Matches in SemOpenAlex for { <https://semopenalex.org/work/W2300901042> ?p ?o ?g. }
Showing items 1 to 85 of
85
with 100 items per page.
- W2300901042 abstract "This paper proposes a method of acoustic modeling for zero-resourced languages speech recognition under mismatch conditions. In those languages, very limited or no transcribed speech is available for traditional monolingual speech recognition. Conventional methods such as IPA based universal acoustic modeling has been proved to be effective under matched acoustic conditions (similar speaking styles, adjacent languages, etc.), while usually poorly preformed when mismatch occurs. Since mismatch problems between languages often appears, in this paper, unsupervised acoustic modeling via cross-lingual knowledge sharing has thus been proposed: first, initial acoustic models (AM) for a target zero-resourced language are trained using Multi-Task Deep Neural Networks (MDNN) – different languages’ speech mapped to the phonemes of the target language (mapped data) is jointly trained together with the same data transcribed language specifically and respectively (specific data); then, automatically transcribed target language data is used in the iterative process to train new AMs, with various auxiliary tasks. Experiment on 100 hour Japanese speech without transcripts achieved a character error rate (CER) of 57.21%, 19.32% absolute improvement compared to baseline (IPA based universal acoustic modeling)." @default.
- W2300901042 created "2016-06-24" @default.
- W2300901042 creator A5015005113 @default.
- W2300901042 creator A5023545041 @default.
- W2300901042 creator A5029522505 @default.
- W2300901042 creator A5061055212 @default.
- W2300901042 date "2016-01-01" @default.
- W2300901042 modified "2023-09-28" @default.
- W2300901042 title "Efficient Acoustic Modeling Method for Unsupervised Speech Recognition using Multi-Task Deep Neural Network" @default.
- W2300901042 cites W1524333225 @default.
- W2300901042 cites W2026369565 @default.
- W2300901042 cites W2040903810 @default.
- W2300901042 cites W2091746061 @default.
- W2300901042 cites W2133267619 @default.
- W2300901042 cites W2160815625 @default.
- W2300901042 cites W2164505566 @default.
- W2300901042 cites W2402741009 @default.
- W2300901042 doi "https://doi.org/10.2991/nceece-15.2016.72" @default.
- W2300901042 hasPublicationYear "2016" @default.
- W2300901042 type Work @default.
- W2300901042 sameAs 2300901042 @default.
- W2300901042 citedByCount "2" @default.
- W2300901042 countsByYear W23009010422016 @default.
- W2300901042 countsByYear W23009010422020 @default.
- W2300901042 crossrefType "proceedings-article" @default.
- W2300901042 hasAuthorship W2300901042A5015005113 @default.
- W2300901042 hasAuthorship W2300901042A5023545041 @default.
- W2300901042 hasAuthorship W2300901042A5029522505 @default.
- W2300901042 hasAuthorship W2300901042A5061055212 @default.
- W2300901042 hasBestOaLocation W23009010421 @default.
- W2300901042 hasConcept C111919701 @default.
- W2300901042 hasConcept C137293760 @default.
- W2300901042 hasConcept C154945302 @default.
- W2300901042 hasConcept C155635449 @default.
- W2300901042 hasConcept C162324750 @default.
- W2300901042 hasConcept C187736073 @default.
- W2300901042 hasConcept C204321447 @default.
- W2300901042 hasConcept C2780451532 @default.
- W2300901042 hasConcept C28490314 @default.
- W2300901042 hasConcept C40969351 @default.
- W2300901042 hasConcept C41008148 @default.
- W2300901042 hasConcept C50644808 @default.
- W2300901042 hasConcept C61328038 @default.
- W2300901042 hasConcept C98045186 @default.
- W2300901042 hasConceptScore W2300901042C111919701 @default.
- W2300901042 hasConceptScore W2300901042C137293760 @default.
- W2300901042 hasConceptScore W2300901042C154945302 @default.
- W2300901042 hasConceptScore W2300901042C155635449 @default.
- W2300901042 hasConceptScore W2300901042C162324750 @default.
- W2300901042 hasConceptScore W2300901042C187736073 @default.
- W2300901042 hasConceptScore W2300901042C204321447 @default.
- W2300901042 hasConceptScore W2300901042C2780451532 @default.
- W2300901042 hasConceptScore W2300901042C28490314 @default.
- W2300901042 hasConceptScore W2300901042C40969351 @default.
- W2300901042 hasConceptScore W2300901042C41008148 @default.
- W2300901042 hasConceptScore W2300901042C50644808 @default.
- W2300901042 hasConceptScore W2300901042C61328038 @default.
- W2300901042 hasConceptScore W2300901042C98045186 @default.
- W2300901042 hasLocation W23009010421 @default.
- W2300901042 hasOpenAccess W2300901042 @default.
- W2300901042 hasPrimaryLocation W23009010421 @default.
- W2300901042 hasRelatedWork W1496183372 @default.
- W2300901042 hasRelatedWork W1513252538 @default.
- W2300901042 hasRelatedWork W1567690964 @default.
- W2300901042 hasRelatedWork W2044577954 @default.
- W2300901042 hasRelatedWork W2094035326 @default.
- W2300901042 hasRelatedWork W2118592515 @default.
- W2300901042 hasRelatedWork W2119665677 @default.
- W2300901042 hasRelatedWork W2121564724 @default.
- W2300901042 hasRelatedWork W2143849997 @default.
- W2300901042 hasRelatedWork W2163045068 @default.
- W2300901042 hasRelatedWork W2183354483 @default.
- W2300901042 hasRelatedWork W2292492052 @default.
- W2300901042 hasRelatedWork W2403553999 @default.
- W2300901042 hasRelatedWork W2511899227 @default.
- W2300901042 hasRelatedWork W2652238127 @default.
- W2300901042 hasRelatedWork W2795349724 @default.
- W2300901042 hasRelatedWork W2901418951 @default.
- W2300901042 hasRelatedWork W2913641852 @default.
- W2300901042 hasRelatedWork W2060312364 @default.
- W2300901042 hasRelatedWork W3056522967 @default.
- W2300901042 isParatext "false" @default.
- W2300901042 isRetracted "false" @default.
- W2300901042 magId "2300901042" @default.
- W2300901042 workType "article" @default.