Extended nonnegative tensor factorisation models for musical sound source separation.
Identifieur interne : 001795 ( Main/Exploration ); précédent : 001794; suivant : 001796Extended nonnegative tensor factorisation models for musical sound source separation.
Auteurs : Derry Fitzgerald [Irlande (pays)] ; Matt Cranitch ; Eugene CoyleSource :
- Computational intelligence and neuroscience [ 1687-5265 ] ; 2008.
Abstract
Recently, shift-invariant tensor factorisation algorithms have been proposed for the purposes of sound source separation of pitched musical instruments. However, in practice, existing algorithms require the use of log-frequency spectrograms to allow shift invariance in frequency which causes problems when attempting to resynthesise the separated sources. Further, it is difficult to impose harmonicity constraints on the recovered basis functions. This paper proposes a new additive synthesis-based approach which allows the use of linear-frequency spectrograms as well as imposing strict harmonic constraints, resulting in an improved model. Further, these additional constraints allow the addition of a source filter model to the factorisation framework, and an extended model which is capable of separating mixtures of pitched and percussive instruments simultaneously.
DOI: 10.1155/2008/872425
PubMed: 18551178
PubMed Central: PMC2409961
Affiliations:
Links toward previous steps (curation, corpus...)
Le document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en">Extended nonnegative tensor factorisation models for musical sound source separation.</title><author><name sortKey="Fitzgerald, Derry" sort="Fitzgerald, Derry" uniqKey="Fitzgerald D" first="Derry" last="Fitzgerald">Derry Fitzgerald</name><affiliation wicri:level="1"><nlm:affiliation>Department of Electronic Engineering, Cork Institute of Technology, Cork, Ireland. derryfitz@eircom.net</nlm:affiliation><country xml:lang="fr">Irlande (pays)</country><wicri:regionArea>Department of Electronic Engineering, Cork Institute of Technology, Cork</wicri:regionArea><wicri:noRegion>Cork</wicri:noRegion></affiliation></author><author><name sortKey="Cranitch, Matt" sort="Cranitch, Matt" uniqKey="Cranitch M" first="Matt" last="Cranitch">Matt Cranitch</name></author><author><name sortKey="Coyle, Eugene" sort="Coyle, Eugene" uniqKey="Coyle E" first="Eugene" last="Coyle">Eugene Coyle</name></author></titleStmt><publicationStmt><idno type="wicri:source">PubMed</idno><date when="2008">2008</date><idno type="RBID">pubmed:18551178</idno><idno type="pmid">18551178</idno><idno type="doi">10.1155/2008/872425</idno><idno type="pmc">PMC2409961</idno><idno type="wicri:Area/Main/Corpus">001777</idno><idno type="wicri:explorRef" wicri:stream="Main" wicri:step="Corpus" wicri:corpus="PubMed">001777</idno><idno type="wicri:Area/Main/Curation">001777</idno><idno type="wicri:explorRef" wicri:stream="Main" wicri:step="Curation">001777</idno><idno type="wicri:Area/Main/Exploration">001777</idno></publicationStmt><sourceDesc><biblStruct><analytic><title xml:lang="en">Extended nonnegative tensor factorisation models for musical sound source separation.</title><author><name sortKey="Fitzgerald, Derry" sort="Fitzgerald, Derry" uniqKey="Fitzgerald D" first="Derry" last="Fitzgerald">Derry Fitzgerald</name><affiliation wicri:level="1"><nlm:affiliation>Department of Electronic Engineering, Cork Institute of Technology, Cork, Ireland. derryfitz@eircom.net</nlm:affiliation><country xml:lang="fr">Irlande (pays)</country><wicri:regionArea>Department of Electronic Engineering, Cork Institute of Technology, Cork</wicri:regionArea><wicri:noRegion>Cork</wicri:noRegion></affiliation></author><author><name sortKey="Cranitch, Matt" sort="Cranitch, Matt" uniqKey="Cranitch M" first="Matt" last="Cranitch">Matt Cranitch</name></author><author><name sortKey="Coyle, Eugene" sort="Coyle, Eugene" uniqKey="Coyle E" first="Eugene" last="Coyle">Eugene Coyle</name></author></analytic><series><title level="j">Computational intelligence and neuroscience</title><idno type="ISSN">1687-5265</idno><imprint><date when="2008" type="published">2008</date></imprint></series></biblStruct></sourceDesc></fileDesc><profileDesc><textClass></textClass></profileDesc></teiHeader><front><div type="abstract" xml:lang="en">Recently, shift-invariant tensor factorisation algorithms have been proposed for the purposes of sound source separation of pitched musical instruments. However, in practice, existing algorithms require the use of log-frequency spectrograms to allow shift invariance in frequency which causes problems when attempting to resynthesise the separated sources. Further, it is difficult to impose harmonicity constraints on the recovered basis functions. This paper proposes a new additive synthesis-based approach which allows the use of linear-frequency spectrograms as well as imposing strict harmonic constraints, resulting in an improved model. Further, these additional constraints allow the addition of a source filter model to the factorisation framework, and an extended model which is capable of separating mixtures of pitched and percussive instruments simultaneously.</div></front></TEI><pubmed><MedlineCitation Status="PubMed-not-MEDLINE" Owner="NLM"><PMID Version="1">18551178</PMID><DateCompleted><Year>2010</Year><Month>05</Month><Day>20</Day></DateCompleted><DateRevised><Year>2018</Year><Month>11</Month><Day>13</Day></DateRevised><Article PubModel="Print"><Journal><ISSN IssnType="Print">1687-5265</ISSN><JournalIssue CitedMedium="Print"><PubDate><Year>2008</Year></PubDate></JournalIssue><Title>Computational intelligence and neuroscience</Title><ISOAbbreviation>Comput Intell Neurosci</ISOAbbreviation></Journal><ArticleTitle>Extended nonnegative tensor factorisation models for musical sound source separation.</ArticleTitle><Pagination><MedlinePgn>872425</MedlinePgn></Pagination><ELocationID EIdType="doi" ValidYN="Y">10.1155/2008/872425</ELocationID><Abstract><AbstractText>Recently, shift-invariant tensor factorisation algorithms have been proposed for the purposes of sound source separation of pitched musical instruments. However, in practice, existing algorithms require the use of log-frequency spectrograms to allow shift invariance in frequency which causes problems when attempting to resynthesise the separated sources. Further, it is difficult to impose harmonicity constraints on the recovered basis functions. This paper proposes a new additive synthesis-based approach which allows the use of linear-frequency spectrograms as well as imposing strict harmonic constraints, resulting in an improved model. Further, these additional constraints allow the addition of a source filter model to the factorisation framework, and an extended model which is capable of separating mixtures of pitched and percussive instruments simultaneously.</AbstractText></Abstract><AuthorList CompleteYN="Y"><Author ValidYN="Y"><LastName>FitzGerald</LastName><ForeName>Derry</ForeName><Initials>D</Initials><AffiliationInfo><Affiliation>Department of Electronic Engineering, Cork Institute of Technology, Cork, Ireland. derryfitz@eircom.net</Affiliation></AffiliationInfo></Author><Author ValidYN="Y"><LastName>Cranitch</LastName><ForeName>Matt</ForeName><Initials>M</Initials></Author><Author ValidYN="Y"><LastName>Coyle</LastName><ForeName>Eugene</ForeName><Initials>E</Initials></Author></AuthorList><Language>eng</Language><PublicationTypeList><PublicationType UI="D016428">Journal Article</PublicationType></PublicationTypeList></Article><MedlineJournalInfo><Country>United States</Country><MedlineTA>Comput Intell Neurosci</MedlineTA><NlmUniqueID>101279357</NlmUniqueID></MedlineJournalInfo></MedlineCitation><PubmedData><History><PubMedPubDate PubStatus="received"><Year>2007</Year><Month>12</Month><Day>18</Day></PubMedPubDate><PubMedPubDate PubStatus="revised"><Year>2008</Year><Month>03</Month><Day>03</Day></PubMedPubDate><PubMedPubDate PubStatus="accepted"><Year>2008</Year><Month>04</Month><Day>17</Day></PubMedPubDate><PubMedPubDate PubStatus="pubmed"><Year>2008</Year><Month>6</Month><Day>14</Day><Hour>9</Hour><Minute>0</Minute></PubMedPubDate><PubMedPubDate PubStatus="medline"><Year>2008</Year><Month>6</Month><Day>14</Day><Hour>9</Hour><Minute>1</Minute></PubMedPubDate><PubMedPubDate PubStatus="entrez"><Year>2008</Year><Month>6</Month><Day>14</Day><Hour>9</Hour><Minute>0</Minute></PubMedPubDate></History><PublicationStatus>ppublish</PublicationStatus><ArticleIdList><ArticleId IdType="pubmed">18551178</ArticleId><ArticleId IdType="doi">10.1155/2008/872425</ArticleId><ArticleId IdType="pmc">PMC2409961</ArticleId></ArticleIdList><ReferenceList><Reference><Citation>Nature. 1999 Oct 21;401(6755):788-91</Citation><ArticleIdList><ArticleId IdType="pubmed">10548103</ArticleId></ArticleIdList></Reference><Reference><Citation>Curr Opin Neurobiol. 2004 Aug;14(4):481-7</Citation><ArticleIdList><ArticleId IdType="pubmed">15321069</ArticleId></ArticleIdList></Reference><Reference><Citation>Neural Comput. 2000 Feb;12(2):337-65</Citation><ArticleIdList><ArticleId IdType="pubmed">10636946</ArticleId></ArticleIdList></Reference></ReferenceList></PubmedData></pubmed><affiliations><list><country><li>Irlande (pays)</li></country></list><tree><noCountry><name sortKey="Coyle, Eugene" sort="Coyle, Eugene" uniqKey="Coyle E" first="Eugene" last="Coyle">Eugene Coyle</name><name sortKey="Cranitch, Matt" sort="Cranitch, Matt" uniqKey="Cranitch M" first="Matt" last="Cranitch">Matt Cranitch</name></noCountry><country name="Irlande (pays)"><noRegion><name sortKey="Fitzgerald, Derry" sort="Fitzgerald, Derry" uniqKey="Fitzgerald D" first="Derry" last="Fitzgerald">Derry Fitzgerald</name></noRegion></country></tree></affiliations></record>Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Sante/explor/SanteMusiqueV1/Data/Main/Exploration
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 001795 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Exploration/biblio.hfd -nk 001795 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien
|wiki= Sante
|area= SanteMusiqueV1
|flux= Main
|étape= Exploration
|type= RBID
|clé= pubmed:18551178
|texte= Extended nonnegative tensor factorisation models for musical sound source separation.
}}
Pour générer des pages wiki
HfdIndexSelect -h $EXPLOR_AREA/Data/Main/Exploration/RBID.i -Sk "pubmed:18551178" \
| HfdSelect -Kh $EXPLOR_AREA/Data/Main/Exploration/biblio.hfd \
| NlmPubMed2Wicri -a SanteMusiqueV1
| This area was generated with Dilib version V0.6.38. |  |