Automatic acoustic identification of individuals in multiple species: improving identification across recording conditions.
Identifieur interne : 000619 ( Main/Exploration ); précédent : 000618; suivant : 000620Automatic acoustic identification of individuals in multiple species: improving identification across recording conditions.
Auteurs : Dan Stowell [Royaume-Uni] ; Tereza Petrusková [République tchèque] ; Martin Šálek [République tchèque] ; Pavel Linhart [Pologne]Source :
- Journal of the Royal Society, Interface [ 1742-5662 ] ; 2019.
Descripteurs français
- KwdFr :
- MESH :
- classification : Animaux, Individualité, Modèles biologiques, Mâle, Oiseaux, Spécificité d'espèce, Vocalisation animale.
English descriptors
- KwdEn :
- MESH :
- classification : Birds, Vocalization, Animal.
- Animals, Individuality, Male, Models, Biological, Species Specificity.
Abstract
Many animals emit vocal sounds which, independently from the sounds' function, contain some individually distinctive signature. Thus the automatic recognition of individuals by sound is a potentially powerful tool for zoology and ecology research and practical monitoring. Here, we present a general automatic identification method that can work across multiple animal species with various levels of complexity in their communication systems. We further introduce new analysis techniques based on dataset manipulations that can evaluate the robustness and generality of a classifier. By using these techniques, we confirmed the presence of experimental confounds in situations resembling those from past studies. We introduce data manipulations that can reduce the impact of these confounds, compatible with any classifier. We suggest that assessment of confounds should become a standard part of future studies to ensure they do not report over-optimistic results. We provide annotated recordings used for analyses along with this study and we call for dataset sharing to be a common practice to enhance the development of methods and comparisons of results.
DOI: 10.1098/rsif.2018.0940
PubMed: 30966953
PubMed Central: PMC6505557
Affiliations:
Links toward previous steps (curation, corpus...)
Le document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en">Automatic acoustic identification of individuals in multiple species: improving identification across recording conditions.</title><author><name sortKey="Stowell, Dan" sort="Stowell, Dan" uniqKey="Stowell D" first="Dan" last="Stowell">Dan Stowell</name><affiliation wicri:level="1"><nlm:affiliation>1 Machine Listening Lab, Centre for Digital Music, Queen Mary University of London , London , UK.</nlm:affiliation><country xml:lang="fr">Royaume-Uni</country><wicri:regionArea>1 Machine Listening Lab, Centre for Digital Music, Queen Mary University of London , London </wicri:regionArea><wicri:noRegion>London </wicri:noRegion></affiliation></author><author><name sortKey="Petruskova, Tereza" sort="Petruskova, Tereza" uniqKey="Petruskova T" first="Tereza" last="Petrusková">Tereza Petrusková</name><affiliation wicri:level="1"><nlm:affiliation>2 Department of Ecology, Faculty of Science, Charles University , Prague , Czech Republic.</nlm:affiliation><country xml:lang="fr">République tchèque</country><wicri:regionArea>2 Department of Ecology, Faculty of Science, Charles University , Prague </wicri:regionArea><wicri:noRegion>Prague </wicri:noRegion></affiliation></author><author><name sortKey="Salek, Martin" sort="Salek, Martin" uniqKey="Salek M" first="Martin" last="Šálek">Martin Šálek</name><affiliation wicri:level="1"><nlm:affiliation>3 Institute of Vertebrate Biology, The Czech Academy of Sciences , Brno , Czech Republic.</nlm:affiliation><country xml:lang="fr">République tchèque</country><wicri:regionArea>3 Institute of Vertebrate Biology, The Czech Academy of Sciences , Brno </wicri:regionArea><wicri:noRegion>Brno </wicri:noRegion></affiliation><affiliation wicri:level="1"><nlm:affiliation>4 Faculty of Environmental Sciences, Czech University of Life Sciences Prague , Prague , Czech Republic.</nlm:affiliation><country xml:lang="fr">République tchèque</country><wicri:regionArea>4 Faculty of Environmental Sciences, Czech University of Life Sciences Prague , Prague </wicri:regionArea><wicri:noRegion>Prague </wicri:noRegion></affiliation></author><author><name sortKey="Linhart, Pavel" sort="Linhart, Pavel" uniqKey="Linhart P" first="Pavel" last="Linhart">Pavel Linhart</name><affiliation wicri:level="1"><nlm:affiliation>5 Department of Behavioural Ecology, Faculty of Biology, Adam Mickiewicz University , Poznań , Poland.</nlm:affiliation><country xml:lang="fr">Pologne</country><wicri:regionArea>5 Department of Behavioural Ecology, Faculty of Biology, Adam Mickiewicz University , Poznań </wicri:regionArea><wicri:noRegion>Poznań </wicri:noRegion></affiliation></author></titleStmt><publicationStmt><idno type="wicri:source">PubMed</idno><date when="2019">2019</date><idno type="RBID">pubmed:30966953</idno><idno type="pmid">30966953</idno><idno type="doi">10.1098/rsif.2018.0940</idno><idno type="pmc">PMC6505557</idno><idno type="wicri:Area/Main/Corpus">000542</idno><idno type="wicri:explorRef" wicri:stream="Main" wicri:step="Corpus" wicri:corpus="PubMed">000542</idno><idno type="wicri:Area/Main/Curation">000542</idno><idno type="wicri:explorRef" wicri:stream="Main" wicri:step="Curation">000542</idno><idno type="wicri:Area/Main/Exploration">000542</idno></publicationStmt><sourceDesc><biblStruct><analytic><title xml:lang="en">Automatic acoustic identification of individuals in multiple species: improving identification across recording conditions.</title><author><name sortKey="Stowell, Dan" sort="Stowell, Dan" uniqKey="Stowell D" first="Dan" last="Stowell">Dan Stowell</name><affiliation wicri:level="1"><nlm:affiliation>1 Machine Listening Lab, Centre for Digital Music, Queen Mary University of London , London , UK.</nlm:affiliation><country xml:lang="fr">Royaume-Uni</country><wicri:regionArea>1 Machine Listening Lab, Centre for Digital Music, Queen Mary University of London , London </wicri:regionArea><wicri:noRegion>London </wicri:noRegion></affiliation></author><author><name sortKey="Petruskova, Tereza" sort="Petruskova, Tereza" uniqKey="Petruskova T" first="Tereza" last="Petrusková">Tereza Petrusková</name><affiliation wicri:level="1"><nlm:affiliation>2 Department of Ecology, Faculty of Science, Charles University , Prague , Czech Republic.</nlm:affiliation><country xml:lang="fr">République tchèque</country><wicri:regionArea>2 Department of Ecology, Faculty of Science, Charles University , Prague </wicri:regionArea><wicri:noRegion>Prague </wicri:noRegion></affiliation></author><author><name sortKey="Salek, Martin" sort="Salek, Martin" uniqKey="Salek M" first="Martin" last="Šálek">Martin Šálek</name><affiliation wicri:level="1"><nlm:affiliation>3 Institute of Vertebrate Biology, The Czech Academy of Sciences , Brno , Czech Republic.</nlm:affiliation><country xml:lang="fr">République tchèque</country><wicri:regionArea>3 Institute of Vertebrate Biology, The Czech Academy of Sciences , Brno </wicri:regionArea><wicri:noRegion>Brno </wicri:noRegion></affiliation><affiliation wicri:level="1"><nlm:affiliation>4 Faculty of Environmental Sciences, Czech University of Life Sciences Prague , Prague , Czech Republic.</nlm:affiliation><country xml:lang="fr">République tchèque</country><wicri:regionArea>4 Faculty of Environmental Sciences, Czech University of Life Sciences Prague , Prague </wicri:regionArea><wicri:noRegion>Prague </wicri:noRegion></affiliation></author><author><name sortKey="Linhart, Pavel" sort="Linhart, Pavel" uniqKey="Linhart P" first="Pavel" last="Linhart">Pavel Linhart</name><affiliation wicri:level="1"><nlm:affiliation>5 Department of Behavioural Ecology, Faculty of Biology, Adam Mickiewicz University , Poznań , Poland.</nlm:affiliation><country xml:lang="fr">Pologne</country><wicri:regionArea>5 Department of Behavioural Ecology, Faculty of Biology, Adam Mickiewicz University , Poznań </wicri:regionArea><wicri:noRegion>Poznań </wicri:noRegion></affiliation></author></analytic><series><title level="j">Journal of the Royal Society, Interface</title><idno type="eISSN">1742-5662</idno><imprint><date when="2019" type="published">2019</date></imprint></series></biblStruct></sourceDesc></fileDesc><profileDesc><textClass><keywords scheme="KwdEn" xml:lang="en"><term>Animals (MeSH)</term><term>Birds (classification)</term><term>Individuality (MeSH)</term><term>Male (MeSH)</term><term>Models, Biological (MeSH)</term><term>Species Specificity (MeSH)</term><term>Vocalization, Animal (classification)</term></keywords><keywords scheme="KwdFr" xml:lang="fr"><term>Animaux (MeSH)</term><term>Individualité (MeSH)</term><term>Modèles biologiques (MeSH)</term><term>Mâle (MeSH)</term><term>Oiseaux (classification)</term><term>Spécificité d'espèce (MeSH)</term><term>Vocalisation animale (classification)</term></keywords><keywords scheme="MESH" qualifier="classification" xml:lang="en"><term>Birds</term><term>Vocalization, Animal</term></keywords><keywords scheme="MESH" xml:lang="en"><term>Animals</term><term>Individuality</term><term>Male</term><term>Models, Biological</term><term>Species Specificity</term></keywords><keywords scheme="MESH" qualifier="classification" xml:lang="fr"><term>Animaux</term><term>Individualité</term><term>Modèles biologiques</term><term>Mâle</term><term>Oiseaux</term><term>Spécificité d'espèce</term><term>Vocalisation animale</term></keywords></textClass></profileDesc></teiHeader><front><div type="abstract" xml:lang="en">Many animals emit vocal sounds which, independently from the sounds' function, contain some individually distinctive signature. Thus the automatic recognition of individuals by sound is a potentially powerful tool for zoology and ecology research and practical monitoring. Here, we present a general automatic identification method that can work across multiple animal species with various levels of complexity in their communication systems. We further introduce new analysis techniques based on dataset manipulations that can evaluate the robustness and generality of a classifier. By using these techniques, we confirmed the presence of experimental confounds in situations resembling those from past studies. We introduce data manipulations that can reduce the impact of these confounds, compatible with any classifier. We suggest that assessment of confounds should become a standard part of future studies to ensure they do not report over-optimistic results. We provide annotated recordings used for analyses along with this study and we call for dataset sharing to be a common practice to enhance the development of methods and comparisons of results.</div></front></TEI><pubmed><MedlineCitation Status="MEDLINE" Owner="NLM"><PMID Version="1">30966953</PMID><DateCompleted><Year>2020</Year><Month>05</Month><Day>29</Day></DateCompleted><DateRevised><Year>2020</Year><Month>05</Month><Day>29</Day></DateRevised><Article PubModel="Print"><Journal><ISSN IssnType="Electronic">1742-5662</ISSN><JournalIssue CitedMedium="Internet"><Volume>16</Volume><Issue>153</Issue><PubDate><Year>2019</Year><Month>04</Month><Day>26</Day></PubDate></JournalIssue><Title>Journal of the Royal Society, Interface</Title><ISOAbbreviation>J R Soc Interface</ISOAbbreviation></Journal><ArticleTitle>Automatic acoustic identification of individuals in multiple species: improving identification across recording conditions.</ArticleTitle><Pagination><MedlinePgn>20180940</MedlinePgn></Pagination><ELocationID EIdType="doi" ValidYN="Y">10.1098/rsif.2018.0940</ELocationID><Abstract><AbstractText>Many animals emit vocal sounds which, independently from the sounds' function, contain some individually distinctive signature. Thus the automatic recognition of individuals by sound is a potentially powerful tool for zoology and ecology research and practical monitoring. Here, we present a general automatic identification method that can work across multiple animal species with various levels of complexity in their communication systems. We further introduce new analysis techniques based on dataset manipulations that can evaluate the robustness and generality of a classifier. By using these techniques, we confirmed the presence of experimental confounds in situations resembling those from past studies. We introduce data manipulations that can reduce the impact of these confounds, compatible with any classifier. We suggest that assessment of confounds should become a standard part of future studies to ensure they do not report over-optimistic results. We provide annotated recordings used for analyses along with this study and we call for dataset sharing to be a common practice to enhance the development of methods and comparisons of results.</AbstractText></Abstract><AuthorList CompleteYN="Y"><Author ValidYN="Y"><LastName>Stowell</LastName><ForeName>Dan</ForeName><Initials>D</Initials><AffiliationInfo><Affiliation>1 Machine Listening Lab, Centre for Digital Music, Queen Mary University of London , London , UK.</Affiliation></AffiliationInfo></Author><Author ValidYN="Y"><LastName>Petrusková</LastName><ForeName>Tereza</ForeName><Initials>T</Initials><AffiliationInfo><Affiliation>2 Department of Ecology, Faculty of Science, Charles University , Prague , Czech Republic.</Affiliation></AffiliationInfo></Author><Author ValidYN="Y"><LastName>Šálek</LastName><ForeName>Martin</ForeName><Initials>M</Initials><AffiliationInfo><Affiliation>3 Institute of Vertebrate Biology, The Czech Academy of Sciences , Brno , Czech Republic.</Affiliation></AffiliationInfo><AffiliationInfo><Affiliation>4 Faculty of Environmental Sciences, Czech University of Life Sciences Prague , Prague , Czech Republic.</Affiliation></AffiliationInfo></Author><Author ValidYN="Y"><LastName>Linhart</LastName><ForeName>Pavel</ForeName><Initials>P</Initials><AffiliationInfo><Affiliation>5 Department of Behavioural Ecology, Faculty of Biology, Adam Mickiewicz University , Poznań , Poland.</Affiliation></AffiliationInfo></Author></AuthorList><Language>eng</Language><PublicationTypeList><PublicationType UI="D016428">Journal Article</PublicationType><PublicationType UI="D013485">Research Support, Non-U.S. Gov't</PublicationType></PublicationTypeList></Article><MedlineJournalInfo><Country>England</Country><MedlineTA>J R Soc Interface</MedlineTA><NlmUniqueID>101217269</NlmUniqueID><ISSNLinking>1742-5662</ISSNLinking></MedlineJournalInfo><CitationSubset>IM</CitationSubset><MeshHeadingList><MeshHeading><DescriptorName UI="D000818" MajorTopicYN="N">Animals</DescriptorName></MeshHeading><MeshHeading><DescriptorName UI="D001717" MajorTopicYN="N">Birds</DescriptorName><QualifierName UI="Q000145" MajorTopicYN="Y">classification</QualifierName></MeshHeading><MeshHeading><DescriptorName UI="D007206" MajorTopicYN="Y">Individuality</DescriptorName></MeshHeading><MeshHeading><DescriptorName UI="D008297" MajorTopicYN="N">Male</DescriptorName></MeshHeading><MeshHeading><DescriptorName UI="D008954" MajorTopicYN="N">Models, Biological</DescriptorName></MeshHeading><MeshHeading><DescriptorName UI="D013045" MajorTopicYN="N">Species Specificity</DescriptorName></MeshHeading><MeshHeading><DescriptorName UI="D014828" MajorTopicYN="N">Vocalization, Animal</DescriptorName><QualifierName UI="Q000145" MajorTopicYN="Y">classification</QualifierName></MeshHeading></MeshHeadingList><KeywordList Owner="NOTNLM"><Keyword MajorTopicYN="Y">acoustic monitoring</Keyword><Keyword MajorTopicYN="Y">animal communication</Keyword><Keyword MajorTopicYN="Y">data augmentation</Keyword><Keyword MajorTopicYN="Y">individuality</Keyword><Keyword MajorTopicYN="Y">song repertoire</Keyword><Keyword MajorTopicYN="Y">vocalization</Keyword></KeywordList></MedlineCitation><PubmedData><History><PubMedPubDate PubStatus="entrez"><Year>2019</Year><Month>4</Month><Day>11</Day><Hour>6</Hour><Minute>0</Minute></PubMedPubDate><PubMedPubDate PubStatus="pubmed"><Year>2019</Year><Month>4</Month><Day>11</Day><Hour>6</Hour><Minute>0</Minute></PubMedPubDate><PubMedPubDate PubStatus="medline"><Year>2020</Year><Month>5</Month><Day>30</Day><Hour>6</Hour><Minute>0</Minute></PubMedPubDate></History><PublicationStatus>ppublish</PublicationStatus><ArticleIdList><ArticleId IdType="pubmed">30966953</ArticleId><ArticleId IdType="doi">10.1098/rsif.2018.0940</ArticleId><ArticleId IdType="pmc">PMC6505557</ArticleId></ArticleIdList><ReferenceList><Reference><Citation>Biol Rev Camb Philos Soc. 2013 Feb;88(1):179-95</Citation><ArticleIdList><ArticleId IdType="pubmed">22978769</ArticleId></ArticleIdList></Reference><Reference><Citation>Trends Ecol Evol. 2005 Mar;20(3):143-9</Citation><ArticleIdList><ArticleId IdType="pubmed">16701358</ArticleId></ArticleIdList></Reference><Reference><Citation>PeerJ. 2014 Jul 17;2:e488</Citation><ArticleIdList><ArticleId IdType="pubmed">25083350</ArticleId></ArticleIdList></Reference><Reference><Citation>Biosystems. 1996;37(1-2):49-66</Citation><ArticleIdList><ArticleId IdType="pubmed">8924639</ArticleId></ArticleIdList></Reference><Reference><Citation>PLoS One. 2017 May 9;12(5):e0177206</Citation><ArticleIdList><ArticleId IdType="pubmed">28486488</ArticleId></ArticleIdList></Reference><Reference><Citation>Curr Biol. 2012 May 8;22(9):801-6</Citation><ArticleIdList><ArticleId IdType="pubmed">22521788</ArticleId></ArticleIdList></Reference><Reference><Citation>Behav Processes. 2013 Nov;100:139-45</Citation><ArticleIdList><ArticleId IdType="pubmed">24050853</ArticleId></ArticleIdList></Reference><Reference><Citation>Curr Zool. 2017 Aug;63(4):467-475</Citation><ArticleIdList><ArticleId IdType="pubmed">29492006</ArticleId></ArticleIdList></Reference><Reference><Citation>Evolution. 2012 Dec;66(12):3736-42</Citation><ArticleIdList><ArticleId IdType="pubmed">23206132</ArticleId></ArticleIdList></Reference><Reference><Citation>Front Zool. 2005 Jun 16;2(1):10</Citation><ArticleIdList><ArticleId IdType="pubmed">15960848</ArticleId></ArticleIdList></Reference><Reference><Citation>Nature. 2000 Jul 27;406(6794):404-5</Citation><ArticleIdList><ArticleId IdType="pubmed">10935635</ArticleId></ArticleIdList></Reference><Reference><Citation>Proc Biol Sci. 2012 Sep 22;279(1743):3749-55</Citation><ArticleIdList><ArticleId IdType="pubmed">22719031</ArticleId></ArticleIdList></Reference><Reference><Citation>Anim Behav. 1997 Jul;54(1):121-30</Citation><ArticleIdList><ArticleId IdType="pubmed">9268442</ArticleId></ArticleIdList></Reference><Reference><Citation>J Acoust Soc Am. 2010 Feb;127(2):874-83</Citation><ArticleIdList><ArticleId IdType="pubmed">20136210</ArticleId></ArticleIdList></Reference><Reference><Citation>J Exp Biol. 2014 Sep 1;217(Pt 17):3169-77</Citation><ArticleIdList><ArticleId IdType="pubmed">24948627</ArticleId></ArticleIdList></Reference><Reference><Citation>Neural Netw. 2012 Aug;32:333-8</Citation><ArticleIdList><ArticleId IdType="pubmed">22386783</ArticleId></ArticleIdList></Reference></ReferenceList></PubmedData></pubmed><affiliations><list><country><li>Pologne</li><li>Royaume-Uni</li><li>République tchèque</li></country></list><tree><country name="Royaume-Uni"><noRegion><name sortKey="Stowell, Dan" sort="Stowell, Dan" uniqKey="Stowell D" first="Dan" last="Stowell">Dan Stowell</name></noRegion></country><country name="République tchèque"><noRegion><name sortKey="Petruskova, Tereza" sort="Petruskova, Tereza" uniqKey="Petruskova T" first="Tereza" last="Petrusková">Tereza Petrusková</name></noRegion><name sortKey="Salek, Martin" sort="Salek, Martin" uniqKey="Salek M" first="Martin" last="Šálek">Martin Šálek</name><name sortKey="Salek, Martin" sort="Salek, Martin" uniqKey="Salek M" first="Martin" last="Šálek">Martin Šálek</name></country><country name="Pologne"><noRegion><name sortKey="Linhart, Pavel" sort="Linhart, Pavel" uniqKey="Linhart P" first="Pavel" last="Linhart">Pavel Linhart</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 000619 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Exploration/biblio.hfd -nk 000619 | 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:30966953
|texte= Automatic acoustic identification of individuals in multiple species: improving identification across recording conditions.
}}
Pour générer des pages wiki
HfdIndexSelect -h $EXPLOR_AREA/Data/Main/Exploration/RBID.i -Sk "pubmed:30966953" \
| HfdSelect -Kh $EXPLOR_AREA/Data/Main/Exploration/biblio.hfd \
| NlmPubMed2Wicri -a SanteMusiqueV1
| This area was generated with Dilib version V0.6.38. |  |