Acoustic model adaptation using splines
Described is a technology by which a speech recognizer is adapted to perform in noisy environments using linear spline interpolation to approximate the nonlinear relationship between clean speech, noise, and noisy speech. Linear spline parameters that minimize the error the between predicted noisy f...
Gespeichert in:
| Hauptverfasser: | , , |
|---|---|
| Format: | Patent |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext bestellen |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | |
|---|---|
| container_issue | |
| container_start_page | |
| container_title | |
| container_volume | |
| creator | SELTZER MICHAEL LEWIS ACERO ALEJANDRO KALGAONKAR KAUSTUBH PRAKASH |
| description | Described is a technology by which a speech recognizer is adapted to perform in noisy environments using linear spline interpolation to approximate the nonlinear relationship between clean speech, noise, and noisy speech. Linear spline parameters that minimize the error the between predicted noisy features and actual noisy features are learned from training data, along with variance data that reflect regression errors. Also described is compensating for linear channel distortion and updating noise and channel parameters during speech recognition decoding. |
| format | Patent |
| fullrecord | <record><control><sourceid>epo_EVB</sourceid><recordid>TN_cdi_epo_espacenet_US8700394B2</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>US8700394B2</sourcerecordid><originalsourceid>FETCH-epo_espacenet_US8700394B23</originalsourceid><addsrcrecordid>eNrjZFB3TM4vLS7JTFbIzU9JzVFITEksKEksyczPUygtzsxLVyguyMnMSy3mYWBNS8wpTuWF0twMCm6uIc4euqkF-fGpxQWJyal5qSXxocEW5gYGxpYmTkbGRCgBAG6FJ_M</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>patent</recordtype></control><display><type>patent</type><title>Acoustic model adaptation using splines</title><source>esp@cenet</source><creator>SELTZER MICHAEL LEWIS ; ACERO ALEJANDRO ; KALGAONKAR KAUSTUBH PRAKASH</creator><creatorcontrib>SELTZER MICHAEL LEWIS ; ACERO ALEJANDRO ; KALGAONKAR KAUSTUBH PRAKASH</creatorcontrib><description>Described is a technology by which a speech recognizer is adapted to perform in noisy environments using linear spline interpolation to approximate the nonlinear relationship between clean speech, noise, and noisy speech. Linear spline parameters that minimize the error the between predicted noisy features and actual noisy features are learned from training data, along with variance data that reflect regression errors. Also described is compensating for linear channel distortion and updating noise and channel parameters during speech recognition decoding.</description><language>eng</language><subject>ACOUSTICS ; MUSICAL INSTRUMENTS ; PHYSICS ; SPEECH ANALYSIS OR SYNTHESIS ; SPEECH OR AUDIO CODING OR DECODING ; SPEECH OR VOICE PROCESSING ; SPEECH RECOGNITION</subject><creationdate>2014</creationdate><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://worldwide.espacenet.com/publicationDetails/biblio?FT=D&date=20140415&DB=EPODOC&CC=US&NR=8700394B2$$EHTML$$P50$$Gepo$$Hfree_for_read</linktohtml><link.rule.ids>230,308,776,881,25543,76294</link.rule.ids><linktorsrc>$$Uhttps://worldwide.espacenet.com/publicationDetails/biblio?FT=D&date=20140415&DB=EPODOC&CC=US&NR=8700394B2$$EView_record_in_European_Patent_Office$$FView_record_in_$$GEuropean_Patent_Office$$Hfree_for_read</linktorsrc></links><search><creatorcontrib>SELTZER MICHAEL LEWIS</creatorcontrib><creatorcontrib>ACERO ALEJANDRO</creatorcontrib><creatorcontrib>KALGAONKAR KAUSTUBH PRAKASH</creatorcontrib><title>Acoustic model adaptation using splines</title><description>Described is a technology by which a speech recognizer is adapted to perform in noisy environments using linear spline interpolation to approximate the nonlinear relationship between clean speech, noise, and noisy speech. Linear spline parameters that minimize the error the between predicted noisy features and actual noisy features are learned from training data, along with variance data that reflect regression errors. Also described is compensating for linear channel distortion and updating noise and channel parameters during speech recognition decoding.</description><subject>ACOUSTICS</subject><subject>MUSICAL INSTRUMENTS</subject><subject>PHYSICS</subject><subject>SPEECH ANALYSIS OR SYNTHESIS</subject><subject>SPEECH OR AUDIO CODING OR DECODING</subject><subject>SPEECH OR VOICE PROCESSING</subject><subject>SPEECH RECOGNITION</subject><fulltext>true</fulltext><rsrctype>patent</rsrctype><creationdate>2014</creationdate><recordtype>patent</recordtype><sourceid>EVB</sourceid><recordid>eNrjZFB3TM4vLS7JTFbIzU9JzVFITEksKEksyczPUygtzsxLVyguyMnMSy3mYWBNS8wpTuWF0twMCm6uIc4euqkF-fGpxQWJyal5qSXxocEW5gYGxpYmTkbGRCgBAG6FJ_M</recordid><startdate>20140415</startdate><enddate>20140415</enddate><creator>SELTZER MICHAEL LEWIS</creator><creator>ACERO ALEJANDRO</creator><creator>KALGAONKAR KAUSTUBH PRAKASH</creator><scope>EVB</scope></search><sort><creationdate>20140415</creationdate><title>Acoustic model adaptation using splines</title><author>SELTZER MICHAEL LEWIS ; ACERO ALEJANDRO ; KALGAONKAR KAUSTUBH PRAKASH</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-epo_espacenet_US8700394B23</frbrgroupid><rsrctype>patents</rsrctype><prefilter>patents</prefilter><language>eng</language><creationdate>2014</creationdate><topic>ACOUSTICS</topic><topic>MUSICAL INSTRUMENTS</topic><topic>PHYSICS</topic><topic>SPEECH ANALYSIS OR SYNTHESIS</topic><topic>SPEECH OR AUDIO CODING OR DECODING</topic><topic>SPEECH OR VOICE PROCESSING</topic><topic>SPEECH RECOGNITION</topic><toplevel>online_resources</toplevel><creatorcontrib>SELTZER MICHAEL LEWIS</creatorcontrib><creatorcontrib>ACERO ALEJANDRO</creatorcontrib><creatorcontrib>KALGAONKAR KAUSTUBH PRAKASH</creatorcontrib><collection>esp@cenet</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>SELTZER MICHAEL LEWIS</au><au>ACERO ALEJANDRO</au><au>KALGAONKAR KAUSTUBH PRAKASH</au><format>patent</format><genre>patent</genre><ristype>GEN</ristype><title>Acoustic model adaptation using splines</title><date>2014-04-15</date><risdate>2014</risdate><abstract>Described is a technology by which a speech recognizer is adapted to perform in noisy environments using linear spline interpolation to approximate the nonlinear relationship between clean speech, noise, and noisy speech. Linear spline parameters that minimize the error the between predicted noisy features and actual noisy features are learned from training data, along with variance data that reflect regression errors. Also described is compensating for linear channel distortion and updating noise and channel parameters during speech recognition decoding.</abstract><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext_linktorsrc |
| identifier | |
| ispartof | |
| issn | |
| language | eng |
| recordid | cdi_epo_espacenet_US8700394B2 |
| source | esp@cenet |
| subjects | ACOUSTICS MUSICAL INSTRUMENTS PHYSICS SPEECH ANALYSIS OR SYNTHESIS SPEECH OR AUDIO CODING OR DECODING SPEECH OR VOICE PROCESSING SPEECH RECOGNITION |
| title | Acoustic model adaptation using splines |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-22T07%3A57%3A02IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-epo_EVB&rft_val_fmt=info:ofi/fmt:kev:mtx:patent&rft.genre=patent&rft.au=SELTZER%20MICHAEL%20LEWIS&rft.date=2014-04-15&rft_id=info:doi/&rft_dat=%3Cepo_EVB%3EUS8700394B2%3C/epo_EVB%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 |