Matches in SemOpenAlex for { <https://semopenalex.org/work/W2810852753> ?p ?o ?g. }
Showing items 1 to 88 of
88
with 100 items per page.
- W2810852753 abstract "Dealing with speech corrupted by noise and reverberation is still an issue for automatic speech recognition. To address this, a solution that can be combined with multi-style learning consists of using multi-task learning, where the acoustic model is trained to solve one main task and at least one auxiliary task simultaneously. In noisy and reverberant environment using clean-speech features estimation as auxiliary task has proven its efficiency, while the main task is speech recognition. Still, recognizing speech in these degraded conditions is all the more difficult when the amount of available data during training is limited. Thus, in this paper, we evaluate robust speech recognition based on multi-task learning when the amount of training data is gradually reduced. We show that using multi-task learning improves recognition and more specifically, that its impact is even more significant if the amount of training data is decreasing (up to 12% relative improvement of the word error rate). All training and testing experiments are carried out on parts of the CHiME4 database." @default.
- W2810852753 created "2018-07-10" @default.
- W2810852753 creator A5040916551 @default.
- W2810852753 creator A5042439255 @default.
- W2810852753 creator A5083563222 @default.
- W2810852753 date "2017-12-01" @default.
- W2810852753 modified "2023-09-25" @default.
- W2810852753 title "Investigating the impact of the training data volume for robust speech recognition using multi-task learning" @default.
- W2810852753 cites W129870797 @default.
- W2810852753 cites W1499999342 @default.
- W2810852753 cites W1524333225 @default.
- W2810852753 cites W2160815625 @default.
- W2810852753 cites W2237224976 @default.
- W2810852753 cites W2291456664 @default.
- W2810852753 cites W2346842942 @default.
- W2810852753 cites W2398723458 @default.
- W2810852753 cites W2402201734 @default.
- W2810852753 cites W2506203739 @default.
- W2810852753 cites W2533523411 @default.
- W2810852753 cites W2550928100 @default.
- W2810852753 cites W2559260703 @default.
- W2810852753 cites W2586665961 @default.
- W2810852753 cites W2607725416 @default.
- W2810852753 cites W2913340405 @default.
- W2810852753 cites W2919115771 @default.
- W2810852753 doi "https://doi.org/10.1109/isspit.2017.8388673" @default.
- W2810852753 hasPublicationYear "2017" @default.
- W2810852753 type Work @default.
- W2810852753 sameAs 2810852753 @default.
- W2810852753 citedByCount "1" @default.
- W2810852753 countsByYear W28108527532019 @default.
- W2810852753 crossrefType "proceedings-article" @default.
- W2810852753 hasAuthorship W2810852753A5040916551 @default.
- W2810852753 hasAuthorship W2810852753A5042439255 @default.
- W2810852753 hasAuthorship W2810852753A5083563222 @default.
- W2810852753 hasConcept C115961682 @default.
- W2810852753 hasConcept C119599485 @default.
- W2810852753 hasConcept C127413603 @default.
- W2810852753 hasConcept C154945302 @default.
- W2810852753 hasConcept C201995342 @default.
- W2810852753 hasConcept C2780451532 @default.
- W2810852753 hasConcept C28006648 @default.
- W2810852753 hasConcept C28490314 @default.
- W2810852753 hasConcept C40969351 @default.
- W2810852753 hasConcept C41008148 @default.
- W2810852753 hasConcept C51632099 @default.
- W2810852753 hasConcept C95851461 @default.
- W2810852753 hasConcept C99498987 @default.
- W2810852753 hasConceptScore W2810852753C115961682 @default.
- W2810852753 hasConceptScore W2810852753C119599485 @default.
- W2810852753 hasConceptScore W2810852753C127413603 @default.
- W2810852753 hasConceptScore W2810852753C154945302 @default.
- W2810852753 hasConceptScore W2810852753C201995342 @default.
- W2810852753 hasConceptScore W2810852753C2780451532 @default.
- W2810852753 hasConceptScore W2810852753C28006648 @default.
- W2810852753 hasConceptScore W2810852753C28490314 @default.
- W2810852753 hasConceptScore W2810852753C40969351 @default.
- W2810852753 hasConceptScore W2810852753C41008148 @default.
- W2810852753 hasConceptScore W2810852753C51632099 @default.
- W2810852753 hasConceptScore W2810852753C95851461 @default.
- W2810852753 hasConceptScore W2810852753C99498987 @default.
- W2810852753 hasLocation W28108527531 @default.
- W2810852753 hasOpenAccess W2810852753 @default.
- W2810852753 hasPrimaryLocation W28108527531 @default.
- W2810852753 hasRelatedWork W147859270 @default.
- W2810852753 hasRelatedWork W1493817255 @default.
- W2810852753 hasRelatedWork W2051727797 @default.
- W2810852753 hasRelatedWork W2096594917 @default.
- W2810852753 hasRelatedWork W2124338691 @default.
- W2810852753 hasRelatedWork W2169813781 @default.
- W2810852753 hasRelatedWork W2526425061 @default.
- W2810852753 hasRelatedWork W2548494187 @default.
- W2810852753 hasRelatedWork W2565624985 @default.
- W2810852753 hasRelatedWork W2576549984 @default.
- W2810852753 hasRelatedWork W2586425531 @default.
- W2810852753 hasRelatedWork W2586665961 @default.
- W2810852753 hasRelatedWork W2892204792 @default.
- W2810852753 hasRelatedWork W2922358226 @default.
- W2810852753 hasRelatedWork W2952206214 @default.
- W2810852753 hasRelatedWork W2962858469 @default.
- W2810852753 hasRelatedWork W2972592847 @default.
- W2810852753 hasRelatedWork W3021188327 @default.
- W2810852753 hasRelatedWork W3105591863 @default.
- W2810852753 hasRelatedWork W3163142165 @default.
- W2810852753 isParatext "false" @default.
- W2810852753 isRetracted "false" @default.
- W2810852753 magId "2810852753" @default.
- W2810852753 workType "article" @default.