000148271 001__ 148271
000148271 005__ 20250117145309.0
000148271 0247_ $$2doi$$a10.1109/LSP.2021.3084501
000148271 0248_ $$2sideral$$a126436
000148271 037__ $$aART-2021-126436
000148271 041__ $$aeng
000148271 100__ $$0(orcid)0000-0002-3142-0708$$aGimeno, P$$uUniversidad de Zaragoza
000148271 245__ $$aGeneralizing AUC Optimization to Multiclass Classification for Audio Segmentation With Limited Training Data
000148271 260__ $$c2021
000148271 5060_ $$aAccess copy available to the general public$$fUnrestricted
000148271 5203_ $$aArea under the ROC curve (AUC) optimisation techniques developed for neural networks have recently demonstrated their capabilities in different audio and speech related tasks. However, due to its intrinsic nature, AUC optimisation has focused only on binary tasks so far. In this paper, we introduce an extension to the AUC optimisation framework so that it can be easily applied to an arbitrary number of classes, aiming to overcome the issues derived from training data limitations in deep learning solutions. Building upon the multiclass definitions of the AUC metric found in the literature, we define two new training objectives using a one-versus-one and a one-versus-rest approach. In order to demonstrate its potential, we apply them in an audio segmentation task with limited training data that aims to differentiate 3 classes: foreground music, background music and no music. Experimental results show that our proposal can improve the performance of audio segmentation systems significantly compared to traditional training criteria such as cross entropy.
000148271 536__ $$9info:eu-repo/grantAgreement/ES/DGA/T36-20R$$9info:eu-repo/grantAgreement/EC/H2020/101007666/EU/Exchanges for SPEech ReseArch aNd TechnOlogies/ESPERANTO$$9This project has received funding from the European Union’s Horizon 2020 research and innovation program under grant agreement No H2020 101007666-ESPERANTO$$9info:eu-repo/grantAgreement/ES/MINECO/TIN2017-85854-C4-1-R
000148271 540__ $$9info:eu-repo/semantics/openAccess$$aAll rights reserved$$uhttp://www.europeana.eu/rights/rr-f/
000148271 590__ $$a3.201$$b2021
000148271 591__ $$aENGINEERING, ELECTRICAL & ELECTRONIC$$b113 / 274 = 0.412$$c2021$$dQ2$$eT2
000148271 592__ $$a1.361$$b2021
000148271 593__ $$aElectrical and Electronic Engineering$$c2021$$dQ1
000148271 593__ $$aApplied Mathematics$$c2021$$dQ1
000148271 594__ $$a6.6$$b2021
000148271 655_4 $$ainfo:eu-repo/semantics/article$$vinfo:eu-repo/semantics/acceptedVersion
000148271 700__ $$0(orcid)0000-0002-3505-0249$$aMingote, V$$uUniversidad de Zaragoza
000148271 700__ $$0(orcid)0000-0002-3886-7748$$aOrtega, A$$uUniversidad de Zaragoza
000148271 700__ $$0(orcid)0000-0001-5803-4316$$aMiguel, A$$uUniversidad de Zaragoza
000148271 700__ $$0(orcid)0000-0001-9137-4013$$aLleida, E$$uUniversidad de Zaragoza
000148271 7102_ $$15008$$2800$$aUniversidad de Zaragoza$$bDpto. Ingeniería Electrón.Com.$$cÁrea Teoría Señal y Comunicac.
000148271 773__ $$g28 (2021), 1135-1139$$pIEEE signal process. lett.$$tIEEE SIGNAL PROCESSING LETTERS$$x1070-9908
000148271 8564_ $$s429490$$uhttps://zaguan.unizar.es/record/148271/files/texto_completo.pdf$$yPostprint
000148271 8564_ $$s3234155$$uhttps://zaguan.unizar.es/record/148271/files/texto_completo.jpg?subformat=icon$$xicon$$yPostprint
000148271 909CO $$ooai:zaguan.unizar.es:148271$$particulos$$pdriver
000148271 951__ $$a2025-01-17-14:45:03
000148271 980__ $$aARTICLE