Automatic paragraph segmentation with lexical and prosodic features

Mostra el registre complet Registre parcial de l'ítem

• dc.contributor.author Lai, Catherineca
• dc.contributor.author Farrús, Mireiaca
• dc.contributor.author Moore, Johanna D.ca
• dc.date.accessioned 2017-02-07T08:31:38Z
• dc.date.available 2017-02-07T08:31:38Z
• dc.date.issued 2016ca
• dc.description Comunicació presentada a la Interspeech 2016, celebrada per la International Speech Communication Association (ISCA) els dies 8 a 12 de septembre de 2016 a San Francisco (EUA).
• dc.description.abstract As long-form spoken documents become more ubiquitous in everyday life, so does the need for automatic discourse segmentation in spoken language processing tasks. Although previous work has focused on broad topic segmentation, detection of finer-grained discourse units, such as paragraphs, is highly desirable for presenting and analyzing spoken content. To better understand how different aspects of speech cue these subtle discourse transitions, we investigate automatic paragraph segmentation of TED talks. We build lexical and prosodic paragraph segmenters using Support Vector Machines, AdaBoost, and Long Short Term Memory (LSTM) recurrent neural networks. In general, we find that induced cue words and supra-sentential prosodic features outperform features based on topical coherence, syntactic form and complexity. However, our best performance is achieved by combining a wide range of individually weak lexical and prosodic features, with the sequence modelling LSTM generally outperforming the other classifiers by a large margin. Moreover, we find that models that allow lower level interactions between different feature types produce better results than treating lexical and prosodic contributions as separate, independent information sources.
• dc.description.sponsorship The second author is funded from the EU’s Horizon 2020 Research and Innovation Programme under the GA H2020-RIA-645012 and the Spanish Ministry of Economy and Competitivity Juan de la Cierva program.
• dc.format.mimetype application/pdfca
• dc.identifier.citation Lai C, Farrús M, Moore JD. Automatic paragraph segmentation with lexical and prosodic features. In: Interspeech 2016; 2016 Sep 08-12; San Francisco (CA). [place unknown]: ISCA; 2016. p. 1034-8. DOI: 10.21437/Interspeech.2016-992ca
• dc.identifier.doi http://dx.doi.org/10.21437/Interspeech.2016-992
• dc.identifier.issn 1990-9772ca
• dc.identifier.uri http://hdl.handle.net/10230/28063
• dc.language.iso engca
• dc.publisher International Speech Communication Association (ISCA)ca
• dc.relation.ispartof Interspeech 2016; 2016 Sep 08-12; San Francisco (CA). [place unknown]: ISCA; 2016. p. 1034-8.
• dc.relation.projectID info:eu-repo/grantAgreement/EC/H2020/645012ca
• dc.rights © 2016 ISCAca
• dc.rights.accessRights info:eu-repo/semantics/openAccessca
• dc.subject.keyword Prosodyen
• dc.subject.keyword Discourseen
• dc.subject.keyword Segmentationen
• dc.subject.keyword Paragraphen
• dc.subject.keyword Coherenceen
• dc.title Automatic paragraph segmentation with lexical and prosodic featuresca
• dc.type info:eu-repo/semantics/conferenceObjectca
• dc.type.version info:eu-repo/semantics/publishedVersionca