An Unsupervised Approach to Cochannel Speech Separation
Cochannel (two-talker) speech separation is predominantly addressed using pretrained speaker dependent models. In this paper, we propose an unsupervised approach to separating cochannel speech. Our approach follows the two main stages of computational auditory scene analysis: segmentation and groupi...
Gespeichert in:
| Veröffentlicht in: | IEEE transactions on audio, speech, and language processing speech, and language processing, 2013-01, Vol.21 (1), p.122-131 |
|---|---|
| Hauptverfasser: | , |
| Format: | Artikel |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext bestellen |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | 131 |
|---|---|
| container_issue | 1 |
| container_start_page | 122 |
| container_title | IEEE transactions on audio, speech, and language processing |
| container_volume | 21 |
| creator | Hu, Ke Wang, DeLiang |
| description | Cochannel (two-talker) speech separation is predominantly addressed using pretrained speaker dependent models. In this paper, we propose an unsupervised approach to separating cochannel speech. Our approach follows the two main stages of computational auditory scene analysis: segmentation and grouping. For voiced speech segregation, the proposed system utilizes a tandem algorithm for simultaneous grouping and then unsupervised clustering for sequential grouping. The clustering is performed by a search to maximize the ratio of between- and within-group speaker distances while penalizing within-group concurrent pitches. To segregate unvoiced speech, we first produce unvoiced speech segments based on onset/offset analysis. The segments are grouped using the complementary binary masks of segregated voiced speech. Despite its simplicity, our approach produces significant SNR improvements across a range of input SNR. The proposed system yields competitive performance in comparison to other speaker-independent and model-based methods. |
| doi_str_mv | 10.1109/TASL.2012.2215591 |
| format | Article |
| fullrecord | <record><control><sourceid>proquest_RIE</sourceid><recordid>TN_cdi_ieee_primary_6303834</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>6303834</ieee_id><sourcerecordid>2797714011</sourcerecordid><originalsourceid>FETCH-LOGICAL-c399t-e3ea29ae5571fdb803c25085e68b88619f20747612b298f6eed1bb97fe4cfe383</originalsourceid><addsrcrecordid>eNpdkE9rwzAMxc3YYF23DzB2CYzBLuksO3bsYyj7B4Ud2p6N4yo0JU0yOxns28-lpYednpB-ekiPkHugMwCqX1bFcjFjFNiMMRBCwwWZRFVprll2ea5BXpObEHaUZlxmMCF50SbrNow9-p864CYp-t531m2ToUvmndvatsUmWfaIsbfE3no71F17S64q2wS8O-mUrN9eV_OPdPH1_jkvFqnjWg8pcrRMWxQih2pTKsodE1QJlKpUSoKuGM2zXAIrmVaVRNxAWeq8wsxVyBWfkuejb7zqe8QwmH0dHDaNbbEbgwEOQuaUMhnRx3_orht9G68zAJDpTFBGIwVHyvkuBI-V6X29t_7XADWHKM0hSnOI0pyijDtPJ2cbnG0qb1tXh_Mik0pwzkXkHo5cjYjnseQ0PpLxP1lQeoE</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>1114945020</pqid></control><display><type>article</type><title>An Unsupervised Approach to Cochannel Speech Separation</title><source>IEEE Electronic Library (IEL)</source><creator>Hu, Ke ; Wang, DeLiang</creator><creatorcontrib>Hu, Ke ; Wang, DeLiang</creatorcontrib><description>Cochannel (two-talker) speech separation is predominantly addressed using pretrained speaker dependent models. In this paper, we propose an unsupervised approach to separating cochannel speech. Our approach follows the two main stages of computational auditory scene analysis: segmentation and grouping. For voiced speech segregation, the proposed system utilizes a tandem algorithm for simultaneous grouping and then unsupervised clustering for sequential grouping. The clustering is performed by a search to maximize the ratio of between- and within-group speaker distances while penalizing within-group concurrent pitches. To segregate unvoiced speech, we first produce unvoiced speech segments based on onset/offset analysis. The segments are grouped using the complementary binary masks of segregated voiced speech. Despite its simplicity, our approach produces significant SNR improvements across a range of input SNR. The proposed system yields competitive performance in comparison to other speaker-independent and model-based methods.</description><identifier>ISSN: 1558-7916</identifier><identifier>ISSN: 2329-9290</identifier><identifier>EISSN: 1558-7924</identifier><identifier>EISSN: 2329-9304</identifier><identifier>DOI: 10.1109/TASL.2012.2215591</identifier><identifier>CODEN: ITASD8</identifier><language>eng</language><publisher>Piscataway, NJ: IEEE</publisher><subject>Algorithm design and analysis ; Applied sciences ; Clustering ; Clustering algorithms ; cochannel speech separation ; Computational auditory scene analysis (CASA) ; Computational modeling ; Exact sciences and technology ; Hidden Markov models ; Image processing ; Information, signal and communications theory ; Mathematical models ; Natural language processing ; Scene analysis ; Segments ; Separation ; sequential grouping ; Signal and communications theory ; Signal processing ; Signal representation. Spectral analysis ; Signal to noise ratio ; Signal, noise ; Speech ; Studies ; Telecommunications and information theory ; Time frequency analysis ; Transaction processing ; unsupervised clustering ; unvoiced speech segregation</subject><ispartof>IEEE transactions on audio, speech, and language processing, 2013-01, Vol.21 (1), p.122-131</ispartof><rights>2014 INIST-CNRS</rights><rights>Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) Jan 2013</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c399t-e3ea29ae5571fdb803c25085e68b88619f20747612b298f6eed1bb97fe4cfe383</citedby><cites>FETCH-LOGICAL-c399t-e3ea29ae5571fdb803c25085e68b88619f20747612b298f6eed1bb97fe4cfe383</cites></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/6303834$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>314,776,780,792,4010,27900,27901,27902,54733</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/6303834$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc><backlink>$$Uhttp://pascal-francis.inist.fr/vibad/index.php?action=getRecordDetail&idt=26853335$$DView record in Pascal Francis$$Hfree_for_read</backlink></links><search><creatorcontrib>Hu, Ke</creatorcontrib><creatorcontrib>Wang, DeLiang</creatorcontrib><title>An Unsupervised Approach to Cochannel Speech Separation</title><title>IEEE transactions on audio, speech, and language processing</title><addtitle>TASL</addtitle><description>Cochannel (two-talker) speech separation is predominantly addressed using pretrained speaker dependent models. In this paper, we propose an unsupervised approach to separating cochannel speech. Our approach follows the two main stages of computational auditory scene analysis: segmentation and grouping. For voiced speech segregation, the proposed system utilizes a tandem algorithm for simultaneous grouping and then unsupervised clustering for sequential grouping. The clustering is performed by a search to maximize the ratio of between- and within-group speaker distances while penalizing within-group concurrent pitches. To segregate unvoiced speech, we first produce unvoiced speech segments based on onset/offset analysis. The segments are grouped using the complementary binary masks of segregated voiced speech. Despite its simplicity, our approach produces significant SNR improvements across a range of input SNR. The proposed system yields competitive performance in comparison to other speaker-independent and model-based methods.</description><subject>Algorithm design and analysis</subject><subject>Applied sciences</subject><subject>Clustering</subject><subject>Clustering algorithms</subject><subject>cochannel speech separation</subject><subject>Computational auditory scene analysis (CASA)</subject><subject>Computational modeling</subject><subject>Exact sciences and technology</subject><subject>Hidden Markov models</subject><subject>Image processing</subject><subject>Information, signal and communications theory</subject><subject>Mathematical models</subject><subject>Natural language processing</subject><subject>Scene analysis</subject><subject>Segments</subject><subject>Separation</subject><subject>sequential grouping</subject><subject>Signal and communications theory</subject><subject>Signal processing</subject><subject>Signal representation. Spectral analysis</subject><subject>Signal to noise ratio</subject><subject>Signal, noise</subject><subject>Speech</subject><subject>Studies</subject><subject>Telecommunications and information theory</subject><subject>Time frequency analysis</subject><subject>Transaction processing</subject><subject>unsupervised clustering</subject><subject>unvoiced speech segregation</subject><issn>1558-7916</issn><issn>2329-9290</issn><issn>1558-7924</issn><issn>2329-9304</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2013</creationdate><recordtype>article</recordtype><sourceid>RIE</sourceid><recordid>eNpdkE9rwzAMxc3YYF23DzB2CYzBLuksO3bsYyj7B4Ud2p6N4yo0JU0yOxns28-lpYednpB-ekiPkHugMwCqX1bFcjFjFNiMMRBCwwWZRFVprll2ea5BXpObEHaUZlxmMCF50SbrNow9-p864CYp-t531m2ToUvmndvatsUmWfaIsbfE3no71F17S64q2wS8O-mUrN9eV_OPdPH1_jkvFqnjWg8pcrRMWxQih2pTKsodE1QJlKpUSoKuGM2zXAIrmVaVRNxAWeq8wsxVyBWfkuejb7zqe8QwmH0dHDaNbbEbgwEOQuaUMhnRx3_orht9G68zAJDpTFBGIwVHyvkuBI-V6X29t_7XADWHKM0hSnOI0pyijDtPJ2cbnG0qb1tXh_Mik0pwzkXkHo5cjYjnseQ0PpLxP1lQeoE</recordid><startdate>201301</startdate><enddate>201301</enddate><creator>Hu, Ke</creator><creator>Wang, DeLiang</creator><general>IEEE</general><general>Institute of Electrical and Electronics Engineers</general><general>The Institute of Electrical and Electronics Engineers, Inc. (IEEE)</general><scope>97E</scope><scope>RIA</scope><scope>RIE</scope><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>201301</creationdate><title>An Unsupervised Approach to Cochannel Speech Separation</title><author>Hu, Ke ; Wang, DeLiang</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c399t-e3ea29ae5571fdb803c25085e68b88619f20747612b298f6eed1bb97fe4cfe383</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2013</creationdate><topic>Algorithm design and analysis</topic><topic>Applied sciences</topic><topic>Clustering</topic><topic>Clustering algorithms</topic><topic>cochannel speech separation</topic><topic>Computational auditory scene analysis (CASA)</topic><topic>Computational modeling</topic><topic>Exact sciences and technology</topic><topic>Hidden Markov models</topic><topic>Image processing</topic><topic>Information, signal and communications theory</topic><topic>Mathematical models</topic><topic>Natural language processing</topic><topic>Scene analysis</topic><topic>Segments</topic><topic>Separation</topic><topic>sequential grouping</topic><topic>Signal and communications theory</topic><topic>Signal processing</topic><topic>Signal representation. Spectral analysis</topic><topic>Signal to noise ratio</topic><topic>Signal, noise</topic><topic>Speech</topic><topic>Studies</topic><topic>Telecommunications and information theory</topic><topic>Time frequency analysis</topic><topic>Transaction processing</topic><topic>unsupervised clustering</topic><topic>unvoiced speech segregation</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Hu, Ke</creatorcontrib><creatorcontrib>Wang, DeLiang</creatorcontrib><collection>IEEE All-Society Periodicals Package (ASPP) 2005-present</collection><collection>IEEE All-Society Periodicals Package (ASPP) 1998-Present</collection><collection>IEEE Electronic Library (IEL)</collection><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>IEEE transactions on audio, speech, and language processing</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Hu, Ke</au><au>Wang, DeLiang</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>An Unsupervised Approach to Cochannel Speech Separation</atitle><jtitle>IEEE transactions on audio, speech, and language processing</jtitle><stitle>TASL</stitle><date>2013-01</date><risdate>2013</risdate><volume>21</volume><issue>1</issue><spage>122</spage><epage>131</epage><pages>122-131</pages><issn>1558-7916</issn><issn>2329-9290</issn><eissn>1558-7924</eissn><eissn>2329-9304</eissn><coden>ITASD8</coden><abstract>Cochannel (two-talker) speech separation is predominantly addressed using pretrained speaker dependent models. In this paper, we propose an unsupervised approach to separating cochannel speech. Our approach follows the two main stages of computational auditory scene analysis: segmentation and grouping. For voiced speech segregation, the proposed system utilizes a tandem algorithm for simultaneous grouping and then unsupervised clustering for sequential grouping. The clustering is performed by a search to maximize the ratio of between- and within-group speaker distances while penalizing within-group concurrent pitches. To segregate unvoiced speech, we first produce unvoiced speech segments based on onset/offset analysis. The segments are grouped using the complementary binary masks of segregated voiced speech. Despite its simplicity, our approach produces significant SNR improvements across a range of input SNR. The proposed system yields competitive performance in comparison to other speaker-independent and model-based methods.</abstract><cop>Piscataway, NJ</cop><pub>IEEE</pub><doi>10.1109/TASL.2012.2215591</doi><tpages>10</tpages><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext_linktorsrc |
| identifier | ISSN: 1558-7916 |
| ispartof | IEEE transactions on audio, speech, and language processing, 2013-01, Vol.21 (1), p.122-131 |
| issn | 1558-7916 2329-9290 1558-7924 2329-9304 |
| language | eng |
| recordid | cdi_ieee_primary_6303834 |
| source | IEEE Electronic Library (IEL) |
| subjects | Algorithm design and analysis Applied sciences Clustering Clustering algorithms cochannel speech separation Computational auditory scene analysis (CASA) Computational modeling Exact sciences and technology Hidden Markov models Image processing Information, signal and communications theory Mathematical models Natural language processing Scene analysis Segments Separation sequential grouping Signal and communications theory Signal processing Signal representation. Spectral analysis Signal to noise ratio Signal, noise Speech Studies Telecommunications and information theory Time frequency analysis Transaction processing unsupervised clustering unvoiced speech segregation |
| title | An Unsupervised Approach to Cochannel Speech Separation |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-28T20%3A29%3A18IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_RIE&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=An%20Unsupervised%20Approach%20to%20Cochannel%20Speech%20Separation&rft.jtitle=IEEE%20transactions%20on%20audio,%20speech,%20and%20language%20processing&rft.au=Hu,%20Ke&rft.date=2013-01&rft.volume=21&rft.issue=1&rft.spage=122&rft.epage=131&rft.pages=122-131&rft.issn=1558-7916&rft.eissn=1558-7924&rft.coden=ITASD8&rft_id=info:doi/10.1109/TASL.2012.2215591&rft_dat=%3Cproquest_RIE%3E2797714011%3C/proquest_RIE%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=1114945020&rft_id=info:pmid/&rft_ieee_id=6303834&rfr_iscdi=true |