Unsupervised Music Structure Annotation by Time Series Structure Features and Segment Similarity : Music Data Mining
Identifieur interne : 000011 ( PascalFrancis/Curation ); précédent : 000010; suivant : 000012Unsupervised Music Structure Annotation by Time Series Structure Features and Segment Similarity : Music Data Mining
Auteurs : Joan Serra [Espagne] ; Meinard Müller [Allemagne] ; Peter Grosche [Allemagne] ; Josep Ll. Arcos [Espagne]Source :
- IEEE transactions on multimedia [ 1520-9210 ] ; 2014.
Descripteurs français
- Pascal (Inist)
- Multimédia, Recherche information, Similitude, Annotation, Présentation document, Délai d'exécution, Réalité terrain, Acoustique musicale, Série temporelle, Solution similitude, Système hiérarchisé, Méthode globale locale, Musique, Etude expérimentale, Acoustique audio, Apprentissage non supervisé, Recherche par contenu.
- Wicri :
- topic : Multimédia, Musique.
English descriptors
- KwdEn :
Abstract
Automatically inferring the structural properties of raw multimedia documents is essential in today's digitized society. Given its hierarchical and multi-faceted organization, musical pieces represent a challenge for current computational systems. In this article, we present a novel approach to music structure annotation based on the combination of structure features with time series similarity. Structure features encapsulate both local and global properties of a time series, and allow us to detect boundaries between homogeneous, novel, or repeated segments. Time series similarity is used to identify equivalent segments, corresponding to musically meaningful parts. Extensive tests with a total of five benchmark music collections and seven different human annotations show that the proposed approach is robust to different ground truth choices and parameter settings. Moreover, we see that it outperforms previous approaches evaluated under the same framework.
pA |
|
---|
Links toward previous steps (curation, corpus...)
- to stream PascalFrancis, to step Corpus: Pour aller vers cette notice dans l'étape Curation :000002
Links to Exploration step
Pascal:14-0220601Le document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en" level="a">Unsupervised Music Structure Annotation by Time Series Structure Features and Segment Similarity : Music Data Mining</title>
<author><name sortKey="Serra, Joan" sort="Serra, Joan" uniqKey="Serra J" first="Joan" last="Serra">Joan Serra</name>
<affiliation wicri:level="1"><inist:fA14 i1="01"><s1>IIIA-CSIC, Campus de la UAB s/n</s1>
<s2>08193 Bellaterra</s2>
<s3>ESP</s3>
<sZ>1 aut.</sZ>
<sZ>4 aut.</sZ>
</inist:fA14>
<country>Espagne</country>
</affiliation>
</author>
<author><name sortKey="Muller, Meinard" sort="Muller, Meinard" uniqKey="Muller M" first="Meinard" last="Müller">Meinard Müller</name>
<affiliation wicri:level="1"><inist:fA14 i1="02"><s1>International Audio Laboratories Erlangen</s1>
<s2>91058 Erlangen</s2>
<s3>DEU</s3>
<sZ>2 aut.</sZ>
</inist:fA14>
<country>Allemagne</country>
</affiliation>
</author>
<author><name sortKey="Grosche, Peter" sort="Grosche, Peter" uniqKey="Grosche P" first="Peter" last="Grosche">Peter Grosche</name>
<affiliation wicri:level="1"><inist:fA14 i1="03"><s1>Saarland University and the Max-Planck Institut für Informatik, Campus E1.4</s1>
<s2>66123 Saarbrücken</s2>
<s3>DEU</s3>
<sZ>3 aut.</sZ>
</inist:fA14>
<country>Allemagne</country>
</affiliation>
</author>
<author><name sortKey="Arcos, Josep Ll" sort="Arcos, Josep Ll" uniqKey="Arcos J" first="Josep Ll." last="Arcos">Josep Ll. Arcos</name>
<affiliation wicri:level="1"><inist:fA14 i1="01"><s1>IIIA-CSIC, Campus de la UAB s/n</s1>
<s2>08193 Bellaterra</s2>
<s3>ESP</s3>
<sZ>1 aut.</sZ>
<sZ>4 aut.</sZ>
</inist:fA14>
<country>Espagne</country>
</affiliation>
</author>
</titleStmt>
<publicationStmt><idno type="wicri:source">INIST</idno>
<idno type="inist">14-0220601</idno>
<date when="2014">2014</date>
<idno type="stanalyst">PASCAL 14-0220601 INIST</idno>
<idno type="RBID">Pascal:14-0220601</idno>
<idno type="wicri:Area/PascalFrancis/Corpus">000002</idno>
<idno type="wicri:Area/PascalFrancis/Curation">000011</idno>
</publicationStmt>
<sourceDesc><biblStruct><analytic><title xml:lang="en" level="a">Unsupervised Music Structure Annotation by Time Series Structure Features and Segment Similarity : Music Data Mining</title>
<author><name sortKey="Serra, Joan" sort="Serra, Joan" uniqKey="Serra J" first="Joan" last="Serra">Joan Serra</name>
<affiliation wicri:level="1"><inist:fA14 i1="01"><s1>IIIA-CSIC, Campus de la UAB s/n</s1>
<s2>08193 Bellaterra</s2>
<s3>ESP</s3>
<sZ>1 aut.</sZ>
<sZ>4 aut.</sZ>
</inist:fA14>
<country>Espagne</country>
</affiliation>
</author>
<author><name sortKey="Muller, Meinard" sort="Muller, Meinard" uniqKey="Muller M" first="Meinard" last="Müller">Meinard Müller</name>
<affiliation wicri:level="1"><inist:fA14 i1="02"><s1>International Audio Laboratories Erlangen</s1>
<s2>91058 Erlangen</s2>
<s3>DEU</s3>
<sZ>2 aut.</sZ>
</inist:fA14>
<country>Allemagne</country>
</affiliation>
</author>
<author><name sortKey="Grosche, Peter" sort="Grosche, Peter" uniqKey="Grosche P" first="Peter" last="Grosche">Peter Grosche</name>
<affiliation wicri:level="1"><inist:fA14 i1="03"><s1>Saarland University and the Max-Planck Institut für Informatik, Campus E1.4</s1>
<s2>66123 Saarbrücken</s2>
<s3>DEU</s3>
<sZ>3 aut.</sZ>
</inist:fA14>
<country>Allemagne</country>
</affiliation>
</author>
<author><name sortKey="Arcos, Josep Ll" sort="Arcos, Josep Ll" uniqKey="Arcos J" first="Josep Ll." last="Arcos">Josep Ll. Arcos</name>
<affiliation wicri:level="1"><inist:fA14 i1="01"><s1>IIIA-CSIC, Campus de la UAB s/n</s1>
<s2>08193 Bellaterra</s2>
<s3>ESP</s3>
<sZ>1 aut.</sZ>
<sZ>4 aut.</sZ>
</inist:fA14>
<country>Espagne</country>
</affiliation>
</author>
</analytic>
<series><title level="j" type="main">IEEE transactions on multimedia</title>
<title level="j" type="abbreviated">IEEE trans. multimedia</title>
<idno type="ISSN">1520-9210</idno>
<imprint><date when="2014">2014</date>
</imprint>
</series>
</biblStruct>
</sourceDesc>
<seriesStmt><title level="j" type="main">IEEE transactions on multimedia</title>
<title level="j" type="abbreviated">IEEE trans. multimedia</title>
<idno type="ISSN">1520-9210</idno>
</seriesStmt>
</fileDesc>
<profileDesc><textClass><keywords scheme="KwdEn" xml:lang="en"><term>Annotation</term>
<term>Audio acoustics</term>
<term>Content-based retrieval</term>
<term>Document layout</term>
<term>Experimental study</term>
<term>Global local method</term>
<term>Ground truth</term>
<term>Hierarchical system</term>
<term>Information retrieval</term>
<term>Multimedia</term>
<term>Music</term>
<term>Musical acoustics</term>
<term>Similarity</term>
<term>Similarity solution</term>
<term>Time allowed</term>
<term>Time series</term>
<term>Unsupervised learning</term>
</keywords>
<keywords scheme="Pascal" xml:lang="fr"><term>Multimédia</term>
<term>Recherche information</term>
<term>Similitude</term>
<term>Annotation</term>
<term>Présentation document</term>
<term>Délai d'exécution</term>
<term>Réalité terrain</term>
<term>Acoustique musicale</term>
<term>Série temporelle</term>
<term>Solution similitude</term>
<term>Système hiérarchisé</term>
<term>Méthode globale locale</term>
<term>Musique</term>
<term>Etude expérimentale</term>
<term>Acoustique audio</term>
<term>Apprentissage non supervisé</term>
<term>Recherche par contenu</term>
</keywords>
<keywords scheme="Wicri" type="topic" xml:lang="fr"><term>Multimédia</term>
<term>Musique</term>
</keywords>
</textClass>
</profileDesc>
</teiHeader>
<front><div type="abstract" xml:lang="en">Automatically inferring the structural properties of raw multimedia documents is essential in today's digitized society. Given its hierarchical and multi-faceted organization, musical pieces represent a challenge for current computational systems. In this article, we present a novel approach to music structure annotation based on the combination of structure features with time series similarity. Structure features encapsulate both local and global properties of a time series, and allow us to detect boundaries between homogeneous, novel, or repeated segments. Time series similarity is used to identify equivalent segments, corresponding to musically meaningful parts. Extensive tests with a total of five benchmark music collections and seven different human annotations show that the proposed approach is robust to different ground truth choices and parameter settings. Moreover, we see that it outperforms previous approaches evaluated under the same framework.</div>
</front>
</TEI>
<inist><standard h6="B"><pA><fA01 i1="01" i2="1"><s0>1520-9210</s0>
</fA01>
<fA03 i2="1"><s0>IEEE trans. multimedia</s0>
</fA03>
<fA05><s2>16</s2>
</fA05>
<fA06><s2>5</s2>
</fA06>
<fA08 i1="01" i2="1" l="ENG"><s1>Unsupervised Music Structure Annotation by Time Series Structure Features and Segment Similarity : Music Data Mining</s1>
</fA08>
<fA11 i1="01" i2="1"><s1>SERRA (Joan)</s1>
</fA11>
<fA11 i1="02" i2="1"><s1>MÜLLER (Meinard)</s1>
</fA11>
<fA11 i1="03" i2="1"><s1>GROSCHE (Peter)</s1>
</fA11>
<fA11 i1="04" i2="1"><s1>ARCOS (Josep Ll.)</s1>
</fA11>
<fA14 i1="01"><s1>IIIA-CSIC, Campus de la UAB s/n</s1>
<s2>08193 Bellaterra</s2>
<s3>ESP</s3>
<sZ>1 aut.</sZ>
<sZ>4 aut.</sZ>
</fA14>
<fA14 i1="02"><s1>International Audio Laboratories Erlangen</s1>
<s2>91058 Erlangen</s2>
<s3>DEU</s3>
<sZ>2 aut.</sZ>
</fA14>
<fA14 i1="03"><s1>Saarland University and the Max-Planck Institut für Informatik, Campus E1.4</s1>
<s2>66123 Saarbrücken</s2>
<s3>DEU</s3>
<sZ>3 aut.</sZ>
</fA14>
<fA20><s1>1229-1240</s1>
</fA20>
<fA21><s1>2014</s1>
</fA21>
<fA23 i1="01"><s0>ENG</s0>
</fA23>
<fA43 i1="01"><s1>INIST</s1>
<s2>26826</s2>
<s5>354000504838830050</s5>
</fA43>
<fA44><s0>0000</s0>
<s1>© 2014 INIST-CNRS. All rights reserved.</s1>
</fA44>
<fA45><s0>44 ref.</s0>
</fA45>
<fA47 i1="01" i2="1"><s0>14-0220601</s0>
</fA47>
<fA60><s1>P</s1>
</fA60>
<fA61><s0>A</s0>
</fA61>
<fA64 i1="01" i2="1"><s0>IEEE transactions on multimedia</s0>
</fA64>
<fA66 i1="01"><s0>USA</s0>
</fA66>
<fC01 i1="01" l="ENG"><s0>Automatically inferring the structural properties of raw multimedia documents is essential in today's digitized society. Given its hierarchical and multi-faceted organization, musical pieces represent a challenge for current computational systems. In this article, we present a novel approach to music structure annotation based on the combination of structure features with time series similarity. Structure features encapsulate both local and global properties of a time series, and allow us to detect boundaries between homogeneous, novel, or repeated segments. Time series similarity is used to identify equivalent segments, corresponding to musically meaningful parts. Extensive tests with a total of five benchmark music collections and seven different human annotations show that the proposed approach is robust to different ground truth choices and parameter settings. Moreover, we see that it outperforms previous approaches evaluated under the same framework.</s0>
</fC01>
<fC02 i1="01" i2="X"><s0>001D02C03</s0>
</fC02>
<fC02 i1="02" i2="X"><s0>001B40C75</s0>
</fC02>
<fC02 i1="03" i2="X"><s0>001D02B07D</s0>
</fC02>
<fC02 i1="04" i2="X"><s0>001B40C38</s0>
</fC02>
<fC03 i1="01" i2="X" l="FRE"><s0>Multimédia</s0>
<s5>06</s5>
</fC03>
<fC03 i1="01" i2="X" l="ENG"><s0>Multimedia</s0>
<s5>06</s5>
</fC03>
<fC03 i1="01" i2="X" l="SPA"><s0>Multimedia</s0>
<s5>06</s5>
</fC03>
<fC03 i1="02" i2="X" l="FRE"><s0>Recherche information</s0>
<s5>07</s5>
</fC03>
<fC03 i1="02" i2="X" l="ENG"><s0>Information retrieval</s0>
<s5>07</s5>
</fC03>
<fC03 i1="02" i2="X" l="SPA"><s0>Búsqueda información</s0>
<s5>07</s5>
</fC03>
<fC03 i1="03" i2="X" l="FRE"><s0>Similitude</s0>
<s5>15</s5>
</fC03>
<fC03 i1="03" i2="X" l="ENG"><s0>Similarity</s0>
<s5>15</s5>
</fC03>
<fC03 i1="03" i2="X" l="SPA"><s0>Similitud</s0>
<s5>15</s5>
</fC03>
<fC03 i1="04" i2="X" l="FRE"><s0>Annotation</s0>
<s5>18</s5>
</fC03>
<fC03 i1="04" i2="X" l="ENG"><s0>Annotation</s0>
<s5>18</s5>
</fC03>
<fC03 i1="04" i2="X" l="SPA"><s0>Anotación</s0>
<s5>18</s5>
</fC03>
<fC03 i1="05" i2="X" l="FRE"><s0>Présentation document</s0>
<s5>19</s5>
</fC03>
<fC03 i1="05" i2="X" l="ENG"><s0>Document layout</s0>
<s5>19</s5>
</fC03>
<fC03 i1="05" i2="X" l="SPA"><s0>Presentación documento</s0>
<s5>19</s5>
</fC03>
<fC03 i1="06" i2="X" l="FRE"><s0>Délai d'exécution</s0>
<s5>20</s5>
</fC03>
<fC03 i1="06" i2="X" l="ENG"><s0>Time allowed</s0>
<s5>20</s5>
</fC03>
<fC03 i1="06" i2="X" l="SPA"><s0>Plazo ejecución</s0>
<s5>20</s5>
</fC03>
<fC03 i1="07" i2="X" l="FRE"><s0>Réalité terrain</s0>
<s5>21</s5>
</fC03>
<fC03 i1="07" i2="X" l="ENG"><s0>Ground truth</s0>
<s5>21</s5>
</fC03>
<fC03 i1="07" i2="X" l="SPA"><s0>Realidad terreno</s0>
<s5>21</s5>
</fC03>
<fC03 i1="08" i2="X" l="FRE"><s0>Acoustique musicale</s0>
<s5>22</s5>
</fC03>
<fC03 i1="08" i2="X" l="ENG"><s0>Musical acoustics</s0>
<s5>22</s5>
</fC03>
<fC03 i1="08" i2="X" l="SPA"><s0>Acústica musical</s0>
<s5>22</s5>
</fC03>
<fC03 i1="09" i2="X" l="FRE"><s0>Série temporelle</s0>
<s5>23</s5>
</fC03>
<fC03 i1="09" i2="X" l="ENG"><s0>Time series</s0>
<s5>23</s5>
</fC03>
<fC03 i1="09" i2="X" l="SPA"><s0>Serie temporal</s0>
<s5>23</s5>
</fC03>
<fC03 i1="10" i2="X" l="FRE"><s0>Solution similitude</s0>
<s5>24</s5>
</fC03>
<fC03 i1="10" i2="X" l="ENG"><s0>Similarity solution</s0>
<s5>24</s5>
</fC03>
<fC03 i1="10" i2="X" l="SPA"><s0>Solución semejanza</s0>
<s5>24</s5>
</fC03>
<fC03 i1="11" i2="X" l="FRE"><s0>Système hiérarchisé</s0>
<s5>25</s5>
</fC03>
<fC03 i1="11" i2="X" l="ENG"><s0>Hierarchical system</s0>
<s5>25</s5>
</fC03>
<fC03 i1="11" i2="X" l="SPA"><s0>Sistema jerarquizado</s0>
<s5>25</s5>
</fC03>
<fC03 i1="12" i2="X" l="FRE"><s0>Méthode globale locale</s0>
<s5>26</s5>
</fC03>
<fC03 i1="12" i2="X" l="ENG"><s0>Global local method</s0>
<s5>26</s5>
</fC03>
<fC03 i1="12" i2="X" l="SPA"><s0>Método global local</s0>
<s5>26</s5>
</fC03>
<fC03 i1="13" i2="X" l="FRE"><s0>Musique</s0>
<s5>27</s5>
</fC03>
<fC03 i1="13" i2="X" l="ENG"><s0>Music</s0>
<s5>27</s5>
</fC03>
<fC03 i1="13" i2="X" l="SPA"><s0>Música</s0>
<s5>27</s5>
</fC03>
<fC03 i1="14" i2="X" l="FRE"><s0>Etude expérimentale</s0>
<s5>33</s5>
</fC03>
<fC03 i1="14" i2="X" l="ENG"><s0>Experimental study</s0>
<s5>33</s5>
</fC03>
<fC03 i1="14" i2="X" l="SPA"><s0>Estudio experimental</s0>
<s5>33</s5>
</fC03>
<fC03 i1="15" i2="X" l="FRE"><s0>Acoustique audio</s0>
<s4>CD</s4>
<s5>96</s5>
</fC03>
<fC03 i1="15" i2="X" l="ENG"><s0>Audio acoustics</s0>
<s4>CD</s4>
<s5>96</s5>
</fC03>
<fC03 i1="15" i2="X" l="SPA"><s0>Acústica audio</s0>
<s4>CD</s4>
<s5>96</s5>
</fC03>
<fC03 i1="16" i2="X" l="FRE"><s0>Apprentissage non supervisé</s0>
<s4>CD</s4>
<s5>97</s5>
</fC03>
<fC03 i1="16" i2="X" l="ENG"><s0>Unsupervised learning</s0>
<s4>CD</s4>
<s5>97</s5>
</fC03>
<fC03 i1="16" i2="X" l="SPA"><s0>Aprendizaje no supervisado</s0>
<s4>CD</s4>
<s5>97</s5>
</fC03>
<fC03 i1="17" i2="X" l="FRE"><s0>Recherche par contenu</s0>
<s4>CD</s4>
<s5>98</s5>
</fC03>
<fC03 i1="17" i2="X" l="ENG"><s0>Content-based retrieval</s0>
<s4>CD</s4>
<s5>98</s5>
</fC03>
<fC03 i1="17" i2="X" l="SPA"><s0>Búsqueda por Contenidos</s0>
<s4>CD</s4>
<s5>98</s5>
</fC03>
<fN21><s1>265</s1>
</fN21>
<fN44 i1="01"><s1>OTO</s1>
</fN44>
<fN82><s1>OTO</s1>
</fN82>
</pA>
</standard>
</inist>
</record>
Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Wicri/Sarre/explor/MusicSarreV3/Data/PascalFrancis/Curation
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 000011 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/PascalFrancis/Curation/biblio.hfd -nk 000011 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien |wiki= Wicri/Sarre |area= MusicSarreV3 |flux= PascalFrancis |étape= Curation |type= RBID |clé= Pascal:14-0220601 |texte= Unsupervised Music Structure Annotation by Time Series Structure Features and Segment Similarity : Music Data Mining }}
This area was generated with Dilib version V0.6.33. |