000071177 001__ 71177
000071177 005__ 20210208180159.0
000071177 0247_ $$2doi$$a10.21437/Interspeech.2017-1314
000071177 0248_ $$2sideral$$a104677
000071177 037__ $$aART-2017-104677
000071177 041__ $$aeng
000071177 100__ $$0(orcid)0000-0001-5803-4316$$aMiguel, A.$$uUniversidad de Zaragoza
000071177 245__ $$aTied hidden factors in neural networks for end-To-end speaker recognition
000071177 260__ $$c2017
000071177 5060_ $$aAccess copy available to the general public$$fUnrestricted
000071177 5203_ $$aIn this paper we propose a method to model speaker and session variability and able to generate likelihood ratios using neural networks in an end-To-end phrase dependent speaker verification system. As in Joint Factor Analysis, the model uses tied hidden variables to model speaker and session variability and a MAP adaptation of some of the parameters of the model. In the training procedure our method jointly estimates the network parameters and the values of the speaker and channel hidden variables. This is done in a two-step backpropagation algorithm, first the network weights and factor loading matrices are updated and then the hidden variables, whose gradients are calculated by aggregating the corresponding speaker or session frames, since these hidden variables are tied. The last layer of the network is defined as a linear regression probabilistic model whose inputs are the previous layer outputs. This choice has the advantage that it produces likelihoods and additionally it can be adapted during the enrolment using MAP without the need of a gradient optimization. The decisions are made based on the ratio of the output likelihoods of two neural network models, speaker adapted and universal background model. The method was evaluated on the RSR2015 database.
000071177 536__ $$9info:eu-repo/grantAgreement/ES/MINEC0/TIN2014-54288-C4-2-R$$9info:eu-repo/grantAgreement/EC/FP7/610986/EU/IRIS: Towards Natural Interaction and Communication/IRIS
000071177 540__ $$9info:eu-repo/semantics/openAccess$$aAll rights reserved$$uhttp://www.europeana.eu/rights/rr-f/
000071177 655_4 $$ainfo:eu-repo/semantics/article$$vinfo:eu-repo/semantics/acceptedVersion
000071177 700__ $$0(orcid)0000-0002-9407-5817$$aLlombart, J.$$uUniversidad de Zaragoza
000071177 700__ $$0(orcid)0000-0002-3886-7748$$aOrtega, A.$$uUniversidad de Zaragoza
000071177 700__ $$0(orcid)0000-0001-9137-4013$$aLleida, E.$$uUniversidad de Zaragoza
000071177 7102_ $$15008$$2800$$aUniversidad de Zaragoza$$bDpto. Ingeniería Electrón.Com.$$cÁrea Teoría Señal y Comunicac.
000071177 773__ $$g2017-August (2017), 2819-2823$$pInterspeech (USB)$$tInterspeech (USB)$$x2308-457X
000071177 85641 $$uhttps://www.researchgate.net/profile/Antonio_Miguel4/publication/319185138_Tied_Hidden_Factors_in_Neural_Networks_for_End-to-End_Speaker_Recognition/links/59a43b2ea6fdcc773a3736b0/Tied-Hidden-Factors-in-Neural-Networks-for-End-to-End-Speaker-Recognition.pdf$$zTexto completo de la revista
000071177 8564_ $$s265718$$uhttps://zaguan.unizar.es/record/71177/files/texto_completo.pdf$$yPostprint
000071177 8564_ $$s126660$$uhttps://zaguan.unizar.es/record/71177/files/texto_completo.jpg?subformat=icon$$xicon$$yPostprint
000071177 909CO $$ooai:zaguan.unizar.es:71177$$particulos$$pdriver
000071177 951__ $$a2021-02-08-17:42:18
000071177 980__ $$aARTICLE