Bioinspired sparse spectro-temporal representation of speech for robust classification

Autores
Martínez, César Ernesto; Goddard, J.; Milone, Diego Humberto; Rufiner, Hugo Leonardo
Año de publicación
2012
Idioma
inglés
Tipo de recurso
artículo
Estado
versión publicada
Descripción
In this work, a first approach to a robust phoneme recognition task by means of a biologically inspired feature extraction method is presented. The proposed technique provides an approximation to the speech signal representation at the auditory cortical level. It is based on an optimal dictionary of atoms, estimated from auditory spectrograms, and the Matching Pursuit algorithm to approximate the cortical activations. This provides a sparse coding with intrinsic noise robustness, which can be therefore exploited when using the system in adverse environments. The recognition task consisted in the classification of a set of 5 easily confused English phonemes, in both clean and noisy conditions. Multilayer perceptrons were trained as classifiers and the performance was compared to other classic and robust parameterizations: the auditory spectrogram, a probabilistic optimum filtering on Mel frequency cepstral coefficients and the perceptual linear prediction coefficients. Results showed a significant improvement in the recognition rate of clean and noisy phonemes by the cortical representation over these other parameterizations.
Fil: Martínez, César Ernesto. Centro de I+d En Señales; Argentina. Universidad Nacional de Entre Ríos; Argentina
Fil: Goddard, J.. Universidad Autónoma Metropolitana - Iztapalapa; México
Fil: Milone, Diego Humberto. Centro de I+d En Señales; Argentina. Consejo Nacional de Investigaciones Científicas y Técnicas; Argentina
Fil: Rufiner, Hugo Leonardo. Universidad Nacional de Entre Ríos; Argentina. Consejo Nacional de Investigaciones Científicas y Técnicas; Argentina. Centro de I+d En Señales; Argentina
Materia
APPROXIMATED AUDITORY CORTICAL REPRESENTATION
ROBUST PHONEME RECOGNITION
SPARSE CODING
Nivel de accesibilidad
acceso abierto
Condiciones de uso
https://creativecommons.org/licenses/by-nc-sa/2.5/ar/
Repositorio
CONICET Digital (CONICET)
Institución
Consejo Nacional de Investigaciones Científicas y Técnicas
OAI Identificador
oai:ri.conicet.gov.ar:11336/96495
Acceder
id CONICETDig_5d31a75007b487fba0442057e897f2c6
oai_identifier_str oai:ri.conicet.gov.ar:11336/96495
network_acronym_str CONICETDig
repository_id_str 3498
network_name_str CONICET Digital (CONICET)
spelling Bioinspired sparse spectro-temporal representation of speech for robust classificationMartínez, César ErnestoGoddard, J.Milone, Diego HumbertoRufiner, Hugo LeonardoAPPROXIMATED AUDITORY CORTICAL REPRESENTATIONROBUST PHONEME RECOGNITIONSPARSE CODINGhttps://purl.org/becyt/ford/2.2https://purl.org/becyt/ford/2In this work, a first approach to a robust phoneme recognition task by means of a biologically inspired feature extraction method is presented. The proposed technique provides an approximation to the speech signal representation at the auditory cortical level. It is based on an optimal dictionary of atoms, estimated from auditory spectrograms, and the Matching Pursuit algorithm to approximate the cortical activations. This provides a sparse coding with intrinsic noise robustness, which can be therefore exploited when using the system in adverse environments. The recognition task consisted in the classification of a set of 5 easily confused English phonemes, in both clean and noisy conditions. Multilayer perceptrons were trained as classifiers and the performance was compared to other classic and robust parameterizations: the auditory spectrogram, a probabilistic optimum filtering on Mel frequency cepstral coefficients and the perceptual linear prediction coefficients. Results showed a significant improvement in the recognition rate of clean and noisy phonemes by the cortical representation over these other parameterizations.Fil: Martínez, César Ernesto. Centro de I+d En Señales; Argentina. Universidad Nacional de Entre Ríos; ArgentinaFil: Goddard, J.. Universidad Autónoma Metropolitana - Iztapalapa; MéxicoFil: Milone, Diego Humberto. Centro de I+d En Señales; Argentina. Consejo Nacional de Investigaciones Científicas y Técnicas; ArgentinaFil: Rufiner, Hugo Leonardo. Universidad Nacional de Entre Ríos; Argentina. Consejo Nacional de Investigaciones Científicas y Técnicas; Argentina. Centro de I+d En Señales; ArgentinaAcademic Press Ltd - Elsevier Science Ltd2012-10info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersionhttp://purl.org/coar/resource_type/c_6501info:ar-repo/semantics/articuloapplication/pdfapplication/pdfapplication/pdfhttp://hdl.handle.net/11336/96495Martínez, César Ernesto; Goddard, J.; Milone, Diego Humberto; Rufiner, Hugo Leonardo; Bioinspired sparse spectro-temporal representation of speech for robust classification; Academic Press Ltd - Elsevier Science Ltd; Computer Speech And Language; 26; 5; 10-2012; 336-3480885-2308CONICET DigitalCONICETenginfo:eu-repo/semantics/altIdentifier/url/http://www.sciencedirect.com/science/article/pii/S0885230812000125info:eu-repo/semantics/altIdentifier/doi/10.1016/j.csl.2012.02.002info:eu-repo/semantics/openAccesshttps://creativecommons.org/licenses/by-nc-sa/2.5/ar/reponame:CONICET Digital (CONICET)instname:Consejo Nacional de Investigaciones Científicas y Técnicas2025-09-29T10:33:10Zoai:ri.conicet.gov.ar:11336/96495instacron:CONICETInstitucionalhttp://ri.conicet.gov.ar/Organismo científico-tecnológicoNo correspondehttp://ri.conicet.gov.ar/oai/requestdasensio@conicet.gov.ar; lcarlino@conicet.gov.arArgentinaNo correspondeNo correspondeNo correspondeopendoar:34982025-09-29 10:33:10.296CONICET Digital (CONICET) - Consejo Nacional de Investigaciones Científicas y Técnicasfalse
dc.title.none.fl_str_mv Bioinspired sparse spectro-temporal representation of speech for robust classification
title Bioinspired sparse spectro-temporal representation of speech for robust classification
spellingShingle Bioinspired sparse spectro-temporal representation of speech for robust classification
Martínez, César Ernesto
APPROXIMATED AUDITORY CORTICAL REPRESENTATION
ROBUST PHONEME RECOGNITION
SPARSE CODING
title_short Bioinspired sparse spectro-temporal representation of speech for robust classification
title_full Bioinspired sparse spectro-temporal representation of speech for robust classification
title_fullStr Bioinspired sparse spectro-temporal representation of speech for robust classification
title_full_unstemmed Bioinspired sparse spectro-temporal representation of speech for robust classification
title_sort Bioinspired sparse spectro-temporal representation of speech for robust classification
dc.creator.none.fl_str_mv Martínez, César Ernesto
Goddard, J.
Milone, Diego Humberto
Rufiner, Hugo Leonardo
author Martínez, César Ernesto
author_facet Martínez, César Ernesto
Goddard, J.
Milone, Diego Humberto
Rufiner, Hugo Leonardo
author_role author
author2 Goddard, J.
Milone, Diego Humberto
Rufiner, Hugo Leonardo
author2_role author
author
author
dc.subject.none.fl_str_mv APPROXIMATED AUDITORY CORTICAL REPRESENTATION
ROBUST PHONEME RECOGNITION
SPARSE CODING
topic APPROXIMATED AUDITORY CORTICAL REPRESENTATION
ROBUST PHONEME RECOGNITION
SPARSE CODING
purl_subject.fl_str_mv https://purl.org/becyt/ford/2.2
https://purl.org/becyt/ford/2
dc.description.none.fl_txt_mv In this work, a first approach to a robust phoneme recognition task by means of a biologically inspired feature extraction method is presented. The proposed technique provides an approximation to the speech signal representation at the auditory cortical level. It is based on an optimal dictionary of atoms, estimated from auditory spectrograms, and the Matching Pursuit algorithm to approximate the cortical activations. This provides a sparse coding with intrinsic noise robustness, which can be therefore exploited when using the system in adverse environments. The recognition task consisted in the classification of a set of 5 easily confused English phonemes, in both clean and noisy conditions. Multilayer perceptrons were trained as classifiers and the performance was compared to other classic and robust parameterizations: the auditory spectrogram, a probabilistic optimum filtering on Mel frequency cepstral coefficients and the perceptual linear prediction coefficients. Results showed a significant improvement in the recognition rate of clean and noisy phonemes by the cortical representation over these other parameterizations.
Fil: Martínez, César Ernesto. Centro de I+d En Señales; Argentina. Universidad Nacional de Entre Ríos; Argentina
Fil: Goddard, J.. Universidad Autónoma Metropolitana - Iztapalapa; México
Fil: Milone, Diego Humberto. Centro de I+d En Señales; Argentina. Consejo Nacional de Investigaciones Científicas y Técnicas; Argentina
Fil: Rufiner, Hugo Leonardo. Universidad Nacional de Entre Ríos; Argentina. Consejo Nacional de Investigaciones Científicas y Técnicas; Argentina. Centro de I+d En Señales; Argentina
description In this work, a first approach to a robust phoneme recognition task by means of a biologically inspired feature extraction method is presented. The proposed technique provides an approximation to the speech signal representation at the auditory cortical level. It is based on an optimal dictionary of atoms, estimated from auditory spectrograms, and the Matching Pursuit algorithm to approximate the cortical activations. This provides a sparse coding with intrinsic noise robustness, which can be therefore exploited when using the system in adverse environments. The recognition task consisted in the classification of a set of 5 easily confused English phonemes, in both clean and noisy conditions. Multilayer perceptrons were trained as classifiers and the performance was compared to other classic and robust parameterizations: the auditory spectrogram, a probabilistic optimum filtering on Mel frequency cepstral coefficients and the perceptual linear prediction coefficients. Results showed a significant improvement in the recognition rate of clean and noisy phonemes by the cortical representation over these other parameterizations.
publishDate 2012
dc.date.none.fl_str_mv 2012-10
dc.type.none.fl_str_mv info:eu-repo/semantics/article
info:eu-repo/semantics/publishedVersion
http://purl.org/coar/resource_type/c_6501
info:ar-repo/semantics/articulo
format article
status_str publishedVersion
dc.identifier.none.fl_str_mv http://hdl.handle.net/11336/96495
Martínez, César Ernesto; Goddard, J.; Milone, Diego Humberto; Rufiner, Hugo Leonardo; Bioinspired sparse spectro-temporal representation of speech for robust classification; Academic Press Ltd - Elsevier Science Ltd; Computer Speech And Language; 26; 5; 10-2012; 336-348
0885-2308
CONICET Digital
CONICET
url http://hdl.handle.net/11336/96495
identifier_str_mv Martínez, César Ernesto; Goddard, J.; Milone, Diego Humberto; Rufiner, Hugo Leonardo; Bioinspired sparse spectro-temporal representation of speech for robust classification; Academic Press Ltd - Elsevier Science Ltd; Computer Speech And Language; 26; 5; 10-2012; 336-348
0885-2308
CONICET Digital
CONICET
dc.language.none.fl_str_mv eng
language eng
dc.relation.none.fl_str_mv info:eu-repo/semantics/altIdentifier/url/http://www.sciencedirect.com/science/article/pii/S0885230812000125
info:eu-repo/semantics/altIdentifier/doi/10.1016/j.csl.2012.02.002
dc.rights.none.fl_str_mv info:eu-repo/semantics/openAccess
https://creativecommons.org/licenses/by-nc-sa/2.5/ar/
eu_rights_str_mv openAccess
rights_invalid_str_mv https://creativecommons.org/licenses/by-nc-sa/2.5/ar/
dc.format.none.fl_str_mv application/pdf
application/pdf
application/pdf
dc.publisher.none.fl_str_mv Academic Press Ltd - Elsevier Science Ltd
publisher.none.fl_str_mv Academic Press Ltd - Elsevier Science Ltd
dc.source.none.fl_str_mv reponame:CONICET Digital (CONICET)
instname:Consejo Nacional de Investigaciones Científicas y Técnicas
reponame_str CONICET Digital (CONICET)
collection CONICET Digital (CONICET)
instname_str Consejo Nacional de Investigaciones Científicas y Técnicas
repository.name.fl_str_mv CONICET Digital (CONICET) - Consejo Nacional de Investigaciones Científicas y Técnicas
repository.mail.fl_str_mv dasensio@conicet.gov.ar; lcarlino@conicet.gov.ar
_version_ 1844614346747936768
score 13.069144

Publicaciones similares