Percussive/harmonic sound separation by non-negative matrix factorization with smoothness/sparseness constraints

In this paper, unsupervised learning is used to separate percussive and harmonic sounds from monaural non-vocal polyphonic signals. Our algorithm is based on a modified non-negative matrix factorization (NMF) procedure that no labeled data is required to distinguish between percussive and harmonic b...

Descripción completa

Detalles Bibliográficos
Autores: Canadas Quesada, Francisco Jesus, Vera Candeas, Pedro, Ruiz Reyes, Nicolas, Carabias Orti, Julio J., Cabanas Molero, Pablo
Tipo de recurso: artículo
Estado:Versión publicada
Fecha de publicación:2014
País:España
Institución:Universitat Pompeu Fabra
Repositorio:Repositorio Digital de la UPF
OAI Identifier:oai:repositori.upf.edu:10230/23259
Acceso en línea:http://hdl.handle.net/10230/23259
http://dx.doi.org/10.1186/s13636-014-0026-5
Access Level:acceso abierto
Palabra clave:So -- Mesurament
Anàlisi harmònica (Música)
id ES_8e7beeb8a8d13e958da263736587ff96
oai_identifier_str oai:repositori.upf.edu:10230/23259
network_acronym_str ES
network_name_str España
repository_id_str
spelling Percussive/harmonic sound separation by non-negative matrix factorization with smoothness/sparseness constraintsCanadas Quesada, Francisco JesusVera Candeas, PedroRuiz Reyes, NicolasCarabias Orti, Julio J.Cabanas Molero, PabloSo -- MesuramentAnàlisi harmònica (Música)In this paper, unsupervised learning is used to separate percussive and harmonic sounds from monaural non-vocal polyphonic signals. Our algorithm is based on a modified non-negative matrix factorization (NMF) procedure that no labeled data is required to distinguish between percussive and harmonic bases because information from percussive and harmonic sounds is integrated into the decomposition process. NMF is performed in this process by assuming that harmonic sounds exhibit spectral sparseness (narrowband sounds) and temporal smoothness (steady sounds), whereas percussive sounds exhibit spectral smoothness (broadband sounds) and temporal sparseness (transient sounds). The evaluation is performed using several real-world excerpts from different musical genres. Comparing the developed approach to three current state-of-the art separation systems produces promising results.This work was supported by the Andalusian Business, Science and Innovation Council under project P2010- TIC-6762 and (FEDER) the Spanish Ministry of Economy and Competitiveness under Project TEC2012-38142-C04-03SpringerOpen201520152014info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersionapplication/pdfapplication/pdfhttp://hdl.handle.net/10230/23259http://dx.doi.org/10.1186/s13636-014-0026-5reponame:Repositorio Digital de la UPFinstname:Universitat Pompeu FabraInglésEURASIP Journal on Audio, Speech, and Music Processing. 2014; 2014: 26info:eu-repo/grantAgreement/ES/3PN/TEC2012-38142© 2014 Canadas-Quesada et al.; licensee Springer This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited.http://creativecommons.org/licenses/by/2.0info:eu-repo/semantics/openAccessoai:repositori.upf.edu:10230/232592026-06-12T07:21:37Z
dc.title.none.fl_str_mv Percussive/harmonic sound separation by non-negative matrix factorization with smoothness/sparseness constraints
title Percussive/harmonic sound separation by non-negative matrix factorization with smoothness/sparseness constraints
spellingShingle Percussive/harmonic sound separation by non-negative matrix factorization with smoothness/sparseness constraints
Canadas Quesada, Francisco Jesus
So -- Mesurament
Anàlisi harmònica (Música)
title_short Percussive/harmonic sound separation by non-negative matrix factorization with smoothness/sparseness constraints
title_full Percussive/harmonic sound separation by non-negative matrix factorization with smoothness/sparseness constraints
title_fullStr Percussive/harmonic sound separation by non-negative matrix factorization with smoothness/sparseness constraints
title_full_unstemmed Percussive/harmonic sound separation by non-negative matrix factorization with smoothness/sparseness constraints
title_sort Percussive/harmonic sound separation by non-negative matrix factorization with smoothness/sparseness constraints
dc.creator.none.fl_str_mv Canadas Quesada, Francisco Jesus
Vera Candeas, Pedro
Ruiz Reyes, Nicolas
Carabias Orti, Julio J.
Cabanas Molero, Pablo
author Canadas Quesada, Francisco Jesus
author_facet Canadas Quesada, Francisco Jesus
Vera Candeas, Pedro
Ruiz Reyes, Nicolas
Carabias Orti, Julio J.
Cabanas Molero, Pablo
author_role author
author2 Vera Candeas, Pedro
Ruiz Reyes, Nicolas
Carabias Orti, Julio J.
Cabanas Molero, Pablo
author2_role author
author
author
author
dc.subject.none.fl_str_mv So -- Mesurament
Anàlisi harmònica (Música)
topic So -- Mesurament
Anàlisi harmònica (Música)
description In this paper, unsupervised learning is used to separate percussive and harmonic sounds from monaural non-vocal polyphonic signals. Our algorithm is based on a modified non-negative matrix factorization (NMF) procedure that no labeled data is required to distinguish between percussive and harmonic bases because information from percussive and harmonic sounds is integrated into the decomposition process. NMF is performed in this process by assuming that harmonic sounds exhibit spectral sparseness (narrowband sounds) and temporal smoothness (steady sounds), whereas percussive sounds exhibit spectral smoothness (broadband sounds) and temporal sparseness (transient sounds). The evaluation is performed using several real-world excerpts from different musical genres. Comparing the developed approach to three current state-of-the art separation systems produces promising results.
publishDate 2014
dc.date.none.fl_str_mv 2014
2015
2015
dc.type.none.fl_str_mv info:eu-repo/semantics/article
info:eu-repo/semantics/publishedVersion
format article
status_str publishedVersion
dc.identifier.none.fl_str_mv http://hdl.handle.net/10230/23259
http://dx.doi.org/10.1186/s13636-014-0026-5
url http://hdl.handle.net/10230/23259
http://dx.doi.org/10.1186/s13636-014-0026-5
dc.language.none.fl_str_mv Inglés
language_invalid_str_mv Inglés
dc.relation.none.fl_str_mv EURASIP Journal on Audio, Speech, and Music Processing. 2014; 2014: 26
info:eu-repo/grantAgreement/ES/3PN/TEC2012-38142
dc.rights.none.fl_str_mv http://creativecommons.org/licenses/by/2.0
info:eu-repo/semantics/openAccess
rights_invalid_str_mv http://creativecommons.org/licenses/by/2.0
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
application/pdf
dc.publisher.none.fl_str_mv SpringerOpen
publisher.none.fl_str_mv SpringerOpen
dc.source.none.fl_str_mv reponame:Repositorio Digital de la UPF
instname:Universitat Pompeu Fabra
instname_str Universitat Pompeu Fabra
reponame_str Repositorio Digital de la UPF
collection Repositorio Digital de la UPF
repository.name.fl_str_mv
repository.mail.fl_str_mv
_version_ 1869413132997230592
score 15.811543

Similares en LA Referencia