On the determination of epsilon during discriminative GMM training
Autor(a) principal: | |
---|---|
Data de Publicação: | 2010 |
Outros Autores: | , , , , , , , , |
Tipo de documento: | Artigo de conferência |
Idioma: | eng |
Título da fonte: | Repositório Institucional da UNESP |
Texto Completo: | http://dx.doi.org/10.1109/ISM.2010.66 http://hdl.handle.net/11449/72054 |
Resumo: | Discriminative training of Gaussian Mixture Models (GMMs) for speech or speaker recognition purposes is usually based on the gradient descent method, in which the iteration step-size, ε, uses to be defined experimentally. In this letter, we derive an equation to adaptively determine ε, by showing that the second-order Newton-Raphson iterative method to find roots of equations is equivalent to the gradient descent algorithm. © 2010 IEEE. |
id |
UNSP_b4293691ab633e41fff1c89b6c28e33d |
---|---|
oai_identifier_str |
oai:repositorio.unesp.br:11449/72054 |
network_acronym_str |
UNSP |
network_name_str |
Repositório Institucional da UNESP |
repository_id_str |
2946 |
spelling |
On the determination of epsilon during discriminative GMM trainingDiscriminative training of Gaussian Mixture Models (GMMs)Markov ModelsSpeaker identificationSpeech recognitionDiscriminative trainingGaussian mixture modelsGradient descent algorithmsGradient Descent methodIteration stepNewton-Raphson iterative methodSecond ordersSpeaker recognitionGaussian distributionIterative methodsLoudspeakersMarkov processesDiscriminative training of Gaussian Mixture Models (GMMs) for speech or speaker recognition purposes is usually based on the gradient descent method, in which the iteration step-size, ε, uses to be defined experimentally. In this letter, we derive an equation to adaptively determine ε, by showing that the second-order Newton-Raphson iterative method to find roots of equations is equivalent to the gradient descent algorithm. © 2010 IEEE.SpeechLab. FFI, Institute of Physics at São Carlos University of São Paulo, Av. Trabalhador São Carlense 400, 13566-590, São Carlos, SPDCCE/IBILCE/UNESP São Paulo State University, Rua Cristovão Colombo 2265, São José do Rio Preto, SPDepartment of Computer Science and Information Engineering Shu-Te University, N.59, Hengshan Rd., Yanchao, Kaohsiung County 82445DCCE/IBILCE/UNESP São Paulo State University, Rua Cristovão Colombo 2265, São José do Rio Preto, SPUniversidade de São Paulo (USP)Universidade Estadual Paulista (Unesp)Shu-Te UniversityGuido, Rodrigo Capobianco [UNESP]Chen, Shi-Huang [UNESP]Junior, Sylvio Barbon [UNESP]Souza, Leonardo Mendes [UNESP]Vieira, Lucimar Sasso [UNESP]Rodrigues, Luciene Cavalcanti [UNESP]Escola, Joao Paulo Lemos [UNESP]Zulato, Paulo Ricardo Franchi [UNESP]Lacerda, Michel Alves [UNESP]Ribeiro, Jussara [UNESP]2014-05-27T11:25:20Z2014-05-27T11:25:20Z2010-12-01info:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/conferenceObject362-364http://dx.doi.org/10.1109/ISM.2010.66Proceedings - 2010 IEEE International Symposium on Multimedia, ISM 2010, p. 362-364.http://hdl.handle.net/11449/7205410.1109/ISM.2010.662-s2.0-7995172800465420862268080670000-0002-0924-8024Scopusreponame:Repositório Institucional da UNESPinstname:Universidade Estadual Paulista (UNESP)instacron:UNESPengProceedings - 2010 IEEE International Symposium on Multimedia, ISM 2010info:eu-repo/semantics/openAccess2021-10-23T21:37:54Zoai:repositorio.unesp.br:11449/72054Repositório InstitucionalPUBhttp://repositorio.unesp.br/oai/requestopendoar:29462024-08-05T20:07:33.912397Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP)false |
dc.title.none.fl_str_mv |
On the determination of epsilon during discriminative GMM training |
title |
On the determination of epsilon during discriminative GMM training |
spellingShingle |
On the determination of epsilon during discriminative GMM training Guido, Rodrigo Capobianco [UNESP] Discriminative training of Gaussian Mixture Models (GMMs) Markov Models Speaker identification Speech recognition Discriminative training Gaussian mixture models Gradient descent algorithms Gradient Descent method Iteration step Newton-Raphson iterative method Second orders Speaker recognition Gaussian distribution Iterative methods Loudspeakers Markov processes |
title_short |
On the determination of epsilon during discriminative GMM training |
title_full |
On the determination of epsilon during discriminative GMM training |
title_fullStr |
On the determination of epsilon during discriminative GMM training |
title_full_unstemmed |
On the determination of epsilon during discriminative GMM training |
title_sort |
On the determination of epsilon during discriminative GMM training |
author |
Guido, Rodrigo Capobianco [UNESP] |
author_facet |
Guido, Rodrigo Capobianco [UNESP] Chen, Shi-Huang [UNESP] Junior, Sylvio Barbon [UNESP] Souza, Leonardo Mendes [UNESP] Vieira, Lucimar Sasso [UNESP] Rodrigues, Luciene Cavalcanti [UNESP] Escola, Joao Paulo Lemos [UNESP] Zulato, Paulo Ricardo Franchi [UNESP] Lacerda, Michel Alves [UNESP] Ribeiro, Jussara [UNESP] |
author_role |
author |
author2 |
Chen, Shi-Huang [UNESP] Junior, Sylvio Barbon [UNESP] Souza, Leonardo Mendes [UNESP] Vieira, Lucimar Sasso [UNESP] Rodrigues, Luciene Cavalcanti [UNESP] Escola, Joao Paulo Lemos [UNESP] Zulato, Paulo Ricardo Franchi [UNESP] Lacerda, Michel Alves [UNESP] Ribeiro, Jussara [UNESP] |
author2_role |
author author author author author author author author author |
dc.contributor.none.fl_str_mv |
Universidade de São Paulo (USP) Universidade Estadual Paulista (Unesp) Shu-Te University |
dc.contributor.author.fl_str_mv |
Guido, Rodrigo Capobianco [UNESP] Chen, Shi-Huang [UNESP] Junior, Sylvio Barbon [UNESP] Souza, Leonardo Mendes [UNESP] Vieira, Lucimar Sasso [UNESP] Rodrigues, Luciene Cavalcanti [UNESP] Escola, Joao Paulo Lemos [UNESP] Zulato, Paulo Ricardo Franchi [UNESP] Lacerda, Michel Alves [UNESP] Ribeiro, Jussara [UNESP] |
dc.subject.por.fl_str_mv |
Discriminative training of Gaussian Mixture Models (GMMs) Markov Models Speaker identification Speech recognition Discriminative training Gaussian mixture models Gradient descent algorithms Gradient Descent method Iteration step Newton-Raphson iterative method Second orders Speaker recognition Gaussian distribution Iterative methods Loudspeakers Markov processes |
topic |
Discriminative training of Gaussian Mixture Models (GMMs) Markov Models Speaker identification Speech recognition Discriminative training Gaussian mixture models Gradient descent algorithms Gradient Descent method Iteration step Newton-Raphson iterative method Second orders Speaker recognition Gaussian distribution Iterative methods Loudspeakers Markov processes |
description |
Discriminative training of Gaussian Mixture Models (GMMs) for speech or speaker recognition purposes is usually based on the gradient descent method, in which the iteration step-size, ε, uses to be defined experimentally. In this letter, we derive an equation to adaptively determine ε, by showing that the second-order Newton-Raphson iterative method to find roots of equations is equivalent to the gradient descent algorithm. © 2010 IEEE. |
publishDate |
2010 |
dc.date.none.fl_str_mv |
2010-12-01 2014-05-27T11:25:20Z 2014-05-27T11:25:20Z |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/conferenceObject |
format |
conferenceObject |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
http://dx.doi.org/10.1109/ISM.2010.66 Proceedings - 2010 IEEE International Symposium on Multimedia, ISM 2010, p. 362-364. http://hdl.handle.net/11449/72054 10.1109/ISM.2010.66 2-s2.0-79951728004 6542086226808067 0000-0002-0924-8024 |
url |
http://dx.doi.org/10.1109/ISM.2010.66 http://hdl.handle.net/11449/72054 |
identifier_str_mv |
Proceedings - 2010 IEEE International Symposium on Multimedia, ISM 2010, p. 362-364. 10.1109/ISM.2010.66 2-s2.0-79951728004 6542086226808067 0000-0002-0924-8024 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.relation.none.fl_str_mv |
Proceedings - 2010 IEEE International Symposium on Multimedia, ISM 2010 |
dc.rights.driver.fl_str_mv |
info:eu-repo/semantics/openAccess |
eu_rights_str_mv |
openAccess |
dc.format.none.fl_str_mv |
362-364 |
dc.source.none.fl_str_mv |
Scopus reponame:Repositório Institucional da UNESP instname:Universidade Estadual Paulista (UNESP) instacron:UNESP |
instname_str |
Universidade Estadual Paulista (UNESP) |
instacron_str |
UNESP |
institution |
UNESP |
reponame_str |
Repositório Institucional da UNESP |
collection |
Repositório Institucional da UNESP |
repository.name.fl_str_mv |
Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP) |
repository.mail.fl_str_mv |
|
_version_ |
1808129162570891264 |