Cue integration with categories: Weighting acoustic cues in speech using unsupervised learning and distributional statistics
Identifieur interne : 001924 ( Ncbi/Merge ); précédent : 001923; suivant : 001925Cue integration with categories: Weighting acoustic cues in speech using unsupervised learning and distributional statistics
Auteurs : Joseph C. Toscano ; Bob McmurraySource :
- Cognitive science [ 0364-0213 ] ; 2010.
Abstract
During speech perception, listeners make judgments about the phonological category of sounds by taking advantage of multiple acoustic cues for each phonological contrast. Perceptual experiments have shown that listeners weight these cues differently. How do listeners weight and combine acoustic cues to arrive at an overall estimate of the category for a speech sound? Here, we present several simulations using mixture of Gaussians (MOG) models that learn cue weights and combine cues on the basis of their distributional statistics. We show that a cue-weighting metric in which cues receive weight as a function of their reliability at distinguishing the phonological categories provides a good fit to the perceptual data obtained from human listeners, but only when these weights emerge through the dynamics of learning. These results suggest that cue weights can be readily extracted from the speech signal through unsupervised learning processes.
Url:
DOI: 10.1111/j.1551-6709.2009.01077.x
PubMed: 21339861
PubMed Central: 3039883
Links toward previous steps (curation, corpus...)
- to stream Pmc, to step Corpus: 000F77
- to stream Pmc, to step Curation: 000F77
- to stream Pmc, to step Checkpoint: 001F27
Links to Exploration step
PMC:3039883Le document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en">Cue integration with categories: Weighting acoustic cues in speech using unsupervised learning and distributional statistics</title><author><name sortKey="Toscano, Joseph C" sort="Toscano, Joseph C" uniqKey="Toscano J" first="Joseph C." last="Toscano">Joseph C. Toscano</name></author><author><name sortKey="Mcmurray, Bob" sort="Mcmurray, Bob" uniqKey="Mcmurray B" first="Bob" last="Mcmurray">Bob Mcmurray</name></author></titleStmt><publicationStmt><idno type="wicri:source">PMC</idno><idno type="pmid">21339861</idno><idno type="pmc">3039883</idno><idno type="url">http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3039883</idno><idno type="RBID">PMC:3039883</idno><idno type="doi">10.1111/j.1551-6709.2009.01077.x</idno><date when="2010">2010</date><idno type="wicri:Area/Pmc/Corpus">000F77</idno><idno type="wicri:Area/Pmc/Curation">000F77</idno><idno type="wicri:Area/Pmc/Checkpoint">001F27</idno><idno type="wicri:Area/Ncbi/Merge">001924</idno></publicationStmt><sourceDesc><biblStruct><analytic><title xml:lang="en" level="a" type="main">Cue integration with categories: Weighting acoustic cues in speech using unsupervised learning and distributional statistics</title><author><name sortKey="Toscano, Joseph C" sort="Toscano, Joseph C" uniqKey="Toscano J" first="Joseph C." last="Toscano">Joseph C. Toscano</name></author><author><name sortKey="Mcmurray, Bob" sort="Mcmurray, Bob" uniqKey="Mcmurray B" first="Bob" last="Mcmurray">Bob Mcmurray</name></author></analytic><series><title level="j">Cognitive science</title><idno type="ISSN">0364-0213</idno><idno type="eISSN">1551-6709</idno><imprint><date when="2010">2010</date></imprint></series></biblStruct></sourceDesc></fileDesc><profileDesc><textClass></textClass></profileDesc></teiHeader><front><div type="abstract" xml:lang="en"><p id="P1">During speech perception, listeners make judgments about the phonological category of sounds by taking advantage of multiple acoustic cues for each phonological contrast. Perceptual experiments have shown that listeners weight these cues differently. How do listeners weight and combine acoustic cues to arrive at an overall estimate of the category for a speech sound? Here, we present several simulations using mixture of Gaussians (MOG) models that learn cue weights and combine cues on the basis of their distributional statistics. We show that a cue-weighting metric in which cues receive weight as a function of their reliability at distinguishing the phonological categories provides a good fit to the perceptual data obtained from human listeners, but only when these weights emerge through the dynamics of learning. These results suggest that cue weights can be readily extracted from the speech signal through unsupervised learning processes.</p></div></front></TEI><pmc article-type="research-article" xml:lang="EN"><pmc-comment>The publisher of this article does not allow downloading of the full text in XML form.</pmc-comment>
<pmc-dir>properties manuscript</pmc-dir>
<front><journal-meta><journal-id journal-id-type="nlm-journal-id">7708195</journal-id><journal-id journal-id-type="pubmed-jr-id">34490</journal-id><journal-id journal-id-type="nlm-ta">Cogn Sci</journal-id><journal-title>Cognitive science</journal-title><issn pub-type="ppub">0364-0213</issn><issn pub-type="epub">1551-6709</issn></journal-meta><article-meta><article-id pub-id-type="pmid">21339861</article-id><article-id pub-id-type="pmc">3039883</article-id><article-id pub-id-type="doi">10.1111/j.1551-6709.2009.01077.x</article-id><article-id pub-id-type="manuscript">NIHMS268122</article-id><article-categories><subj-group subj-group-type="heading"><subject>Article</subject></subj-group></article-categories><title-group><article-title>Cue integration with categories: Weighting acoustic cues in speech using unsupervised learning and distributional statistics</article-title></title-group><contrib-group><contrib contrib-type="author"><name><surname>Toscano</surname><given-names>Joseph C.</given-names></name></contrib><contrib contrib-type="author"><name><surname>McMurray</surname><given-names>Bob</given-names></name></contrib><aff id="A1">Dept. of Psychology and Delta Center, University of Iowa</aff></contrib-group><author-notes><corresp id="cor1">Corresponding Author: Joseph Toscano, Dept. of Psychology, E11 SSH, University of Iowa, Iowa City, IA 52242, <email>joseph-toscano@uiowa.edu</email>, 319-335-0692 (voice), 319-335-0191 (fax)</corresp></author-notes><pub-date pub-type="nihms-submitted"><day>27</day><month>1</month><year>2011</year></pub-date><pub-date pub-type="ppub"><month>4</month><year>2010</year></pub-date><pub-date pub-type="pmc-release"><day>1</day><month>4</month><year>2011</year></pub-date><volume>34</volume><issue>3</issue><fpage>434</fpage><lpage>464</lpage><abstract><p id="P1">During speech perception, listeners make judgments about the phonological category of sounds by taking advantage of multiple acoustic cues for each phonological contrast. Perceptual experiments have shown that listeners weight these cues differently. How do listeners weight and combine acoustic cues to arrive at an overall estimate of the category for a speech sound? Here, we present several simulations using mixture of Gaussians (MOG) models that learn cue weights and combine cues on the basis of their distributional statistics. We show that a cue-weighting metric in which cues receive weight as a function of their reliability at distinguishing the phonological categories provides a good fit to the perceptual data obtained from human listeners, but only when these weights emerge through the dynamics of learning. These results suggest that cue weights can be readily extracted from the speech signal through unsupervised learning processes.</p></abstract><kwd-group><kwd>speech perception</kwd><kwd>speech development</kwd><kwd>cue weighting</kwd><kwd>reliability</kwd><kwd>categorization</kwd><kwd>statistical learning</kwd><kwd>unsupervised learning</kwd><kwd>mixture of Gaussians</kwd></kwd-group><contract-num rid="DC1">R01 DC008089-01A1
||DC</contract-num><contract-sponsor id="DC1">National Institute on Deafness and Other Communication Disorders : NIDCD</contract-sponsor></article-meta></front></pmc><affiliations><list></list><tree><noCountry><name sortKey="Mcmurray, Bob" sort="Mcmurray, Bob" uniqKey="Mcmurray B" first="Bob" last="Mcmurray">Bob Mcmurray</name><name sortKey="Toscano, Joseph C" sort="Toscano, Joseph C" uniqKey="Toscano J" first="Joseph C." last="Toscano">Joseph C. Toscano</name></noCountry></tree></affiliations></record>Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Ticri/CIDE/explor/HapticV1/Data/Ncbi/Merge
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 001924 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Ncbi/Merge/biblio.hfd -nk 001924 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien
|wiki= Ticri/CIDE
|area= HapticV1
|flux= Ncbi
|étape= Merge
|type= RBID
|clé= PMC:3039883
|texte= Cue integration with categories: Weighting acoustic cues in speech using unsupervised learning and distributional statistics
}}
Pour générer des pages wiki
HfdIndexSelect -h $EXPLOR_AREA/Data/Ncbi/Merge/RBID.i -Sk "pubmed:21339861" \
| HfdSelect -Kh $EXPLOR_AREA/Data/Ncbi/Merge/biblio.hfd \
| NlmPubMed2Wicri -a HapticV1
| This area was generated with Dilib version V0.6.23. |  |