84697 20230914083238.0 doi 10.3390/app9183697 sideral 114013 ART-2019-114013 eng Viñals, Ignacio Universidad de Zaragoza (orcid)0000-0003-1772-0605 An analysis of the short utterance problem for speaker characterization 2019 Access copy available to the general public Unrestricted Speaker characterization has always been conditioned by the length of the evaluated utterances. Despite performing well with large amounts of audio, significant degradations in performance are obtained when short utterances are considered. In this work we present an analysis of the short utterance problem providing an alternative point of view. From our perspective the performance in the evaluation of short utterances is highly influenced by the phonetic similarity between enrollment and test utterances. Both enrollment and test should contain similar phonemes to properly discriminate, being degraded otherwise. In this study we also interpret short utterances as incomplete long utterances where some acoustic units are either unbalanced or just missing. These missing units are responsible for the speaker representations to be unreliable. These unreliable representations are biased with respect to the reference counterparts, obtained from long utterances. These undesired shifts increase the intra-speaker variability, causing a significant loss of performance. According to our experiments, short utterances (3-60 s) can perform as accurate as if long utterances were involved by just reassuring the phonetic distributions. This analysis is determined by the current embedding extraction approach, based on the accumulation of local short-time information. Thus it is applicable to most of the state-of-the-art embeddings, including traditional i-vectors and Deep Neural Network (DNN) xvectors. info:eu-repo/grantAgreement/ES/DGA-FEDER/T36-17R info:eu-repo/grantAgreement/ES/DGA-FEDER/2014-2020 info:eu-repo/grantAgreement/ES/MINECO/TIN2017-85854-C4-1-R info:eu-repo/semantics/openAccess by http://creativecommons.org/licenses/by/3.0/es/ 2.474 2019 PHYSICS, APPLIED 62 / 154 = 0.403 2019 Q2 T2 ENGINEERING, MULTIDISCIPLINARY 32 / 91 = 0.352 2019 Q2 T2 CHEMISTRY, MULTIDISCIPLINARY 88 / 176 = 0.5 2019 Q2 T2 MATERIALS SCIENCE, MULTIDISCIPLINARY 161 / 314 = 0.513 2019 Q3 T2 0.418 2019 Engineering (miscellaneous) 2019 Q1 Fluid Flow and Transfer Processes 2019 Q2 Process Chemistry and Technology 2019 Q2 Instrumentation 2019 Q2 Materials Science (miscellaneous) 2019 Q2 Computer Science Applications 2019 Q3 info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion Ortega, Alfonso Universidad de Zaragoza (orcid)0000-0002-3886-7748 Miguel, Antonio Universidad de Zaragoza (orcid)0000-0001-5803-4316 Lleida, Eduardo Universidad de Zaragoza (orcid)0000-0001-9137-4013 5008 800 Universidad de Zaragoza Dpto. Ingeniería Electrón.Com. Área Teoría Señal y Comunicac. 9, 18 (2019), 3697 [19 pp.] Appl. sci. Applied Sciences (Switzerland) 2076-3417 437566 http://zaguan.unizar.es/record/84697/files/texto_completo.pdf Versión publicada 109124 http://zaguan.unizar.es/record/84697/files/texto_completo.jpg?subformat=icon icon Versión publicada oai:zaguan.unizar.es:84697 articulos driver 2023-09-13-10:47:12 ARTICLE