Transfer learning with audioSet to voice pathologies identification in continuous speech

Guedes, Victor; Teixeira, Felipe; Oliveira, Alessa Anjos de; Fernandes, Joana Filipa Teixeira; Silva, Letícia; Candido Junior, Arnaldo; Teixeira, João Paulo

Publication

Transfer learning with audioSet to voice pathologies identification in continuous speech

2019Conference paper

dc.contributor.author	Guedes, Victor
dc.contributor.author	Teixeira, Felipe
dc.contributor.author	Oliveira, Alessa Anjos de
dc.contributor.author	Fernandes, Joana Filipa Teixeira
dc.contributor.author	Silva, Letícia
dc.contributor.author	Candido Junior, Arnaldo
dc.contributor.author	Teixeira, João Paulo
dc.date.accessioned	2020-04-23T09:46:13Z
dc.date.available	2020-04-23T09:46:13Z
dc.date.issued	2019
dc.description.abstract	The classification of pathological diseases with the implementation of concepts of Deep Learning has been increasing considerably in recent times. Among the works developed there are good results for the classification in sustained speech with vowels, but few related works for the classification in continuous speech. This work uses the German Saarbrücken Voice Database with the phrase “Guten Morgen, wie geht es Ihnen?” to classify four classes: dysphonia, laryngitis, paralysis of vocal cords and healthy voices. Transfer learning concepts were used with the AudioSet database. Two models were developed based on Long-Short-Term-Memory and Convolutional Network for classification of extracted embeddings and comparison of the best results, using cross-validation. The final results allowed to obtaining 40% of f1-score for the four classes, 66% f1-score for Dysphonia x Healthy, 67% for Laryngitis x healthy and 80% for Paralysis x Healthy.	pt_PT
dc.description.version	info:eu-repo/semantics/publishedVersion	pt_PT
dc.identifier.citation	Guedes, Victor; Teixeira, Felipe; Oliveira, Alessa; Fernandes, Joana; Silva, Leticia; Junior, Arnaldo; Teixeira, João Paulo (2019). Transfer learning with audioSet to voice pathologies identification in continuous speech. In International Conference on ENTERprise Information Systems, International Conference on Project MANagement. Tunisia. 164, p. 662-669	pt_PT
dc.identifier.doi	10.1016/j.procs.2019.12.233	pt_PT
dc.identifier.uri	http://hdl.handle.net/10198/21796
dc.language.iso	eng	pt_PT
dc.peerreviewed	yes	pt_PT
dc.publisher	Elsevier	pt_PT
dc.rights.uri	http://creativecommons.org/licenses/by/4.0/	pt_PT
dc.subject	Long short term memory	pt_PT
dc.subject	Convolutional neural network	pt_PT
dc.subject	SVD	pt_PT
dc.subject	Deep learning	pt_PT
dc.subject	Voice pathologies diagnose	pt_PT
dc.title	Transfer learning with audioSet to voice pathologies identification in continuous speech	pt_PT
dc.type	conference paper
dspace.entity.type	Publication
oaire.citation.conferencePlace	Tunisia	pt_PT
oaire.citation.endPage	669	pt_PT
oaire.citation.startPage	662	pt_PT
oaire.citation.title	International Conference on ENTERprise Information Systems, International Conference on Project MANagement	pt_PT
oaire.citation.volume	164	pt_PT
person.familyName	Teixeira
person.familyName	Fernandes
person.familyName	Silva
person.familyName	Teixeira
person.givenName	Felipe
person.givenName	Joana Filipa Teixeira
person.givenName	Letícia
person.givenName	João Paulo
person.identifier	ABC-9055-2020
person.identifier	663194
person.identifier.ciencia-id	0E17-62FB-AA17
person.identifier.ciencia-id	AE12-440A-299D
person.identifier.ciencia-id	C01E-87BA-67D7
person.identifier.ciencia-id	4F15-B322-59B4
person.identifier.orcid	0000-0002-0618-4627
person.identifier.orcid	0000-0003-3812-2794
person.identifier.orcid	0000-0002-6679-5702
person.identifier.rid	N-6576-2013
person.identifier.scopus-author-id	57069567500
rcaap.rights	openAccess	pt_PT
rcaap.type	conferenceObject	pt_PT
relation.isAuthorOfPublication	764c5209-b9ab-479e-b5be-59fbe07c784b
relation.isAuthorOfPublication	a6f7a119-fbc9-439f-8dd9-0bbc9ec82fad
relation.isAuthorOfPublication	a2aa1be8-574c-4e0b-afd6-d3c61efad820
relation.isAuthorOfPublication	33f4af65-7ddf-46f0-8b44-a7470a8ba2bf
relation.isAuthorOfPublication.latestForDiscovery	33f4af65-7ddf-46f0-8b44-a7470a8ba2bf