Matches in SemOpenAlex for { <https://semopenalex.org/work/W3141756469> ?p ?o ?g. }
Showing items 1 to 92 of
92
with 100 items per page.
- W3141756469 abstract "Sequence to Sequence models, in particular the Transformer, achieve state of the art results in Automatic Speech Recognition. Practical usage is however limited to cases where full utterance latency is acceptable. In this work we introduce Taris, a Transformer-based online speech recognition system aided by an auxiliary task of incremental word counting. We use the cumulative word sum to dynamically segment speech and enable its eager decoding into words. Experiments performed on the LRS2, LibriSpeech, and Aishell-1 datasets of English and Mandarin speech show that the online system performs comparable with the offline one when having a dynamic algorithmic delay of 5 segments. Furthermore, we show that the estimated segment length distribution resembles the word length distribution obtained with forced alignment, although our system does not require an exact segment-to-word equivalence. Taris introduces a negligible overhead compared to a standard Transformer, while the local relationship modelling between inputs and outputs grants invariance to sequence length by design." @default.
- W3141756469 created "2021-04-13" @default.
- W3141756469 creator A5034046565 @default.
- W3141756469 creator A5042231269 @default.
- W3141756469 creator A5042852762 @default.
- W3141756469 date "2021-01-19" @default.
- W3141756469 modified "2023-09-27" @default.
- W3141756469 title "Learning to Count Words in Fluent Speech Enables Online Speech Recognition" @default.
- W3141756469 cites W1494198834 @default.
- W3141756469 cites W1995562189 @default.
- W3141756469 cites W2038056950 @default.
- W3141756469 cites W2143612262 @default.
- W3141756469 cites W2157149948 @default.
- W3141756469 cites W2327501763 @default.
- W3141756469 cites W2741480356 @default.
- W3141756469 cites W2747874407 @default.
- W3141756469 cites W2889187401 @default.
- W3141756469 cites W2936123380 @default.
- W3141756469 cites W2949975180 @default.
- W3141756469 cites W2953190524 @default.
- W3141756469 cites W2962742956 @default.
- W3141756469 cites W2962824709 @default.
- W3141756469 cites W2963158258 @default.
- W3141756469 cites W2963242190 @default.
- W3141756469 cites W2963414781 @default.
- W3141756469 cites W2963827914 @default.
- W3141756469 cites W2972439411 @default.
- W3141756469 cites W2973215447 @default.
- W3141756469 cites W3008181812 @default.
- W3141756469 cites W3008898571 @default.
- W3141756469 cites W3011234510 @default.
- W3141756469 cites W3013690751 @default.
- W3141756469 cites W3016190221 @default.
- W3141756469 doi "https://doi.org/10.1109/slt48900.2021.9383563" @default.
- W3141756469 hasPublicationYear "2021" @default.
- W3141756469 type Work @default.
- W3141756469 sameAs 3141756469 @default.
- W3141756469 citedByCount "4" @default.
- W3141756469 countsByYear W31417564692020 @default.
- W3141756469 countsByYear W31417564692021 @default.
- W3141756469 countsByYear W31417564692022 @default.
- W3141756469 crossrefType "proceedings-article" @default.
- W3141756469 hasAuthorship W3141756469A5034046565 @default.
- W3141756469 hasAuthorship W3141756469A5042231269 @default.
- W3141756469 hasAuthorship W3141756469A5042852762 @default.
- W3141756469 hasBestOaLocation W31417564692 @default.
- W3141756469 hasConcept C11413529 @default.
- W3141756469 hasConcept C121332964 @default.
- W3141756469 hasConcept C154945302 @default.
- W3141756469 hasConcept C165801399 @default.
- W3141756469 hasConcept C204321447 @default.
- W3141756469 hasConcept C23224414 @default.
- W3141756469 hasConcept C2775852435 @default.
- W3141756469 hasConcept C28490314 @default.
- W3141756469 hasConcept C41008148 @default.
- W3141756469 hasConcept C57273362 @default.
- W3141756469 hasConcept C62520636 @default.
- W3141756469 hasConcept C66322947 @default.
- W3141756469 hasConcept C76155785 @default.
- W3141756469 hasConcept C82876162 @default.
- W3141756469 hasConceptScore W3141756469C11413529 @default.
- W3141756469 hasConceptScore W3141756469C121332964 @default.
- W3141756469 hasConceptScore W3141756469C154945302 @default.
- W3141756469 hasConceptScore W3141756469C165801399 @default.
- W3141756469 hasConceptScore W3141756469C204321447 @default.
- W3141756469 hasConceptScore W3141756469C23224414 @default.
- W3141756469 hasConceptScore W3141756469C2775852435 @default.
- W3141756469 hasConceptScore W3141756469C28490314 @default.
- W3141756469 hasConceptScore W3141756469C41008148 @default.
- W3141756469 hasConceptScore W3141756469C57273362 @default.
- W3141756469 hasConceptScore W3141756469C62520636 @default.
- W3141756469 hasConceptScore W3141756469C66322947 @default.
- W3141756469 hasConceptScore W3141756469C76155785 @default.
- W3141756469 hasConceptScore W3141756469C82876162 @default.
- W3141756469 hasLocation W31417564691 @default.
- W3141756469 hasLocation W31417564692 @default.
- W3141756469 hasOpenAccess W3141756469 @default.
- W3141756469 hasPrimaryLocation W31417564691 @default.
- W3141756469 hasRelatedWork W1966737826 @default.
- W3141756469 hasRelatedWork W2012060148 @default.
- W3141756469 hasRelatedWork W2058877998 @default.
- W3141756469 hasRelatedWork W2137833003 @default.
- W3141756469 hasRelatedWork W2225817525 @default.
- W3141756469 hasRelatedWork W2540351954 @default.
- W3141756469 hasRelatedWork W2739991629 @default.
- W3141756469 hasRelatedWork W3107474891 @default.
- W3141756469 hasRelatedWork W3156915121 @default.
- W3141756469 hasRelatedWork W60887625 @default.
- W3141756469 isParatext "false" @default.
- W3141756469 isRetracted "false" @default.
- W3141756469 magId "3141756469" @default.
- W3141756469 workType "article" @default.