Discrete wavelet packet transform and ensembles of lazy and eager learners for music genre classification
Identifieur interne : 000037 ( Istex/Corpus ); précédent : 000036; suivant : 000038Discrete wavelet packet transform and ensembles of lazy and eager learners for music genre classification
Auteurs : Marco Grimaldi ; P Draig Cunningham ; Anil KokaramSource :
- Multimedia Systems [ 0942-4962 ] ; 2006-06-01.
English descriptors
- KwdEn :
Abstract
Abstract: This paper presents a process for determining the music genre of an item using a new set of descriptors. A discrete wavelet packet transform is applied to obtain the signal representation at two different resolutions: a frequency resolution and a time resolution tuned to encode music notes and their onset and offset. These features are tested on a number of data sets as descriptors for music genre classification. Lazy learning classifiers (k-nearest neighbor) and eager learners (neural networks and support vector machines) are applied in order to assess the classification power of the proposed features. Different feature selection techniques and ensemble methods are explored to maximize the accuracy of the classifiers and stabilize their behavior. Our evaluation shows that these frequency descriptors perform better than a standard approach based on Mel-Frequency Cepstral Coefficients and on the Short Time Fourier Transform in music genre classification. Moreover, our work confirms that a parameterization of the music rhythm based on the beat-histogram provides some meaningful information in the context of music classification by genre.Finally, our evaluation suggests that multi-class support vector machines with a linear kernel and round-robin binarization are the simplest and more effective process for music genre classification.
Url:
DOI: 10.1007/s00530-006-0027-z
Links to Exploration step
ISTEX:58FFDED8033DA0EC9ECB5BC3D171F6293F5AEA18Le document en format XML
<record><TEI wicri:istexFullTextTei="biblStruct"><teiHeader><fileDesc><titleStmt><title xml:lang="en">Discrete wavelet packet transform and ensembles of lazy and eager learners for music genre classification</title><author><name sortKey="Grimaldi, Marco" sort="Grimaldi, Marco" uniqKey="Grimaldi M" first="Marco" last="Grimaldi">Marco Grimaldi</name><affiliation><mods:affiliation>Computer Science Department, University College Dublin, Dublin, Ireland</mods:affiliation></affiliation><affiliation><mods:affiliation>E-mail: Marco.Grimaldi@ucd.ie</mods:affiliation></affiliation></author><author><name sortKey="Cunningham, P Draig" sort="Cunningham, P Draig" uniqKey="Cunningham P" first="P Draig" last="Cunningham">P Draig Cunningham</name><affiliation><mods:affiliation>Computer Science Department, Trinity College Dublin, Dublin, Ireland</mods:affiliation></affiliation><affiliation><mods:affiliation>E-mail: Padraig.Cunningham@cs.tcd.ie</mods:affiliation></affiliation></author><author><name sortKey="Kokaram, Anil" sort="Kokaram, Anil" uniqKey="Kokaram A" first="Anil" last="Kokaram">Anil Kokaram</name><affiliation><mods:affiliation>Electronic and Electrical Engineering Department, Trinity College Dublin, Dublin, Ireland</mods:affiliation></affiliation><affiliation><mods:affiliation>E-mail: Anil.Kokaram@tcd.ie</mods:affiliation></affiliation></author></titleStmt><publicationStmt><idno type="wicri:source">ISTEX</idno><idno type="RBID">ISTEX:58FFDED8033DA0EC9ECB5BC3D171F6293F5AEA18</idno><date when="2006" year="2006">2006</date><idno type="doi">10.1007/s00530-006-0027-z</idno><idno type="url">https://api.istex.fr/document/58FFDED8033DA0EC9ECB5BC3D171F6293F5AEA18/fulltext/pdf</idno><idno type="wicri:Area/Istex/Corpus">000037</idno><idno type="wicri:explorRef" wicri:stream="Istex" wicri:step="Corpus" wicri:corpus="ISTEX">000037</idno></publicationStmt><sourceDesc><biblStruct><analytic><title level="a" type="main" xml:lang="en">Discrete wavelet packet transform and ensembles of lazy and eager learners for music genre classification</title><author><name sortKey="Grimaldi, Marco" sort="Grimaldi, Marco" uniqKey="Grimaldi M" first="Marco" last="Grimaldi">Marco Grimaldi</name><affiliation><mods:affiliation>Computer Science Department, University College Dublin, Dublin, Ireland</mods:affiliation></affiliation><affiliation><mods:affiliation>E-mail: Marco.Grimaldi@ucd.ie</mods:affiliation></affiliation></author><author><name sortKey="Cunningham, P Draig" sort="Cunningham, P Draig" uniqKey="Cunningham P" first="P Draig" last="Cunningham">P Draig Cunningham</name><affiliation><mods:affiliation>Computer Science Department, Trinity College Dublin, Dublin, Ireland</mods:affiliation></affiliation><affiliation><mods:affiliation>E-mail: Padraig.Cunningham@cs.tcd.ie</mods:affiliation></affiliation></author><author><name sortKey="Kokaram, Anil" sort="Kokaram, Anil" uniqKey="Kokaram A" first="Anil" last="Kokaram">Anil Kokaram</name><affiliation><mods:affiliation>Electronic and Electrical Engineering Department, Trinity College Dublin, Dublin, Ireland</mods:affiliation></affiliation><affiliation><mods:affiliation>E-mail: Anil.Kokaram@tcd.ie</mods:affiliation></affiliation></author></analytic><monogr></monogr><series><title level="j">Multimedia Systems</title><title level="j" type="abbrev">Multimedia Systems</title><idno type="ISSN">0942-4962</idno><idno type="eISSN">1432-1882</idno><imprint><publisher>Springer-Verlag</publisher><pubPlace>Berlin/Heidelberg</pubPlace><date type="published" when="2006-06-01">2006-06-01</date><biblScope unit="volume">11</biblScope><biblScope unit="issue">5</biblScope><biblScope unit="page" from="422">422</biblScope><biblScope unit="page" to="437">437</biblScope></imprint><idno type="ISSN">0942-4962</idno></series></biblStruct></sourceDesc><seriesStmt><idno type="ISSN">0942-4962</idno></seriesStmt></fileDesc><profileDesc><textClass><keywords scheme="KwdEn" xml:lang="en"><term>Eager learners</term><term>Ensemble techniques</term><term>Feature extraction</term><term>Features selection</term><term>Lazy learners</term><term>Music genre classification</term><term>Wavelet analysis</term></keywords></textClass><langUsage><language ident="en">en</language></langUsage></profileDesc></teiHeader><front><div type="abstract" xml:lang="en">Abstract: This paper presents a process for determining the music genre of an item using a new set of descriptors. A discrete wavelet packet transform is applied to obtain the signal representation at two different resolutions: a frequency resolution and a time resolution tuned to encode music notes and their onset and offset. These features are tested on a number of data sets as descriptors for music genre classification. Lazy learning classifiers (k-nearest neighbor) and eager learners (neural networks and support vector machines) are applied in order to assess the classification power of the proposed features. Different feature selection techniques and ensemble methods are explored to maximize the accuracy of the classifiers and stabilize their behavior. Our evaluation shows that these frequency descriptors perform better than a standard approach based on Mel-Frequency Cepstral Coefficients and on the Short Time Fourier Transform in music genre classification. Moreover, our work confirms that a parameterization of the music rhythm based on the beat-histogram provides some meaningful information in the context of music classification by genre.Finally, our evaluation suggests that multi-class support vector machines with a linear kernel and round-robin binarization are the simplest and more effective process for music genre classification.</div></front></TEI><istex><corpusName>springer-journals</corpusName><author><json:item><name>Marco Grimaldi</name><affiliations><json:string>Computer Science Department, University College Dublin, Dublin, Ireland</json:string><json:string>E-mail: Marco.Grimaldi@ucd.ie</json:string></affiliations></json:item><json:item><name>P´draig Cunningham</name><affiliations><json:string>Computer Science Department, Trinity College Dublin, Dublin, Ireland</json:string><json:string>E-mail: Padraig.Cunningham@cs.tcd.ie</json:string></affiliations></json:item><json:item><name>Anil Kokaram</name><affiliations><json:string>Electronic and Electrical Engineering Department, Trinity College Dublin, Dublin, Ireland</json:string><json:string>E-mail: Anil.Kokaram@tcd.ie</json:string></affiliations></json:item></author><subject><json:item><lang><json:string>eng</json:string></lang><value>Music genre classification</value></json:item><json:item><lang><json:string>eng</json:string></lang><value>Wavelet analysis</value></json:item><json:item><lang><json:string>eng</json:string></lang><value>Feature extraction</value></json:item><json:item><lang><json:string>eng</json:string></lang><value>Ensemble techniques</value></json:item><json:item><lang><json:string>eng</json:string></lang><value>Features selection</value></json:item><json:item><lang><json:string>eng</json:string></lang><value>Lazy learners</value></json:item><json:item><lang><json:string>eng</json:string></lang><value>Eager learners</value></json:item></subject><articleId><json:string>27</json:string><json:string>s00530-006-0027-z</json:string></articleId><arkIstex>ark:/67375/VQC-HR13F38Q-H</arkIstex><language><json:string>eng</json:string></language><originalGenre><json:string>OriginalPaper</json:string></originalGenre><abstract>Abstract: This paper presents a process for determining the music genre of an item using a new set of descriptors. A discrete wavelet packet transform is applied to obtain the signal representation at two different resolutions: a frequency resolution and a time resolution tuned to encode music notes and their onset and offset. These features are tested on a number of data sets as descriptors for music genre classification. Lazy learning classifiers (k-nearest neighbor) and eager learners (neural networks and support vector machines) are applied in order to assess the classification power of the proposed features. Different feature selection techniques and ensemble methods are explored to maximize the accuracy of the classifiers and stabilize their behavior. Our evaluation shows that these frequency descriptors perform better than a standard approach based on Mel-Frequency Cepstral Coefficients and on the Short Time Fourier Transform in music genre classification. Moreover, our work confirms that a parameterization of the music rhythm based on the beat-histogram provides some meaningful information in the context of music classification by genre.Finally, our evaluation suggests that multi-class support vector machines with a linear kernel and round-robin binarization are the simplest and more effective process for music genre classification.</abstract><qualityIndicators><score>9.376</score><pdfWordCount>9313</pdfWordCount><pdfCharCount>57468</pdfCharCount><pdfVersion>1.3</pdfVersion><pdfPageCount>16</pdfPageCount><pdfPageSize>595 x 785 pts</pdfPageSize><refBibsNative>false</refBibsNative><abstractWordCount>198</abstractWordCount><abstractCharCount>1362</abstractCharCount><keywordCount>7</keywordCount></qualityIndicators><title>Discrete wavelet packet transform and ensembles of lazy and eager learners for music genre classification</title><genre><json:string>research-article</json:string></genre><host><title>Multimedia Systems</title><language><json:string>unknown</json:string></language><publicationDate>2006</publicationDate><copyrightDate>2006</copyrightDate><issn><json:string>0942-4962</json:string></issn><eissn><json:string>1432-1882</json:string></eissn><journalId><json:string>530</json:string></journalId><volume>11</volume><issue>5</issue><pages><first>422</first><last>437</last></pages><genre><json:string>journal</json:string></genre></host><ark><json:string>ark:/67375/VQC-HR13F38Q-H</json:string></ark><publicationDate>2006</publicationDate><copyrightDate>2006</copyrightDate><doi><json:string>10.1007/s00530-006-0027-z</json:string></doi><id>58FFDED8033DA0EC9ECB5BC3D171F6293F5AEA18</id><score>1</score><fulltext><json:item><extension>pdf</extension><original>true</original><mimetype>application/pdf</mimetype><uri>https://api.istex.fr/document/58FFDED8033DA0EC9ECB5BC3D171F6293F5AEA18/fulltext/pdf</uri></json:item><json:item><extension>zip</extension><original>false</original><mimetype>application/zip</mimetype><uri>https://api.istex.fr/document/58FFDED8033DA0EC9ECB5BC3D171F6293F5AEA18/fulltext/zip</uri></json:item><json:item><extension>txt</extension><original>false</original><mimetype>text/plain</mimetype><uri>https://api.istex.fr/document/58FFDED8033DA0EC9ECB5BC3D171F6293F5AEA18/fulltext/txt</uri></json:item><istex:fulltextTEI uri="https://api.istex.fr/document/58FFDED8033DA0EC9ECB5BC3D171F6293F5AEA18/fulltext/tei"><teiHeader><fileDesc><titleStmt><title level="a" type="main" xml:lang="en">Discrete wavelet packet transform and ensembles of lazy and eager learners for music genre classification</title><respStmt><resp>Références bibliographiques récupérées via GROBID</resp><name resp="ISTEX-API">ISTEX-API (INIST-CNRS)</name></respStmt></titleStmt><publicationStmt><authority>ISTEX</authority><publisher scheme="https://publisher-list.data.istex.fr">Springer-Verlag</publisher><pubPlace>Berlin/Heidelberg</pubPlace><availability><licence><p>Springer-Verlag, 2006</p></licence><p scheme="https://loaded-corpus.data.istex.fr/ark:/67375/XBH-3XSW68JL-F">springer</p></availability><date>2006</date></publicationStmt><notesStmt><note type="research-article" scheme="https://content-type.data.istex.fr/ark:/67375/XTP-1JC4F85T-7">research-article</note><note type="journal" scheme="https://publication-type.data.istex.fr/ark:/67375/JMC-0GLKJH51-B">journal</note><note>Regular Paper</note></notesStmt><sourceDesc><biblStruct type="inbook"><analytic><title level="a" type="main" xml:lang="en">Discrete wavelet packet transform and ensembles of lazy and eager learners for music genre classification</title><author xml:id="author-0000" corresp="yes"><persName><forename type="first">Marco</forename><surname>Grimaldi</surname></persName><email>Marco.Grimaldi@ucd.ie</email><affiliation>Computer Science Department, University College Dublin, Dublin, Ireland</affiliation></author><author xml:id="author-0001"><persName><forename type="first">P´draig</forename><surname>Cunningham</surname></persName><email>Padraig.Cunningham@cs.tcd.ie</email><affiliation>Computer Science Department, Trinity College Dublin, Dublin, Ireland</affiliation></author><author xml:id="author-0002"><persName><forename type="first">Anil</forename><surname>Kokaram</surname></persName><email>Anil.Kokaram@tcd.ie</email><affiliation>Electronic and Electrical Engineering Department, Trinity College Dublin, Dublin, Ireland</affiliation></author><idno type="istex">58FFDED8033DA0EC9ECB5BC3D171F6293F5AEA18</idno><idno type="ark">ark:/67375/VQC-HR13F38Q-H</idno><idno type="DOI">10.1007/s00530-006-0027-z</idno><idno type="article-id">27</idno><idno type="article-id">s00530-006-0027-z</idno></analytic><monogr><title level="j">Multimedia Systems</title><title level="j" type="abbrev">Multimedia Systems</title><idno type="pISSN">0942-4962</idno><idno type="eISSN">1432-1882</idno><idno type="journal-ID">true</idno><idno type="issue-article-count">8</idno><idno type="volume-issue-count">6</idno><imprint><publisher>Springer-Verlag</publisher><pubPlace>Berlin/Heidelberg</pubPlace><date type="published" when="2006-06-01"></date><biblScope unit="volume">11</biblScope><biblScope unit="issue">5</biblScope><biblScope unit="page" from="422">422</biblScope><biblScope unit="page" to="437">437</biblScope></imprint></monogr></biblStruct></sourceDesc></fileDesc><profileDesc><creation><date>2006</date></creation><langUsage><language ident="en">en</language></langUsage><abstract xml:lang="en"><p>Abstract: This paper presents a process for determining the music genre of an item using a new set of descriptors. A discrete wavelet packet transform is applied to obtain the signal representation at two different resolutions: a frequency resolution and a time resolution tuned to encode music notes and their onset and offset. These features are tested on a number of data sets as descriptors for music genre classification. Lazy learning classifiers (k-nearest neighbor) and eager learners (neural networks and support vector machines) are applied in order to assess the classification power of the proposed features. Different feature selection techniques and ensemble methods are explored to maximize the accuracy of the classifiers and stabilize their behavior. Our evaluation shows that these frequency descriptors perform better than a standard approach based on Mel-Frequency Cepstral Coefficients and on the Short Time Fourier Transform in music genre classification. Moreover, our work confirms that a parameterization of the music rhythm based on the beat-histogram provides some meaningful information in the context of music classification by genre.Finally, our evaluation suggests that multi-class support vector machines with a linear kernel and round-robin binarization are the simplest and more effective process for music genre classification.</p></abstract><textClass xml:lang="en"><keywords scheme="keyword"><list><head>Keywords</head><item><term>Music genre classification</term></item><item><term>Wavelet analysis</term></item><item><term>Feature extraction</term></item><item><term>Ensemble techniques</term></item><item><term>Features selection</term></item><item><term>Lazy learners</term></item><item><term>Eager learners</term></item></list></keywords></textClass></profileDesc><revisionDesc><change when="2006-06-01">Published</change><change xml:id="refBibs-istex" who="#ISTEX-API" when="2017-12-1">References added</change></revisionDesc></teiHeader></istex:fulltextTEI></fulltext><metadata><istex:metadataXml wicri:clean="corpus springer-journals not found" wicri:toSee="no header"><istex:xmlDeclaration>version="1.0" encoding="UTF-8"</istex:xmlDeclaration><istex:docType PUBLIC="-//Springer-Verlag//DTD A++ V2.4//EN" URI="http://devel.springer.de/A++/V2.4/DTD/A++V2.4.dtd" name="istex:docType"></istex:docType><istex:document><Publisher><PublisherInfo><PublisherName>Springer-Verlag</PublisherName><PublisherLocation>Berlin/Heidelberg</PublisherLocation></PublisherInfo><Journal OutputMedium="All"><JournalInfo JournalProductType="ArchiveJournal" NumberingStyle="ContentOnly"><JournalID>530</JournalID><JournalPrintISSN>0942-4962</JournalPrintISSN><JournalElectronicISSN>1432-1882</JournalElectronicISSN><JournalTitle>Multimedia Systems</JournalTitle><JournalAbbreviatedTitle>Multimedia Systems</JournalAbbreviatedTitle></JournalInfo><Volume OutputMedium="All"><VolumeInfo TocLevels="0" VolumeType="Regular"><VolumeIDStart>11</VolumeIDStart><VolumeIDEnd>11</VolumeIDEnd><VolumeIssueCount>6</VolumeIssueCount></VolumeInfo><Issue IssueType="Regular" OutputMedium="All"><IssueInfo TocLevels="0"><IssueIDStart>5</IssueIDStart><IssueIDEnd>5</IssueIDEnd><IssueArticleCount>8</IssueArticleCount><IssueHistory><OnlineDate><Year>2006</Year><Month>6</Month><Day>2</Day></OnlineDate><PrintDate><Year>2006</Year><Month>6</Month><Day>2</Day></PrintDate><CoverDate><Year>2006</Year><Month>6</Month></CoverDate></IssueHistory><IssueCopyright><CopyrightHolderName>Springer-Verlag</CopyrightHolderName><CopyrightYear>2006</CopyrightYear></IssueCopyright></IssueInfo><Article ID="s00530-006-0027-z" OutputMedium="All"><ArticleInfo ArticleType="OriginalPaper" ContainsESM="No" Language="En" NumberingStyle="ContentOnly" TocLevels="0"><ArticleID>27</ArticleID><ArticleDOI>10.1007/s00530-006-0027-z</ArticleDOI><ArticleSequenceNumber>3</ArticleSequenceNumber><ArticleTitle Language="En">Discrete wavelet packet transform and ensembles of lazy and eager learners for music genre classification</ArticleTitle><ArticleCategory>Regular Paper</ArticleCategory><ArticleFirstPage>422</ArticleFirstPage><ArticleLastPage>437</ArticleLastPage><ArticleHistory><RegistrationDate><Year>2006</Year><Month>2</Month><Day>24</Day></RegistrationDate><OnlineDate><Year>2006</Year><Month>4</Month><Day>20</Day></OnlineDate></ArticleHistory><ArticleCopyright><CopyrightHolderName>Springer-Verlag</CopyrightHolderName><CopyrightYear>2006</CopyrightYear></ArticleCopyright><ArticleGrants Type="Regular"><MetadataGrant Grant="OpenAccess"></MetadataGrant><AbstractGrant Grant="OpenAccess"></AbstractGrant><BodyPDFGrant Grant="Restricted"></BodyPDFGrant><BodyHTMLGrant Grant="Restricted"></BodyHTMLGrant><BibliographyGrant Grant="Restricted"></BibliographyGrant><ESMGrant Grant="Restricted"></ESMGrant></ArticleGrants></ArticleInfo><ArticleHeader><AuthorGroup><Author AffiliationIDS="Aff1" CorrespondingAffiliationID="Aff1"><AuthorName DisplayOrder="Western"><GivenName>Marco</GivenName><FamilyName>Grimaldi</FamilyName></AuthorName><Contact><Email>Marco.Grimaldi@ucd.ie</Email></Contact></Author><Author AffiliationIDS="Aff2"><AuthorName DisplayOrder="Western"><GivenName>P´draig</GivenName><FamilyName>Cunningham</FamilyName></AuthorName><Contact><Email>Padraig.Cunningham@cs.tcd.ie</Email></Contact></Author><Author AffiliationIDS="Aff3"><AuthorName DisplayOrder="Western"><GivenName>Anil</GivenName><FamilyName>Kokaram</FamilyName></AuthorName><Contact><Email>Anil.Kokaram@tcd.ie</Email></Contact></Author><Affiliation ID="Aff1"><OrgDivision>Computer Science Department</OrgDivision><OrgName>University College Dublin</OrgName><OrgAddress><City>Dublin</City><Country Code="IE">Ireland</Country></OrgAddress></Affiliation><Affiliation ID="Aff2"><OrgDivision>Computer Science Department</OrgDivision><OrgName>Trinity College Dublin</OrgName><OrgAddress><City>Dublin</City><Country Code="IE">Ireland</Country></OrgAddress></Affiliation><Affiliation ID="Aff3"><OrgDivision>Electronic and Electrical Engineering Department</OrgDivision><OrgName>Trinity College Dublin</OrgName><OrgAddress><City>Dublin</City><Country Code="IE">Ireland</Country></OrgAddress></Affiliation></AuthorGroup><Abstract ID="Abs1" Language="En"><Heading>Abstract</Heading><Para>This paper presents a process for determining the music genre of an item using a new set of descriptors. A discrete wavelet packet transform is applied to obtain the signal representation at two different resolutions: a frequency resolution and a time resolution tuned to encode music notes and their onset and offset. These features are tested on a number of data sets as descriptors for music genre classification. Lazy learning classifiers (<Emphasis Type="Italic">k</Emphasis>-nearest neighbor) and eager learners (neural networks and support vector machines) are applied in order to assess the classification power of the proposed features. Different feature selection techniques and ensemble methods are explored to maximize the accuracy of the classifiers and stabilize their behavior. Our evaluation shows that these frequency descriptors perform better than a standard approach based on Mel-Frequency Cepstral Coefficients and on the Short Time Fourier Transform in music genre classification. Moreover, our work confirms that a parameterization of the music rhythm based on the beat-histogram provides some meaningful information in the context of music classification by genre.Finally, our evaluation suggests that multi-class support vector machines with a linear kernel and round-robin binarization are the simplest and more effective process for music genre classification.</Para></Abstract><KeywordGroup Language="En"><Heading>Keywords</Heading><Keyword>Music genre classification</Keyword><Keyword>Wavelet analysis</Keyword><Keyword>Feature extraction</Keyword><Keyword>Ensemble techniques</Keyword><Keyword>Features selection</Keyword><Keyword>Lazy learners</Keyword><Keyword>Eager learners</Keyword></KeywordGroup></ArticleHeader><NoBody></NoBody></Article></Issue></Volume></Journal></Publisher></istex:document></istex:metadataXml><mods version="3.6"><titleInfo lang="en"><title>Discrete wavelet packet transform and ensembles of lazy and eager learners for music genre classification</title></titleInfo><titleInfo type="alternative" contentType="CDATA" lang="en"><title>Discrete wavelet packet transform and ensembles of lazy and eager learners for music genre classification</title></titleInfo><name type="personal" displayLabel="corresp"><namePart type="given">Marco</namePart><namePart type="family">Grimaldi</namePart><affiliation>Computer Science Department, University College Dublin, Dublin, Ireland</affiliation><affiliation>E-mail: Marco.Grimaldi@ucd.ie</affiliation><role><roleTerm type="text">author</roleTerm></role></name><name type="personal"><namePart type="given">P´draig</namePart><namePart type="family">Cunningham</namePart><affiliation>Computer Science Department, Trinity College Dublin, Dublin, Ireland</affiliation><affiliation>E-mail: Padraig.Cunningham@cs.tcd.ie</affiliation><role><roleTerm type="text">author</roleTerm></role></name><name type="personal"><namePart type="given">Anil</namePart><namePart type="family">Kokaram</namePart><affiliation>Electronic and Electrical Engineering Department, Trinity College Dublin, Dublin, Ireland</affiliation><affiliation>E-mail: Anil.Kokaram@tcd.ie</affiliation><role><roleTerm type="text">author</roleTerm></role></name><typeOfResource>text</typeOfResource><genre type="research-article" displayLabel="OriginalPaper" authority="ISTEX" authorityURI="https://content-type.data.istex.fr" valueURI="https://content-type.data.istex.fr/ark:/67375/XTP-1JC4F85T-7">research-article</genre><originInfo><publisher>Springer-Verlag</publisher><place><placeTerm type="text">Berlin/Heidelberg</placeTerm></place><dateIssued encoding="w3cdtf">2006-06-01</dateIssued><dateIssued encoding="w3cdtf">2006</dateIssued><copyrightDate encoding="w3cdtf">2006</copyrightDate></originInfo><language><languageTerm type="code" authority="rfc3066">en</languageTerm><languageTerm type="code" authority="iso639-2b">eng</languageTerm></language><abstract lang="en">Abstract: This paper presents a process for determining the music genre of an item using a new set of descriptors. A discrete wavelet packet transform is applied to obtain the signal representation at two different resolutions: a frequency resolution and a time resolution tuned to encode music notes and their onset and offset. These features are tested on a number of data sets as descriptors for music genre classification. Lazy learning classifiers (k-nearest neighbor) and eager learners (neural networks and support vector machines) are applied in order to assess the classification power of the proposed features. Different feature selection techniques and ensemble methods are explored to maximize the accuracy of the classifiers and stabilize their behavior. Our evaluation shows that these frequency descriptors perform better than a standard approach based on Mel-Frequency Cepstral Coefficients and on the Short Time Fourier Transform in music genre classification. Moreover, our work confirms that a parameterization of the music rhythm based on the beat-histogram provides some meaningful information in the context of music classification by genre.Finally, our evaluation suggests that multi-class support vector machines with a linear kernel and round-robin binarization are the simplest and more effective process for music genre classification.</abstract><note>Regular Paper</note><subject lang="en"><genre>Keywords</genre><topic>Music genre classification</topic><topic>Wavelet analysis</topic><topic>Feature extraction</topic><topic>Ensemble techniques</topic><topic>Features selection</topic><topic>Lazy learners</topic><topic>Eager learners</topic></subject><relatedItem type="host"><titleInfo><title>Multimedia Systems</title></titleInfo><titleInfo type="abbreviated"><title>Multimedia Systems</title></titleInfo><genre type="journal" displayLabel="Archive Journal" authority="ISTEX" valueURI="https://publication-type.data.istex.fr/ark:/67375/JMC-0GLKJH51-B">journal</genre><originInfo><publisher>Springer</publisher><dateIssued encoding="w3cdtf">2006-06-02</dateIssued><copyrightDate encoding="w3cdtf">2006</copyrightDate></originInfo><identifier type="ISSN">0942-4962</identifier><identifier type="eISSN">1432-1882</identifier><identifier type="JournalID">530</identifier><identifier type="IssueArticleCount">8</identifier><identifier type="VolumeIssueCount">6</identifier><part><date>2006</date><detail type="volume"><number>11</number><caption>vol.</caption></detail><detail type="issue"><number>5</number><caption>no.</caption></detail><extent unit="pages"><start>422</start><end>437</end></extent></part><recordInfo><recordOrigin>Springer-Verlag, 2006</recordOrigin></recordInfo></relatedItem><identifier type="istex">58FFDED8033DA0EC9ECB5BC3D171F6293F5AEA18</identifier><identifier type="ark">ark:/67375/VQC-HR13F38Q-H</identifier><identifier type="DOI">10.1007/s00530-006-0027-z</identifier><identifier type="ArticleID">27</identifier><identifier type="ArticleID">s00530-006-0027-z</identifier><accessCondition type="use and reproduction" contentType="copyright">Springer-Verlag, 2006</accessCondition><recordInfo><recordContentSource authority="ISTEX" authorityURI="https://loaded-corpus.data.istex.fr" valueURI="https://loaded-corpus.data.istex.fr/ark:/67375/XBH-3XSW68JL-F">springer</recordContentSource><recordOrigin>Springer-Verlag, 2006</recordOrigin></recordInfo></mods><json:item><extension>json</extension><original>false</original><mimetype>application/json</mimetype><uri>https://api.istex.fr/document/58FFDED8033DA0EC9ECB5BC3D171F6293F5AEA18/metadata/json</uri></json:item></metadata><serie></serie></istex></record>Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Wicri/Musique/explor/MusiqueCeltiqueV1/Data/Istex/Corpus
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 000037 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Istex/Corpus/biblio.hfd -nk 000037 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien
|wiki= Wicri/Musique
|area= MusiqueCeltiqueV1
|flux= Istex
|étape= Corpus
|type= RBID
|clé= ISTEX:58FFDED8033DA0EC9ECB5BC3D171F6293F5AEA18
|texte= Discrete wavelet packet transform and ensembles of lazy and eager learners for music genre classification
}}
| This area was generated with Dilib version V0.6.38. |  |