Matches in SemOpenAlex for { <https://semopenalex.org/work/W4372267638> ?p ?o ?g. }
- W4372267638 abstract "The black-box nature of end-to-end speech-to-text translation (E2E ST) makes it difficult to understand how source language inputs are being mapped to the target language. To solve this problem, we propose to simultaneously generate automatic speech recognition (ASR) and ST predictions such that each source language word is explicitly mapped to a target language word. A major challenge arises from the fact that translation is a non-monotonic sequence transduction task due to word ordering differences between languages – this clashes with the monotonic nature of ASR. Therefore, we propose to generate ST tokens out-of-order while remembering how to re-order them later. We achieve this by predicting a sequence of tuples consisting of a source word, the corresponding target words, and post-editing operations dictating the correct insertion points for the target word. We examine two variants of such operation sequences which enable generation of monotonic transcriptions and non-monotonic translations from the same speech input simultaneously. We apply our approach to offline and real-time streaming models, demonstrating that we can provide explainable translations without sacrificing quality or latency. In fact, the delayed re-ordering ability of our approach improves performance during streaming. As an added benefit, our method performs ASR and ST simultaneously, making it faster than using two separate systems to perform these tasks." @default.
- W4372267638 created "2023-05-07" @default.
- W4372267638 creator A5001291873 @default.
- W4372267638 creator A5021201726 @default.
- W4372267638 creator A5068873086 @default.
- W4372267638 creator A5079822291 @default.
- W4372267638 creator A5088171135 @default.
- W4372267638 date "2023-06-04" @default.
- W4372267638 modified "2023-09-27" @default.
- W4372267638 title "Align, Write, Re-Order: Explainable End-to-End Speech Translation via Operation Sequence Generation" @default.
- W4372267638 cites W2080373976 @default.
- W4372267638 cites W2100398002 @default.
- W4372267638 cites W2124807415 @default.
- W4372267638 cites W2605131327 @default.
- W4372267638 cites W2936969148 @default.
- W4372267638 cites W2951456627 @default.
- W4372267638 cites W2962757645 @default.
- W4372267638 cites W2962780374 @default.
- W4372267638 cites W2963532001 @default.
- W4372267638 cites W2964172053 @default.
- W4372267638 cites W2985694911 @default.
- W4372267638 cites W2998386507 @default.
- W4372267638 cites W3007142233 @default.
- W4372267638 cites W3015583403 @default.
- W4372267638 cites W3037217258 @default.
- W4372267638 cites W3097777922 @default.
- W4372267638 cites W3102811925 @default.
- W4372267638 cites W3105681039 @default.
- W4372267638 cites W3113676066 @default.
- W4372267638 cites W3143377973 @default.
- W4372267638 cites W3148654612 @default.
- W4372267638 cites W3152534858 @default.
- W4372267638 cites W3156246620 @default.
- W4372267638 cites W3160950270 @default.
- W4372267638 cites W3162919436 @default.
- W4372267638 cites W3163793923 @default.
- W4372267638 cites W3172862365 @default.
- W4372267638 cites W3174032041 @default.
- W4372267638 cites W3175164646 @default.
- W4372267638 cites W3186200218 @default.
- W4372267638 cites W3186672448 @default.
- W4372267638 cites W4285229778 @default.
- W4372267638 cites W4285242950 @default.
- W4372267638 doi "https://doi.org/10.1109/icassp49357.2023.10095896" @default.
- W4372267638 hasPublicationYear "2023" @default.
- W4372267638 type Work @default.
- W4372267638 citedByCount "0" @default.
- W4372267638 crossrefType "proceedings-article" @default.
- W4372267638 hasAuthorship W4372267638A5001291873 @default.
- W4372267638 hasAuthorship W4372267638A5021201726 @default.
- W4372267638 hasAuthorship W4372267638A5068873086 @default.
- W4372267638 hasAuthorship W4372267638A5079822291 @default.
- W4372267638 hasAuthorship W4372267638A5088171135 @default.
- W4372267638 hasBestOaLocation W43722676381 @default.
- W4372267638 hasConcept C104317684 @default.
- W4372267638 hasConcept C105580179 @default.
- W4372267638 hasConcept C118615104 @default.
- W4372267638 hasConcept C118930307 @default.
- W4372267638 hasConcept C134306372 @default.
- W4372267638 hasConcept C137293760 @default.
- W4372267638 hasConcept C138885662 @default.
- W4372267638 hasConcept C149364088 @default.
- W4372267638 hasConcept C154945302 @default.
- W4372267638 hasConcept C185592680 @default.
- W4372267638 hasConcept C203005215 @default.
- W4372267638 hasConcept C204321447 @default.
- W4372267638 hasConcept C2778112365 @default.
- W4372267638 hasConcept C2780366754 @default.
- W4372267638 hasConcept C28490314 @default.
- W4372267638 hasConcept C33923547 @default.
- W4372267638 hasConcept C41008148 @default.
- W4372267638 hasConcept C41895202 @default.
- W4372267638 hasConcept C54355233 @default.
- W4372267638 hasConcept C55493867 @default.
- W4372267638 hasConcept C70777604 @default.
- W4372267638 hasConcept C72169020 @default.
- W4372267638 hasConcept C76155785 @default.
- W4372267638 hasConcept C82876162 @default.
- W4372267638 hasConcept C86803240 @default.
- W4372267638 hasConcept C90805587 @default.
- W4372267638 hasConceptScore W4372267638C104317684 @default.
- W4372267638 hasConceptScore W4372267638C105580179 @default.
- W4372267638 hasConceptScore W4372267638C118615104 @default.
- W4372267638 hasConceptScore W4372267638C118930307 @default.
- W4372267638 hasConceptScore W4372267638C134306372 @default.
- W4372267638 hasConceptScore W4372267638C137293760 @default.
- W4372267638 hasConceptScore W4372267638C138885662 @default.
- W4372267638 hasConceptScore W4372267638C149364088 @default.
- W4372267638 hasConceptScore W4372267638C154945302 @default.
- W4372267638 hasConceptScore W4372267638C185592680 @default.
- W4372267638 hasConceptScore W4372267638C203005215 @default.
- W4372267638 hasConceptScore W4372267638C204321447 @default.
- W4372267638 hasConceptScore W4372267638C2778112365 @default.
- W4372267638 hasConceptScore W4372267638C2780366754 @default.
- W4372267638 hasConceptScore W4372267638C28490314 @default.
- W4372267638 hasConceptScore W4372267638C33923547 @default.
- W4372267638 hasConceptScore W4372267638C41008148 @default.
- W4372267638 hasConceptScore W4372267638C41895202 @default.
- W4372267638 hasConceptScore W4372267638C54355233 @default.
- W4372267638 hasConceptScore W4372267638C55493867 @default.