Publication:
A saliency-based attention LSTM model for cognitive load classification from speech

Simple item page
dc.affiliation.dptoUC3M. Departamento de Teoría de la Señal y Comunicacioneses
dc.affiliation.grupoinvUC3M. Grupo de Investigación: Procesado Multimediaes
dc.contributor.authorGallardo Antolín, Ascensión
dc.contributor.authorMontero Martínez, Juan Manuel
dc.contributor.funderMinisterio de Economía y Competitividad (España)es
dc.date.accessioned2020-12-22T12:49:53Z
dc.date.available2020-12-22T12:49:53Z
dc.date.issued2019
dc.descriptionProceeding of: Interspeech 2019, 20th Annual Conference of the International Speech Communication Association, Graz, Austria, 15-19 September 2019en
dc.description.abstractCognitive Load (CL) refers to the amount of mental demand that a given task imposes on an individual's cognitive system and it can affect his/her productivity in very high load situations. In this paper, we propose an automatic system capable of classifying the CL level of a speaker by analyzing his/her voice. Our research on this topic goes into two main directions. In the first one, we focus on the use of Long Short-Term Memory (LSTM) networks with different weighted pooling strategies for CL level classification. In the second contribution, for overcoming the need of a large amount of training data, we propose a novel attention mechanism that uses the Kalinli's auditory saliency model. Experiments show that our proposal outperforms significantly both, a baseline system based on Support Vector Machines (SVM) and a LSTM-based system with logistic regression attention model.en
dc.description.sponsorshipThe work leading to these results has been partly supported by Spanish Government grants TEC2017-84395-P and TEC2017-84593-C2-1-R.en
dc.format.extent5es
dc.identifier.bibliographicCitationInterspeech 2019, 20th Annual Conference of the International Speech Communication Association, Graz, Austria, 15-19 September 2019 [Proceedings]. ISCA, 2019, Pp. 216-220en
dc.identifier.doihttps://doi.org/10.21437/Interspeech.2019-1603
dc.identifier.publicationfirstpage216es
dc.identifier.publicationlastpage220es
dc.identifier.publicationtitleInterspeech 2019, 20th Annual Conference of the International Speech Communication Association, Graz, Austria, 15-19 September 2019 [Proceedings]en
dc.identifier.urihttps://hdl.handle.net/10016/31660
dc.identifier.uxxiCC/0000030601
dc.language.isoengen
dc.publisherInternational Speech Communication Association (ISCA)en
dc.relation.eventdate2019-09-15es
dc.relation.eventplaceGraz, AUSTRIAen
dc.relation.eventtitle20th Annual Conference of the International Speech Communication Association (INTERSPEECH 2019)en
dc.relation.projectIDGobierno de España. TEC2017-84395-Pes
dc.relation.projectIDGobierno de España. TEC2017-84593-C2-1-Res
dc.rights© 2019 ISCAen
dc.rights.accessRightsopen accesses
dc.subject.ecienciaElectrónicaes
dc.subject.ecienciaTelecomunicacioneses
dc.subject.otherCognitive loaden
dc.subject.otherSpeechen
dc.subject.otherLSTMen
dc.subject.otherWeigthed poolingen
dc.subject.otherAuditory saliencyen
dc.subject.otherAttention modelen
dc.titleA saliency-based attention LSTM model for cognitive load classification from speechen
dc.typeconference paper*
dc.type.hasVersionVoR*
dspace.entity.typePublication
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
saliency_INTERSPEECH_2019.pdf
Size:
289.22 KB
Format:
Adobe Portable Document Format
Download
Collections
DTSC - GPM - Comunicaciones en congresos y otros eventos
DTSC - GPM - Capítulos de Monografías