Matches in SemOpenAlex for { <https://semopenalex.org/work/W3167946479> ?p ?o ?g. }
- W3167946479 abstract "In this paper, we demonstrate the efficacy of transfer learning and continuous learning for various automatic speech recognition (ASR) tasks using end-to-end models trained with CTC loss. We start with a large pre-trained English ASR model and show that transfer learning can be effectively and easily performed on: (1) different English accents, (2) different languages (from English to German, Spanish, Russian, or from Mandarin to Cantonese) and (3) application-specific domains. Our extensive set of experiments demonstrate that in all three cases, transfer learning from a good base model has higher accuracy than a model trained from scratch. Our results indicate that, for fine-tuning, larger pre-trained models are better than small pre-trained models, even if the dataset for fine-tuning is small. We also show that transfer learning significantly speeds up convergence, which could result in significant cost savings when training with large datasets." @default.
- W3167946479 created "2021-06-22" @default.
- W3167946479 creator A5001258739 @default.
- W3167946479 creator A5005325892 @default.
- W3167946479 creator A5016038454 @default.
- W3167946479 creator A5026088310 @default.
- W3167946479 creator A5032957280 @default.
- W3167946479 creator A5033043101 @default.
- W3167946479 creator A5036658158 @default.
- W3167946479 creator A5040747392 @default.
- W3167946479 creator A5048403564 @default.
- W3167946479 creator A5074472751 @default.
- W3167946479 creator A5076982760 @default.
- W3167946479 creator A5077957584 @default.
- W3167946479 creator A5080235829 @default.
- W3167946479 creator A5083696320 @default.
- W3167946479 creator A5084554132 @default.
- W3167946479 date "2021-07-05" @default.
- W3167946479 modified "2023-10-02" @default.
- W3167946479 title "Cross-Language Transfer Learning and Domain Adaptation for End-to-End Automatic Speech Recognition" @default.
- W3167946479 cites W1494198834 @default.
- W3167946479 cites W1682403713 @default.
- W3167946479 cites W1994606281 @default.
- W3167946479 cites W2024490156 @default.
- W3167946479 cites W2025198378 @default.
- W3167946479 cites W2033436836 @default.
- W3167946479 cites W2041823554 @default.
- W3167946479 cites W2127141656 @default.
- W3167946479 cites W2147768505 @default.
- W3167946479 cites W2154820175 @default.
- W3167946479 cites W2402146185 @default.
- W3167946479 cites W2604173493 @default.
- W3167946479 cites W2786459654 @default.
- W3167946479 cites W2963303951 @default.
- W3167946479 cites W2963522845 @default.
- W3167946479 cites W2973072704 @default.
- W3167946479 cites W3015537910 @default.
- W3167946479 cites W3150269179 @default.
- W3167946479 doi "https://doi.org/10.1109/icme51207.2021.9428334" @default.
- W3167946479 hasPublicationYear "2021" @default.
- W3167946479 type Work @default.
- W3167946479 sameAs 3167946479 @default.
- W3167946479 citedByCount "5" @default.
- W3167946479 countsByYear W31679464792021 @default.
- W3167946479 countsByYear W31679464792022 @default.
- W3167946479 countsByYear W31679464792023 @default.
- W3167946479 crossrefType "proceedings-article" @default.
- W3167946479 hasAuthorship W3167946479A5001258739 @default.
- W3167946479 hasAuthorship W3167946479A5005325892 @default.
- W3167946479 hasAuthorship W3167946479A5016038454 @default.
- W3167946479 hasAuthorship W3167946479A5026088310 @default.
- W3167946479 hasAuthorship W3167946479A5032957280 @default.
- W3167946479 hasAuthorship W3167946479A5033043101 @default.
- W3167946479 hasAuthorship W3167946479A5036658158 @default.
- W3167946479 hasAuthorship W3167946479A5040747392 @default.
- W3167946479 hasAuthorship W3167946479A5048403564 @default.
- W3167946479 hasAuthorship W3167946479A5074472751 @default.
- W3167946479 hasAuthorship W3167946479A5076982760 @default.
- W3167946479 hasAuthorship W3167946479A5077957584 @default.
- W3167946479 hasAuthorship W3167946479A5080235829 @default.
- W3167946479 hasAuthorship W3167946479A5083696320 @default.
- W3167946479 hasAuthorship W3167946479A5084554132 @default.
- W3167946479 hasConcept C120665830 @default.
- W3167946479 hasConcept C121332964 @default.
- W3167946479 hasConcept C137293760 @default.
- W3167946479 hasConcept C138885662 @default.
- W3167946479 hasConcept C138954614 @default.
- W3167946479 hasConcept C139807058 @default.
- W3167946479 hasConcept C150899416 @default.
- W3167946479 hasConcept C154775046 @default.
- W3167946479 hasConcept C154945302 @default.
- W3167946479 hasConcept C166957645 @default.
- W3167946479 hasConcept C173608175 @default.
- W3167946479 hasConcept C177264268 @default.
- W3167946479 hasConcept C199360897 @default.
- W3167946479 hasConcept C204321447 @default.
- W3167946479 hasConcept C2776175482 @default.
- W3167946479 hasConcept C2776434776 @default.
- W3167946479 hasConcept C28490314 @default.
- W3167946479 hasConcept C41008148 @default.
- W3167946479 hasConcept C41895202 @default.
- W3167946479 hasConcept C51632099 @default.
- W3167946479 hasConcept C95457728 @default.
- W3167946479 hasConcept C95623464 @default.
- W3167946479 hasConceptScore W3167946479C120665830 @default.
- W3167946479 hasConceptScore W3167946479C121332964 @default.
- W3167946479 hasConceptScore W3167946479C137293760 @default.
- W3167946479 hasConceptScore W3167946479C138885662 @default.
- W3167946479 hasConceptScore W3167946479C138954614 @default.
- W3167946479 hasConceptScore W3167946479C139807058 @default.
- W3167946479 hasConceptScore W3167946479C150899416 @default.
- W3167946479 hasConceptScore W3167946479C154775046 @default.
- W3167946479 hasConceptScore W3167946479C154945302 @default.
- W3167946479 hasConceptScore W3167946479C166957645 @default.
- W3167946479 hasConceptScore W3167946479C173608175 @default.
- W3167946479 hasConceptScore W3167946479C177264268 @default.
- W3167946479 hasConceptScore W3167946479C199360897 @default.
- W3167946479 hasConceptScore W3167946479C204321447 @default.
- W3167946479 hasConceptScore W3167946479C2776175482 @default.
- W3167946479 hasConceptScore W3167946479C2776434776 @default.