000112775 001__ 112775
000112775 005__ 20240319080954.0
000112775 0247_ $$2doi$$a10.3390/app12041889
000112775 0248_ $$2sideral$$a128058
000112775 037__ $$aART-2022-128058
000112775 041__ $$aeng
000112775 100__ $$aÁlvarez, A.
000112775 245__ $$aEvaluating Novel Speech Transcription Architectures on the Spanish RTVE2020 Database
000112775 260__ $$c2022
000112775 5060_ $$aAccess copy available to the general public$$fUnrestricted
000112775 5203_ $$aThis work presents three novel speech recognition architectures evaluated on the Spanish RTVE2020 dataset, employed as the main evaluation set in the Albayzín S2T Transcription Challenge 2020. The main objective was to improve the performance of the systems previously submitted by the authors to the challenge, in which the primary system scored the second position. The novel systems are based on both DNN-HMM and E2E acoustic models, for which fully-and self-supervised learning methods were included. As a result, the new speech recognition engines clearly outper-formed the performance of the initial systems from the previous best WER of 19.27 to the new best of 17.60 achieved by the DNN-HMM based system. This work therefore describes an interesting benchmark of the latest acoustic models over a highly challenging dataset, and identifies the most optimal ones depending on the expected quality, the available resources and the required latency.
000112775 540__ $$9info:eu-repo/semantics/openAccess$$aby$$uhttp://creativecommons.org/licenses/by/3.0/es/
000112775 590__ $$a2.7$$b2022
000112775 592__ $$a0.492$$b2022
000112775 591__ $$aPHYSICS, APPLIED$$b78 / 160 = 0.488$$c2022$$dQ2$$eT2
000112775 591__ $$aENGINEERING, MULTIDISCIPLINARY$$b42 / 90 = 0.467$$c2022$$dQ2$$eT2
000112775 591__ $$aCHEMISTRY, MULTIDISCIPLINARY$$b100 / 178 = 0.562$$c2022$$dQ3$$eT2
000112775 591__ $$aMATERIALS SCIENCE, MULTIDISCIPLINARY$$b208 / 343 = 0.606$$c2022$$dQ3$$eT2
000112775 593__ $$aFluid Flow and Transfer Processes$$c2022$$dQ2
000112775 593__ $$aMaterials Science (miscellaneous)$$c2022$$dQ2
000112775 593__ $$aEngineering (miscellaneous)$$c2022$$dQ2
000112775 593__ $$aInstrumentation$$c2022$$dQ2
000112775 593__ $$aProcess Chemistry and Technology$$c2022$$dQ3
000112775 593__ $$aComputer Science Applications$$c2022$$dQ3
000112775 594__ $$a4.5$$b2022
000112775 655_4 $$ainfo:eu-repo/semantics/article$$vinfo:eu-repo/semantics/publishedVersion
000112775 700__ $$aArzelus, H.
000112775 700__ $$aTorre, I.G.
000112775 700__ $$aGonzález-Docasal, A.
000112775 773__ $$g12, 4 (2022), 1889 [16 pp.]$$pAppl. sci.$$tApplied Sciences (Switzerland)$$x2076-3417
000112775 8564_ $$s1035168$$uhttps://zaguan.unizar.es/record/112775/files/texto_completo.pdf$$yVersión publicada
000112775 8564_ $$s2765422$$uhttps://zaguan.unizar.es/record/112775/files/texto_completo.jpg?subformat=icon$$xicon$$yVersión publicada
000112775 909CO $$ooai:zaguan.unizar.es:112775$$particulos$$pdriver
000112775 951__ $$a2024-03-18-13:24:59
000112775 980__ $$aARTICLE