T03%3A22%3A02IST&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=Adaptive%20Speaker%20Recognition%20Based%20on%20Hidden%20Markov%20Model%20Parameter%20Optimization&rft.jtitle=IEEE%20access&rft.au=Wei,%20Yangjie&rft.date=2020&rft.volume=8&rft.spage=34942&rft.epage=34948&rft.pages=34942-34948&rft.issn=2169-3536&rft.eissn=2169-3536&rft.coden=IAECCG&rft_id=info:doi/10.1109/ACCESS.2020.2972511&rft_dat=%3Cproquest_cross%3E2454872378%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=2454872378&rft_id=info:pmid/&rft_ieee_id=8995577&rft_doaj_id=oai_doaj_org_article_aaaff5dbd8df498ea6cefde427e47c23&rfr_iscdi=true |