Speech recognition with segmental conditional random fields: a summary of the JHU CLSP 2010 summer workshop
This paper summarizes the 2010 CLSP Summer Workshop on speech recognition at Johns Hopkins University. The key theme of the workshop was to improve on state-of-the-art speech recognition systems by using Segmental Conditional Random Fields (SCRFs) to integrate multiple types of information. This app...
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , , , , , , , |
|---|---|
| Format: | Tagungsbericht |
| Sprache: | eng |
| Online-Zugang: | Volltext bestellen |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | 5047 |
|---|---|
| container_issue | |
| container_start_page | 5044 |
| container_title | |
| container_volume | |
| creator | Zweig, G Nguyen, P Van Compernolle, Dirk Demuynck, Kris Atlas, L Clark, P Sell, G Wang, M Sha, F Hermansky, H Karakos, D Jansen, A Thomas, S Sivaram, G.S.V.S Bowman, S Kao, J |
| description | This paper summarizes the 2010 CLSP Summer Workshop on speech recognition at Johns Hopkins University. The key theme of the workshop was to improve on state-of-the-art speech recognition systems by using Segmental Conditional Random Fields (SCRFs) to integrate multiple types of information. This approach uses a state-of-the-art baseline as a springboard from which to add a suite of novel features including ones derived from acoustic templates, deep neural net phoneme detections, duration models, modulation features, and whole word point-process models. The SCRF framework is able to appropriately weight these different information sources to produce significant gains on both the Broadcast News and Wall Street Journal tasks. |
| format | Conference Proceeding |
| fullrecord | <record><control><sourceid>kuleuven_6IE</sourceid><recordid>TN_cdi_kuleuven_dspace_123456789_344395</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>123456789_344395</sourcerecordid><originalsourceid>FETCH-kuleuven_dspace_123456789_3443953</originalsourceid><addsrcrecordid>eNqVitFqwjAYRgM6mGjf4b_bxRCSJmmMtzIR8WLgvC6h_Wu7tklJUt3eXhEfYPtuzgfnTEii1YoJqRSVXKspmTGZ0mXGhH4lSQjf9L4sVUrqGWmPA2JRg8fCnW0TG2fh2sQaAp57tNF0UDhbPsT9e2NL10PVYFeGNRgIY98b_wuuglgj7Hcn2ByOn5BSRh8SPVydb0PthgV5qUwXMHlyTt62H1-b3bIdOxwvaPMyDKbAnKVcyEytdM6F4Fry_5Tvfyvz-BP5DXNtWT0</addsrcrecordid><sourcetype>Institutional Repository</sourcetype><iscdi>true</iscdi><recordtype>conference_proceeding</recordtype></control><display><type>conference_proceeding</type><title>Speech recognition with segmental conditional random fields: a summary of the JHU CLSP 2010 summer workshop</title><source>IEEE Electronic Library (IEL) Conference Proceedings</source><creator>Zweig, G ; Nguyen, P ; Van Compernolle, Dirk ; Demuynck, Kris ; Atlas, L ; Clark, P ; Sell, G ; Wang, M ; Sha, F ; Hermansky, H ; Karakos, D ; Jansen, A ; Thomas, S ; Sivaram, G.S.V.S ; Bowman, S ; Kao, J</creator><creatorcontrib>Zweig, G ; Nguyen, P ; Van Compernolle, Dirk ; Demuynck, Kris ; Atlas, L ; Clark, P ; Sell, G ; Wang, M ; Sha, F ; Hermansky, H ; Karakos, D ; Jansen, A ; Thomas, S ; Sivaram, G.S.V.S ; Bowman, S ; Kao, J</creatorcontrib><description>This paper summarizes the 2010 CLSP Summer Workshop on speech recognition at Johns Hopkins University. The key theme of the workshop was to improve on state-of-the-art speech recognition systems by using Segmental Conditional Random Fields (SCRFs) to integrate multiple types of information. This approach uses a state-of-the-art baseline as a springboard from which to add a suite of novel features including ones derived from acoustic templates, deep neural net phoneme detections, duration models, modulation features, and whole word point-process models. The SCRF framework is able to appropriately weight these different information sources to produce significant gains on both the Broadcast News and Wall Street Journal tasks.</description><identifier>ISSN: 1520-6149</identifier><identifier>ISBN: 9781457705397</identifier><identifier>ISBN: 1457705397</identifier><language>eng</language><publisher>345 E 47TH ST, NEW YORK, NY 10017 USA: Ieee</publisher><ispartof>Proceedings 36th international conference on acoustics, speech and signal processing - ICASSP'2011, 2011, p.5044-5047</ispartof><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>309,310,315,776,780,785,786,4036,4037,27837</link.rule.ids><linktorsrc>$$Uhttps://lirias.kuleuven.be/handle/123456789/344395$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc></links><search><creatorcontrib>Zweig, G</creatorcontrib><creatorcontrib>Nguyen, P</creatorcontrib><creatorcontrib>Van Compernolle, Dirk</creatorcontrib><creatorcontrib>Demuynck, Kris</creatorcontrib><creatorcontrib>Atlas, L</creatorcontrib><creatorcontrib>Clark, P</creatorcontrib><creatorcontrib>Sell, G</creatorcontrib><creatorcontrib>Wang, M</creatorcontrib><creatorcontrib>Sha, F</creatorcontrib><creatorcontrib>Hermansky, H</creatorcontrib><creatorcontrib>Karakos, D</creatorcontrib><creatorcontrib>Jansen, A</creatorcontrib><creatorcontrib>Thomas, S</creatorcontrib><creatorcontrib>Sivaram, G.S.V.S</creatorcontrib><creatorcontrib>Bowman, S</creatorcontrib><creatorcontrib>Kao, J</creatorcontrib><title>Speech recognition with segmental conditional random fields: a summary of the JHU CLSP 2010 summer workshop</title><title>Proceedings 36th international conference on acoustics, speech and signal processing - ICASSP'2011</title><description>This paper summarizes the 2010 CLSP Summer Workshop on speech recognition at Johns Hopkins University. The key theme of the workshop was to improve on state-of-the-art speech recognition systems by using Segmental Conditional Random Fields (SCRFs) to integrate multiple types of information. This approach uses a state-of-the-art baseline as a springboard from which to add a suite of novel features including ones derived from acoustic templates, deep neural net phoneme detections, duration models, modulation features, and whole word point-process models. The SCRF framework is able to appropriately weight these different information sources to produce significant gains on both the Broadcast News and Wall Street Journal tasks.</description><issn>1520-6149</issn><isbn>9781457705397</isbn><isbn>1457705397</isbn><fulltext>true</fulltext><rsrctype>conference_proceeding</rsrctype><creationdate>2011</creationdate><recordtype>conference_proceeding</recordtype><sourceid>FZOIL</sourceid><recordid>eNqVitFqwjAYRgM6mGjf4b_bxRCSJmmMtzIR8WLgvC6h_Wu7tklJUt3eXhEfYPtuzgfnTEii1YoJqRSVXKspmTGZ0mXGhH4lSQjf9L4sVUrqGWmPA2JRg8fCnW0TG2fh2sQaAp57tNF0UDhbPsT9e2NL10PVYFeGNRgIY98b_wuuglgj7Hcn2ByOn5BSRh8SPVydb0PthgV5qUwXMHlyTt62H1-b3bIdOxwvaPMyDKbAnKVcyEytdM6F4Fry_5Tvfyvz-BP5DXNtWT0</recordid><startdate>2011</startdate><enddate>2011</enddate><creator>Zweig, G</creator><creator>Nguyen, P</creator><creator>Van Compernolle, Dirk</creator><creator>Demuynck, Kris</creator><creator>Atlas, L</creator><creator>Clark, P</creator><creator>Sell, G</creator><creator>Wang, M</creator><creator>Sha, F</creator><creator>Hermansky, H</creator><creator>Karakos, D</creator><creator>Jansen, A</creator><creator>Thomas, S</creator><creator>Sivaram, G.S.V.S</creator><creator>Bowman, S</creator><creator>Kao, J</creator><general>Ieee</general><scope>FZOIL</scope></search><sort><creationdate>2011</creationdate><title>Speech recognition with segmental conditional random fields: a summary of the JHU CLSP 2010 summer workshop</title><author>Zweig, G ; Nguyen, P ; Van Compernolle, Dirk ; Demuynck, Kris ; Atlas, L ; Clark, P ; Sell, G ; Wang, M ; Sha, F ; Hermansky, H ; Karakos, D ; Jansen, A ; Thomas, S ; Sivaram, G.S.V.S ; Bowman, S ; Kao, J</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-kuleuven_dspace_123456789_3443953</frbrgroupid><rsrctype>conference_proceedings</rsrctype><prefilter>conference_proceedings</prefilter><language>eng</language><creationdate>2011</creationdate><toplevel>online_resources</toplevel><creatorcontrib>Zweig, G</creatorcontrib><creatorcontrib>Nguyen, P</creatorcontrib><creatorcontrib>Van Compernolle, Dirk</creatorcontrib><creatorcontrib>Demuynck, Kris</creatorcontrib><creatorcontrib>Atlas, L</creatorcontrib><creatorcontrib>Clark, P</creatorcontrib><creatorcontrib>Sell, G</creatorcontrib><creatorcontrib>Wang, M</creatorcontrib><creatorcontrib>Sha, F</creatorcontrib><creatorcontrib>Hermansky, H</creatorcontrib><creatorcontrib>Karakos, D</creatorcontrib><creatorcontrib>Jansen, A</creatorcontrib><creatorcontrib>Thomas, S</creatorcontrib><creatorcontrib>Sivaram, G.S.V.S</creatorcontrib><creatorcontrib>Bowman, S</creatorcontrib><creatorcontrib>Kao, J</creatorcontrib><collection>Lirias (KU Leuven Association)</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Zweig, G</au><au>Nguyen, P</au><au>Van Compernolle, Dirk</au><au>Demuynck, Kris</au><au>Atlas, L</au><au>Clark, P</au><au>Sell, G</au><au>Wang, M</au><au>Sha, F</au><au>Hermansky, H</au><au>Karakos, D</au><au>Jansen, A</au><au>Thomas, S</au><au>Sivaram, G.S.V.S</au><au>Bowman, S</au><au>Kao, J</au><format>book</format><genre>proceeding</genre><ristype>CONF</ristype><atitle>Speech recognition with segmental conditional random fields: a summary of the JHU CLSP 2010 summer workshop</atitle><btitle>Proceedings 36th international conference on acoustics, speech and signal processing - ICASSP'2011</btitle><date>2011</date><risdate>2011</risdate><spage>5044</spage><epage>5047</epage><pages>5044-5047</pages><issn>1520-6149</issn><isbn>9781457705397</isbn><isbn>1457705397</isbn><abstract>This paper summarizes the 2010 CLSP Summer Workshop on speech recognition at Johns Hopkins University. The key theme of the workshop was to improve on state-of-the-art speech recognition systems by using Segmental Conditional Random Fields (SCRFs) to integrate multiple types of information. This approach uses a state-of-the-art baseline as a springboard from which to add a suite of novel features including ones derived from acoustic templates, deep neural net phoneme detections, duration models, modulation features, and whole word point-process models. The SCRF framework is able to appropriately weight these different information sources to produce significant gains on both the Broadcast News and Wall Street Journal tasks.</abstract><cop>345 E 47TH ST, NEW YORK, NY 10017 USA</cop><pub>Ieee</pub></addata></record> |
| fulltext | fulltext_linktorsrc |
| identifier | ISSN: 1520-6149 |
| ispartof | Proceedings 36th international conference on acoustics, speech and signal processing - ICASSP'2011, 2011, p.5044-5047 |
| issn | 1520-6149 |
| language | eng |
| recordid | cdi_kuleuven_dspace_123456789_344395 |
| source | IEEE Electronic Library (IEL) Conference Proceedings |
| title | Speech recognition with segmental conditional random fields: a summary of the JHU CLSP 2010 summer workshop |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-07T22%3A22%3A16IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-kuleuven_6IE&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=proceeding&rft.atitle=Speech%20recognition%20with%20segmental%20conditional%20random%20fields:%20a%20summary%20of%20the%20JHU%20CLSP%202010%20summer%20workshop&rft.btitle=Proceedings%2036th%20international%20conference%20on%20acoustics,%20speech%20and%20signal%20processing%20-%20ICASSP'2011&rft.au=Zweig,%20G&rft.date=2011&rft.spage=5044&rft.epage=5047&rft.pages=5044-5047&rft.issn=1520-6149&rft.isbn=9781457705397&rft.isbn_list=1457705397&rft_id=info:doi/&rft_dat=%3Ckuleuven_6IE%3E123456789_344395%3C/kuleuven_6IE%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true |