dc.contributor.author | Weninger, Felix | de |
dc.contributor.author | Wagner, Claudia | de |
dc.contributor.author | Wöllmer, Martin | de |
dc.contributor.author | Schuller, Björn | de |
dc.contributor.author | Morency, Louis-Philipp | de |
dc.date.accessioned | 2020-01-15T14:13:38Z | |
dc.date.available | 2020-01-15T14:13:38Z | |
dc.date.issued | 2013 | de |
dc.identifier.isbn | 978-1-4799-0356-6 | de |
dc.identifier.issn | 2379-190X | de |
dc.identifier.uri | https://www.ssoar.info/ssoar/handle/document/66084 | |
dc.description.abstract | We present a multi-modal approach to speaker characterization using
acoustic, visual and linguistic features. Full realism is provided by
evaluation on a database of real-life web videos and automatic feature
extraction including face and eye detection, and automatic speech
recognition. Different segmentations are evaluated for the audio and
video streams, and the statistical relevance of Linguistic Inquiry and
Word Count (LIWC) features is confirmed. In the result, late multimodal
fusion delivers 73, 92 and 73% average recall in binary age,
gender and race classification on unseen test subjects, outperforming
the best single modalities for age and race. | de |
dc.language | en | de |
dc.publisher | IEEE | de |
dc.subject.ddc | Naturwissenschaften | de |
dc.subject.ddc | Science | en |
dc.subject.other | speaker classification; computational paralinguistics; multi-modal fusion; Linguistic Inquiry and Word Count; LIWC | de |
dc.title | Speaker trait characterization in web videos: Uniting speech, language, and facial features | de |
dc.description.review | begutachtet (peer reviewed) | de |
dc.description.review | peer reviewed | en |
dc.source.collection | Proceedings of the 38th International Conference on Acoustics, Speech and Signal Processing (ICASSP 2013) | de |
dc.publisher.country | USA | |
dc.subject.classoz | Naturwissenschaften, Technik(wissenschaften), angewandte Wissenschaften | de |
dc.subject.classoz | Natural Science and Engineering, Applied Sciences | en |
dc.subject.thesoz | Video | de |
dc.subject.thesoz | video | en |
dc.subject.thesoz | Video-Clip | de |
dc.subject.thesoz | video clip | en |
dc.subject.thesoz | Aufzeichnung | de |
dc.subject.thesoz | recording | en |
dc.subject.thesoz | Computerlinguistik | de |
dc.subject.thesoz | computational linguistics | en |
dc.subject.thesoz | Internet | de |
dc.subject.thesoz | Internet | en |
dc.subject.thesoz | Evaluation | de |
dc.subject.thesoz | evaluation | en |
dc.subject.thesoz | Soziale Medien | de |
dc.subject.thesoz | social media | en |
dc.subject.thesoz | Experiment | de |
dc.subject.thesoz | experiment | en |
dc.subject.thesoz | audiovisuelle Medien | de |
dc.subject.thesoz | audiovisual media | en |
dc.identifier.urn | urn:nbn:de:0168-ssoar-66084-2 | |
dc.rights.licence | Deposit Licence - Keine Weiterverbreitung, keine Bearbeitung | de |
dc.rights.licence | Deposit Licence - No Redistribution, No Modifications | en |
internal.status | noch nicht fertig erschlossen | de |
internal.identifier.thesoz | 10061598 | |
internal.identifier.thesoz | 10063356 | |
internal.identifier.thesoz | 10037027 | |
internal.identifier.thesoz | 10040387 | |
internal.identifier.thesoz | 10040528 | |
internal.identifier.thesoz | 10039188 | |
internal.identifier.thesoz | 10094228 | |
internal.identifier.thesoz | 10043015 | |
internal.identifier.thesoz | 10036934 | |
dc.type.stock | incollection | de |
dc.type.document | Konferenzbeitrag | de |
dc.type.document | conference paper | en |
dc.source.pageinfo | 3647-3651 | de |
internal.identifier.classoz | 50200 | |
internal.identifier.document | 16 | |
dc.source.conference | International Conference on Acoustics, Speech and Signal Processing (ICASSP 2013) | de |
dc.event.city | Vancouver | de |
internal.identifier.ddc | 500 | |
dc.identifier.doi | https://doi.org/10.1109/ICASSP.2013.6638338 | de |
dc.date.conference | 2013 | de |
dc.source.conferencenumber | 38 | de |
dc.description.pubstatus | Veröffentlichungsversion | de |
dc.description.pubstatus | Published Version | en |
internal.identifier.licence | 3 | |
internal.identifier.pubstatus | 1 | |
internal.identifier.review | 1 | |
internal.pdf.wellformed | false | |
internal.pdf.encrypted | false | |