Adapta??o ao locutor usando a t?cnica MLLR
Autor(a) principal: | |
---|---|
Data de Publicação: | 2011 |
Tipo de documento: | Dissertação |
Idioma: | por |
Título da fonte: | Biblioteca Digital de Teses e Dissertações da INATEL |
Texto Completo: | http://tede.inatel.br:8080/tede/handle/tede/113 |
Resumo: | In this work a study of the technique of speaker adaptation called MLLR, Maximum Likelihood Linear Regression was made. The tests have been done using continuous speech applications and only the means of continuous hidden Markov models (HMM) have been adapted. The basic point of the technique is the partition of these means in regression classes for the generation of the transformation matrix. Moreover, the amount of material for adaptation of a speaker independet system is very important. Being thus, some alternatives for regression classes construction have been explored. Methods based on phonetic classification and based on distance metrics have been tested, varying also the number of regression classes. After tests, with a varied number of adaptation sentences, was verified that the better approach is to use only three regression classes with four adaptation sentences, but research still must be made in the area. |
id |
INAT_51102648711cce10271d49cb8bde7a63 |
---|---|
oai_identifier_str |
oai:localhost:tede/113 |
network_acronym_str |
INAT |
network_name_str |
Biblioteca Digital de Teses e Dissertações da INATEL |
repository_id_str |
|
spelling |
Ynoguti, Carlos Alberto156.167.778-70http://lattes.cnpq.br/5678667205895840Ynoguti, Carlos Alberto156.167.778-70http://lattes.cnpq.br/5678667205895840Violara, F?biohttp://lattes.cnpq.br/1810833808219352Ram?rez, Miguel Arjonahttp://lattes.cnpq.br/0057571113012412027.338.506-28http://lattes.cnpq.br/1273180157174302Fernandes, Daniela Barude2017-02-21T13:16:16Z2011-11-16Fernandes, Daniela Barude. Adapta??o ao locutor usando a t?cnica MLLR. 2011. [71]. disserta??o( Mestrado em Engenharia de Telecomunica??es) - Instituto Nacional de Telecomunica??es, [Santa Rita do Sapuca?] .http://tede.inatel.br:8080/tede/handle/tede/113In this work a study of the technique of speaker adaptation called MLLR, Maximum Likelihood Linear Regression was made. The tests have been done using continuous speech applications and only the means of continuous hidden Markov models (HMM) have been adapted. The basic point of the technique is the partition of these means in regression classes for the generation of the transformation matrix. Moreover, the amount of material for adaptation of a speaker independet system is very important. Being thus, some alternatives for regression classes construction have been explored. Methods based on phonetic classification and based on distance metrics have been tested, varying also the number of regression classes. After tests, with a varied number of adaptation sentences, was verified that the better approach is to use only three regression classes with four adaptation sentences, but research still must be made in the area.Neste trabalho realizou-se um estudo da t?cnica de adapta??o ao locutor chamada MLLR, Regress?o Linear de M?xima Verossimilhan?a. Os testes foram realizados utilizando fala cont?nua e somente as m?dias das componentes gaussianas dos Modelos Ocultos de Markov (HMMs) foram adaptadas. O ponto fundamental da t?cnica ? a parti??o dessas m?dias em classes de regress?o para a gera??o da matriz de transforma??o. Al?m disso, a quantidade de material para adapta??o de um sistema independente de locutor ? muito importante. Sendo assim, diversas alternativas para a forma??o das classes de regress?o foram exploradas. Foram testados m?todos baseados em classifica??o fon?tica e em medidas de dist?ncia, variando-se tamb?m o n?mero de classes de regress?o. Ap?s a realiza??o dos testes, com um n?mero variado de locu??es de adapta??o, verificou-se que o melhor resultado foi obtido utilizando-se quatro locu??es de adapta??o e tr?s classes de regress?o, mas pesquisas ainda devem ser feitas na ?rea.Submitted by Tede Dspace (tede@inatel.br) on 2017-02-21T13:16:15Z No. of bitstreams: 2 Disserta??o V.Final Daniela Barude.pdf: 757133 bytes, checksum: 7ca09cc73dbc7d5cb640008358fd6f0c (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5)Made available in DSpace on 2017-02-21T13:16:16Z (GMT). No. of bitstreams: 2 Disserta??o V.Final Daniela Barude.pdf: 757133 bytes, checksum: 7ca09cc73dbc7d5cb640008358fd6f0c (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Previous issue date: 2011-11-16application/pdfhttp://tede.inatel.br:8080/jspui/retrieve/944/Disserta%c3%a7%c3%a3o%20V.Final%20Daniela%20Barude.pdf.jpgporInstituto Nacional de Telecomunica??esMestrado em Engenharia de Telecomunica??esINATELBrasilInstituto Nacional de Telecomunica??eshttp://creativecommons.org/licenses/by-nd/4.0/info:eu-repo/semantics/openAccessSistemas de reconhecimento de fala; adapta??o ao locutor; t?cnica MLLR, classes de regress?oEngenharia - Telecomunica??esAdapta??o ao locutor usando a t?cnica MLLRinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/masterThesisreponame:Biblioteca Digital de Teses e Dissertações da INATELinstname:Instituto Nacional de Telecomunicações (INATEL)instacron:INATELTEXTDisserta??o V.Final Daniela Barude.pdf.txtDisserta??o V.Final Daniela Barude.pdf.txttext/plain83459http://localhost:8080/tede/bitstream/tede/113/6/Disserta%C3%A7%C3%A3o+V.Final+Daniela+Barude.pdf.txt91871c07aa57b57c40239885aca2799aMD56THUMBNAILDisserta??o V.Final Daniela Barude.pdf.jpgDisserta??o V.Final Daniela Barude.pdf.jpgimage/jpeg3358http://localhost:8080/tede/bitstream/tede/113/7/Disserta%C3%A7%C3%A3o+V.Final+Daniela+Barude.pdf.jpgf7237859c013fe52ab59370367994a90MD57ORIGINALDisserta??o V.Final Daniela Barude.pdfDisserta??o V.Final Daniela Barude.pdfapplication/pdf757133http://localhost:8080/tede/bitstream/tede/113/5/Disserta%C3%A7%C3%A3o+V.Final+Daniela+Barude.pdf7ca09cc73dbc7d5cb640008358fd6f0cMD55LICENSElicense.txtlicense.txttext/plain; charset=utf-8112http://localhost:8080/tede/bitstream/tede/113/1/license.txtc6279291b293f0db82678eaa73a27769MD51CC-LICENSElicense_urllicense_urltext/plain; charset=utf-846http://localhost:8080/tede/bitstream/tede/113/2/license_url587cd8ffae15c8598ed3c46d248a3f38MD52license_textlicense_texttext/html; charset=utf-80http://localhost:8080/tede/bitstream/tede/113/3/license_textd41d8cd98f00b204e9800998ecf8427eMD53license_rdflicense_rdfapplication/rdf+xml; charset=utf-80http://localhost:8080/tede/bitstream/tede/113/4/license_rdfd41d8cd98f00b204e9800998ecf8427eMD54tede/1132018-04-17 09:59:03.837oai:localhost:tede/113QXV0b3Jpem8gYSBwdWJsaWNhPz9vIGRhIG1pbmhhIERpc3NlcnRhPz9vIGRlIE1lc3RyYWRvLCBlbSBmb3JtYXRvIFBERiwgY29tIGJsb3F1ZWlvIGRlIGVkaT8/bywgY29sYWdlbSBlIGM/cGlhLg==Biblioteca Digital de Teses e Dissertaçõeshttp://tede.inatel.br:8080/jspui/PUBhttp://tede.inatel.br:8080/oai/requestbiblioteca@inatel.br || biblioteca.atendimento@inatel.bropendoar:2018-04-17T12:59:03Biblioteca Digital de Teses e Dissertações da INATEL - Instituto Nacional de Telecomunicações (INATEL)false |
dc.title.por.fl_str_mv |
Adapta??o ao locutor usando a t?cnica MLLR |
title |
Adapta??o ao locutor usando a t?cnica MLLR |
spellingShingle |
Adapta??o ao locutor usando a t?cnica MLLR Fernandes, Daniela Barude Sistemas de reconhecimento de fala; adapta??o ao locutor; t?cnica MLLR, classes de regress?o Engenharia - Telecomunica??es |
title_short |
Adapta??o ao locutor usando a t?cnica MLLR |
title_full |
Adapta??o ao locutor usando a t?cnica MLLR |
title_fullStr |
Adapta??o ao locutor usando a t?cnica MLLR |
title_full_unstemmed |
Adapta??o ao locutor usando a t?cnica MLLR |
title_sort |
Adapta??o ao locutor usando a t?cnica MLLR |
author |
Fernandes, Daniela Barude |
author_facet |
Fernandes, Daniela Barude |
author_role |
author |
dc.contributor.advisor1.fl_str_mv |
Ynoguti, Carlos Alberto |
dc.contributor.advisor1ID.fl_str_mv |
156.167.778-70 |
dc.contributor.advisor1Lattes.fl_str_mv |
http://lattes.cnpq.br/5678667205895840 |
dc.contributor.referee1.fl_str_mv |
Ynoguti, Carlos Alberto |
dc.contributor.referee1ID.fl_str_mv |
156.167.778-70 |
dc.contributor.referee1Lattes.fl_str_mv |
http://lattes.cnpq.br/5678667205895840 |
dc.contributor.referee2.fl_str_mv |
Violara, F?bio |
dc.contributor.referee2Lattes.fl_str_mv |
http://lattes.cnpq.br/1810833808219352 |
dc.contributor.referee3.fl_str_mv |
Ram?rez, Miguel Arjona |
dc.contributor.referee3Lattes.fl_str_mv |
http://lattes.cnpq.br/0057571113012412 |
dc.contributor.authorID.fl_str_mv |
027.338.506-28 |
dc.contributor.authorLattes.fl_str_mv |
http://lattes.cnpq.br/1273180157174302 |
dc.contributor.author.fl_str_mv |
Fernandes, Daniela Barude |
contributor_str_mv |
Ynoguti, Carlos Alberto Ynoguti, Carlos Alberto Violara, F?bio Ram?rez, Miguel Arjona |
dc.subject.por.fl_str_mv |
Sistemas de reconhecimento de fala; adapta??o ao locutor; t?cnica MLLR, classes de regress?o |
topic |
Sistemas de reconhecimento de fala; adapta??o ao locutor; t?cnica MLLR, classes de regress?o Engenharia - Telecomunica??es |
dc.subject.cnpq.fl_str_mv |
Engenharia - Telecomunica??es |
description |
In this work a study of the technique of speaker adaptation called MLLR, Maximum Likelihood Linear Regression was made. The tests have been done using continuous speech applications and only the means of continuous hidden Markov models (HMM) have been adapted. The basic point of the technique is the partition of these means in regression classes for the generation of the transformation matrix. Moreover, the amount of material for adaptation of a speaker independet system is very important. Being thus, some alternatives for regression classes construction have been explored. Methods based on phonetic classification and based on distance metrics have been tested, varying also the number of regression classes. After tests, with a varied number of adaptation sentences, was verified that the better approach is to use only three regression classes with four adaptation sentences, but research still must be made in the area. |
publishDate |
2011 |
dc.date.issued.fl_str_mv |
2011-11-16 |
dc.date.accessioned.fl_str_mv |
2017-02-21T13:16:16Z |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/masterThesis |
format |
masterThesis |
status_str |
publishedVersion |
dc.identifier.citation.fl_str_mv |
Fernandes, Daniela Barude. Adapta??o ao locutor usando a t?cnica MLLR. 2011. [71]. disserta??o( Mestrado em Engenharia de Telecomunica??es) - Instituto Nacional de Telecomunica??es, [Santa Rita do Sapuca?] . |
dc.identifier.uri.fl_str_mv |
http://tede.inatel.br:8080/tede/handle/tede/113 |
identifier_str_mv |
Fernandes, Daniela Barude. Adapta??o ao locutor usando a t?cnica MLLR. 2011. [71]. disserta??o( Mestrado em Engenharia de Telecomunica??es) - Instituto Nacional de Telecomunica??es, [Santa Rita do Sapuca?] . |
url |
http://tede.inatel.br:8080/tede/handle/tede/113 |
dc.language.iso.fl_str_mv |
por |
language |
por |
dc.rights.driver.fl_str_mv |
http://creativecommons.org/licenses/by-nd/4.0/ info:eu-repo/semantics/openAccess |
rights_invalid_str_mv |
http://creativecommons.org/licenses/by-nd/4.0/ |
eu_rights_str_mv |
openAccess |
dc.format.none.fl_str_mv |
application/pdf |
dc.publisher.none.fl_str_mv |
Instituto Nacional de Telecomunica??es |
dc.publisher.program.fl_str_mv |
Mestrado em Engenharia de Telecomunica??es |
dc.publisher.initials.fl_str_mv |
INATEL |
dc.publisher.country.fl_str_mv |
Brasil |
dc.publisher.department.fl_str_mv |
Instituto Nacional de Telecomunica??es |
publisher.none.fl_str_mv |
Instituto Nacional de Telecomunica??es |
dc.source.none.fl_str_mv |
reponame:Biblioteca Digital de Teses e Dissertações da INATEL instname:Instituto Nacional de Telecomunicações (INATEL) instacron:INATEL |
instname_str |
Instituto Nacional de Telecomunicações (INATEL) |
instacron_str |
INATEL |
institution |
INATEL |
reponame_str |
Biblioteca Digital de Teses e Dissertações da INATEL |
collection |
Biblioteca Digital de Teses e Dissertações da INATEL |
bitstream.url.fl_str_mv |
http://localhost:8080/tede/bitstream/tede/113/6/Disserta%C3%A7%C3%A3o+V.Final+Daniela+Barude.pdf.txt http://localhost:8080/tede/bitstream/tede/113/7/Disserta%C3%A7%C3%A3o+V.Final+Daniela+Barude.pdf.jpg http://localhost:8080/tede/bitstream/tede/113/5/Disserta%C3%A7%C3%A3o+V.Final+Daniela+Barude.pdf http://localhost:8080/tede/bitstream/tede/113/1/license.txt http://localhost:8080/tede/bitstream/tede/113/2/license_url http://localhost:8080/tede/bitstream/tede/113/3/license_text http://localhost:8080/tede/bitstream/tede/113/4/license_rdf |
bitstream.checksum.fl_str_mv |
91871c07aa57b57c40239885aca2799a f7237859c013fe52ab59370367994a90 7ca09cc73dbc7d5cb640008358fd6f0c c6279291b293f0db82678eaa73a27769 587cd8ffae15c8598ed3c46d248a3f38 d41d8cd98f00b204e9800998ecf8427e d41d8cd98f00b204e9800998ecf8427e |
bitstream.checksumAlgorithm.fl_str_mv |
MD5 MD5 MD5 MD5 MD5 MD5 MD5 |
repository.name.fl_str_mv |
Biblioteca Digital de Teses e Dissertações da INATEL - Instituto Nacional de Telecomunicações (INATEL) |
repository.mail.fl_str_mv |
biblioteca@inatel.br || biblioteca.atendimento@inatel.br |
_version_ |
1800214191054782464 |