Matches in SemOpenAlex for { <https://semopenalex.org/work/W3199016780> ?p ?o ?g. }
Showing items 1 to 82 of
82
with 100 items per page.
- W3199016780 abstract "The Recurrent Neural Network Transducer (RNN-T) extends Connectionist Temporal Classification (CTC) by jointly modeling both input-output and output-output dependencies, and it has been successfully applied in end-to-end speech recognition. RNN-T adds language information by expanding its dimensions in this way, but the model training is difficult. And some paths of the model are unreasonable when decoding. Therefore, we present the Alignment RNN Transducer (ART) algorithm, which uses the forward-backward algorithm in CTC to find the best path alignment information as context-related information for training. We not only reduce the training dimensions but also adds context information to the model. And we still use CTC as the loss function to avoid decoding the impossible path in the RNN-T. We verify the algorithm on a Mandarin speech corpus AIShell-1, and it achieves a 13.65% CER, compared with the 16.79% CER given by the RNN-T model." @default.
- W3199016780 created "2021-09-27" @default.
- W3199016780 creator A5017058806 @default.
- W3199016780 creator A5017353282 @default.
- W3199016780 creator A5020496465 @default.
- W3199016780 creator A5037611974 @default.
- W3199016780 creator A5063151866 @default.
- W3199016780 creator A5074244244 @default.
- W3199016780 date "2021-07-18" @default.
- W3199016780 modified "2023-10-15" @default.
- W3199016780 title "End-to-end speech recognition with Alignment RNN-Transducer" @default.
- W3199016780 cites W1979001355 @default.
- W3199016780 cites W2127141656 @default.
- W3199016780 cites W2293858598 @default.
- W3199016780 cites W2515439472 @default.
- W3199016780 cites W2750499125 @default.
- W3199016780 cites W2761434131 @default.
- W3199016780 cites W2911544293 @default.
- W3199016780 cites W2949975180 @default.
- W3199016780 cites W2963242190 @default.
- W3199016780 cites W2963882470 @default.
- W3199016780 cites W2972470300 @default.
- W3199016780 cites W2973122799 @default.
- W3199016780 cites W4239447739 @default.
- W3199016780 cites W4253573210 @default.
- W3199016780 doi "https://doi.org/10.1109/ijcnn52387.2021.9533348" @default.
- W3199016780 hasPublicationYear "2021" @default.
- W3199016780 type Work @default.
- W3199016780 sameAs 3199016780 @default.
- W3199016780 citedByCount "0" @default.
- W3199016780 crossrefType "proceedings-article" @default.
- W3199016780 hasAuthorship W3199016780A5017058806 @default.
- W3199016780 hasAuthorship W3199016780A5017353282 @default.
- W3199016780 hasAuthorship W3199016780A5020496465 @default.
- W3199016780 hasAuthorship W3199016780A5037611974 @default.
- W3199016780 hasAuthorship W3199016780A5063151866 @default.
- W3199016780 hasAuthorship W3199016780A5074244244 @default.
- W3199016780 hasConcept C11413529 @default.
- W3199016780 hasConcept C137293760 @default.
- W3199016780 hasConcept C147168706 @default.
- W3199016780 hasConcept C151730666 @default.
- W3199016780 hasConcept C154945302 @default.
- W3199016780 hasConcept C199360897 @default.
- W3199016780 hasConcept C2777735758 @default.
- W3199016780 hasConcept C2779343474 @default.
- W3199016780 hasConcept C28490314 @default.
- W3199016780 hasConcept C41008148 @default.
- W3199016780 hasConcept C50644808 @default.
- W3199016780 hasConcept C57273362 @default.
- W3199016780 hasConcept C8521452 @default.
- W3199016780 hasConcept C86803240 @default.
- W3199016780 hasConceptScore W3199016780C11413529 @default.
- W3199016780 hasConceptScore W3199016780C137293760 @default.
- W3199016780 hasConceptScore W3199016780C147168706 @default.
- W3199016780 hasConceptScore W3199016780C151730666 @default.
- W3199016780 hasConceptScore W3199016780C154945302 @default.
- W3199016780 hasConceptScore W3199016780C199360897 @default.
- W3199016780 hasConceptScore W3199016780C2777735758 @default.
- W3199016780 hasConceptScore W3199016780C2779343474 @default.
- W3199016780 hasConceptScore W3199016780C28490314 @default.
- W3199016780 hasConceptScore W3199016780C41008148 @default.
- W3199016780 hasConceptScore W3199016780C50644808 @default.
- W3199016780 hasConceptScore W3199016780C57273362 @default.
- W3199016780 hasConceptScore W3199016780C8521452 @default.
- W3199016780 hasConceptScore W3199016780C86803240 @default.
- W3199016780 hasLocation W31990167801 @default.
- W3199016780 hasOpenAccess W3199016780 @default.
- W3199016780 hasPrimaryLocation W31990167801 @default.
- W3199016780 hasRelatedWork W2263232528 @default.
- W3199016780 hasRelatedWork W2407648438 @default.
- W3199016780 hasRelatedWork W2782005958 @default.
- W3199016780 hasRelatedWork W2888131732 @default.
- W3199016780 hasRelatedWork W2944322787 @default.
- W3199016780 hasRelatedWork W2944384275 @default.
- W3199016780 hasRelatedWork W2978673814 @default.
- W3199016780 hasRelatedWork W2996122240 @default.
- W3199016780 hasRelatedWork W3112480982 @default.
- W3199016780 hasRelatedWork W3199016780 @default.
- W3199016780 isParatext "false" @default.
- W3199016780 isRetracted "false" @default.
- W3199016780 magId "3199016780" @default.
- W3199016780 workType "article" @default.