Matches in SemOpenAlex for { <https://semopenalex.org/work/W2952976827> ?p ?o ?g. }
Showing items 1 to 88 of
88
with 100 items per page.
- W2952976827 abstract "Furui first demonstrated that the identity of both consonant and vowel can be perceived from the C-V transition; later, Stevens proposed that acoustic landmarks are the primary cues for speech perception, and that steady-state regions are secondary or supplemental. Acoustic landmarks are perceptually salient, even in a language one doesn't speak, and it has been demonstrated that non-speakers of the language can identify features such as the primary articulator of the landmark. These factors suggest a strategy for developing language-independent automatic speech recognition: landmarks can potentially be learned once from a suitably labeled corpus and rapidly applied to many other languages. This paper proposes enhancing the cross-lingual portability of a neural network by using landmarks as the secondary task in multi-task learning (MTL). The network is trained in a well-resourced source language with both phone and landmark labels (English), then adapted to an under-resourced target language with only word labels (Iban). Landmark-tasked MTL reduces source-language phone error rate by 2.9% relative, and reduces target-language word error rate by 1.9%-5.9% depending on the amount of target-language training data. These results suggest that landmark-tasked MTL causes the DNN to learn hidden-node features that are useful for cross-lingual adaptation." @default.
- W2952976827 created "2019-06-27" @default.
- W2952976827 creator A5004778663 @default.
- W2952976827 creator A5026225097 @default.
- W2952976827 creator A5040125616 @default.
- W2952976827 creator A5056321228 @default.
- W2952976827 creator A5059726557 @default.
- W2952976827 date "2018-05-15" @default.
- W2952976827 modified "2023-09-24" @default.
- W2952976827 title "Improved ASR for Under-Resourced Languages Through Multi-Task Learning with Acoustic Landmarks" @default.
- W2952976827 cites W1436754857 @default.
- W2952976827 cites W1499332833 @default.
- W2952976827 cites W1631260214 @default.
- W2952976827 cites W16621805 @default.
- W2952976827 cites W1967177300 @default.
- W2952976827 cites W1993882792 @default.
- W2952976827 cites W2001496355 @default.
- W2952976827 cites W2076395366 @default.
- W2952976827 cites W2086667971 @default.
- W2952976827 cites W2094035326 @default.
- W2952976827 cites W2119155265 @default.
- W2952976827 cites W2137630452 @default.
- W2952976827 cites W2396466159 @default.
- W2952976827 cites W2787228318 @default.
- W2952976827 cites W3104406212 @default.
- W2952976827 doi "https://doi.org/10.48550/arxiv.1805.05574" @default.
- W2952976827 hasPublicationYear "2018" @default.
- W2952976827 type Work @default.
- W2952976827 sameAs 2952976827 @default.
- W2952976827 citedByCount "2" @default.
- W2952976827 countsByYear W29529768272019 @default.
- W2952976827 countsByYear W29529768272020 @default.
- W2952976827 crossrefType "posted-content" @default.
- W2952976827 hasAuthorship W2952976827A5004778663 @default.
- W2952976827 hasAuthorship W2952976827A5026225097 @default.
- W2952976827 hasAuthorship W2952976827A5040125616 @default.
- W2952976827 hasAuthorship W2952976827A5056321228 @default.
- W2952976827 hasAuthorship W2952976827A5059726557 @default.
- W2952976827 hasBestOaLocation W29529768271 @default.
- W2952976827 hasConcept C138885662 @default.
- W2952976827 hasConcept C154945302 @default.
- W2952976827 hasConcept C162324750 @default.
- W2952976827 hasConcept C187736073 @default.
- W2952976827 hasConcept C204321447 @default.
- W2952976827 hasConcept C2778415344 @default.
- W2952976827 hasConcept C2778707766 @default.
- W2952976827 hasConcept C2780297707 @default.
- W2952976827 hasConcept C2780451532 @default.
- W2952976827 hasConcept C2780719617 @default.
- W2952976827 hasConcept C28490314 @default.
- W2952976827 hasConcept C29694066 @default.
- W2952976827 hasConcept C40969351 @default.
- W2952976827 hasConcept C41008148 @default.
- W2952976827 hasConcept C41895202 @default.
- W2952976827 hasConcept C71924100 @default.
- W2952976827 hasConceptScore W2952976827C138885662 @default.
- W2952976827 hasConceptScore W2952976827C154945302 @default.
- W2952976827 hasConceptScore W2952976827C162324750 @default.
- W2952976827 hasConceptScore W2952976827C187736073 @default.
- W2952976827 hasConceptScore W2952976827C204321447 @default.
- W2952976827 hasConceptScore W2952976827C2778415344 @default.
- W2952976827 hasConceptScore W2952976827C2778707766 @default.
- W2952976827 hasConceptScore W2952976827C2780297707 @default.
- W2952976827 hasConceptScore W2952976827C2780451532 @default.
- W2952976827 hasConceptScore W2952976827C2780719617 @default.
- W2952976827 hasConceptScore W2952976827C28490314 @default.
- W2952976827 hasConceptScore W2952976827C29694066 @default.
- W2952976827 hasConceptScore W2952976827C40969351 @default.
- W2952976827 hasConceptScore W2952976827C41008148 @default.
- W2952976827 hasConceptScore W2952976827C41895202 @default.
- W2952976827 hasConceptScore W2952976827C71924100 @default.
- W2952976827 hasLocation W29529768271 @default.
- W2952976827 hasOpenAccess W2952976827 @default.
- W2952976827 hasPrimaryLocation W29529768271 @default.
- W2952976827 hasRelatedWork W1536476694 @default.
- W2952976827 hasRelatedWork W1967477266 @default.
- W2952976827 hasRelatedWork W1973215129 @default.
- W2952976827 hasRelatedWork W2003890417 @default.
- W2952976827 hasRelatedWork W2061937230 @default.
- W2952976827 hasRelatedWork W2080458333 @default.
- W2952976827 hasRelatedWork W2081647779 @default.
- W2952976827 hasRelatedWork W2097989758 @default.
- W2952976827 hasRelatedWork W2285081780 @default.
- W2952976827 hasRelatedWork W3185852197 @default.
- W2952976827 isParatext "false" @default.
- W2952976827 isRetracted "false" @default.
- W2952976827 magId "2952976827" @default.
- W2952976827 workType "article" @default.