Matches in SemOpenAlex for { <https://semopenalex.org/work/W52100036> ?p ?o ?g. }
Showing items 1 to 60 of
60
with 100 items per page.
- W52100036 abstract "An adequate approach to speech translation for small to medium sized tasks is the use of subsequential transducers —a finite state model— as language model for a speech recognizer. These transducers can be automatically trained from sample corpora. The use of manually defined categories improves the training of the subsequential transducers when the available data are scarce. These categories depend on the source and target languages we want to translate. We introduce an automatic approach to derive categories that can be used in training subsequential transducers. This approach extends monolingual word clustering methods to the bilingual case using alignments obtained from statistical models. Experimental results indicate that the models trained with these categories have lower translation errors." @default.
- W52100036 created "2016-06-24" @default.
- W52100036 creator A5038505623 @default.
- W52100036 creator A5090675958 @default.
- W52100036 date "1999-09-05" @default.
- W52100036 modified "2023-09-27" @default.
- W52100036 title "Automatically deriving categories for translation" @default.
- W52100036 cites W1520943203 @default.
- W52100036 cites W154961936 @default.
- W52100036 cites W2006969979 @default.
- W52100036 cites W2080018251 @default.
- W52100036 cites W2118172714 @default.
- W52100036 cites W2121227244 @default.
- W52100036 cites W2162157640 @default.
- W52100036 cites W3088638701 @default.
- W52100036 cites W47415966 @default.
- W52100036 cites W55092401 @default.
- W52100036 cites W87890863 @default.
- W52100036 cites W328769118 @default.
- W52100036 doi "https://doi.org/10.21437/eurospeech.1999-530" @default.
- W52100036 hasPublicationYear "1999" @default.
- W52100036 type Work @default.
- W52100036 sameAs 52100036 @default.
- W52100036 citedByCount "1" @default.
- W52100036 crossrefType "proceedings-article" @default.
- W52100036 hasAuthorship W52100036A5038505623 @default.
- W52100036 hasAuthorship W52100036A5090675958 @default.
- W52100036 hasConcept C104317684 @default.
- W52100036 hasConcept C105580179 @default.
- W52100036 hasConcept C149364088 @default.
- W52100036 hasConcept C154945302 @default.
- W52100036 hasConcept C185592680 @default.
- W52100036 hasConcept C204321447 @default.
- W52100036 hasConcept C41008148 @default.
- W52100036 hasConcept C55493867 @default.
- W52100036 hasConceptScore W52100036C104317684 @default.
- W52100036 hasConceptScore W52100036C105580179 @default.
- W52100036 hasConceptScore W52100036C149364088 @default.
- W52100036 hasConceptScore W52100036C154945302 @default.
- W52100036 hasConceptScore W52100036C185592680 @default.
- W52100036 hasConceptScore W52100036C204321447 @default.
- W52100036 hasConceptScore W52100036C41008148 @default.
- W52100036 hasConceptScore W52100036C55493867 @default.
- W52100036 hasLocation W521000361 @default.
- W52100036 hasOpenAccess W52100036 @default.
- W52100036 hasPrimaryLocation W521000361 @default.
- W52100036 hasRelatedWork W1552159754 @default.
- W52100036 hasRelatedWork W2131420137 @default.
- W52100036 hasRelatedWork W2148757832 @default.
- W52100036 hasRelatedWork W2151447942 @default.
- W52100036 hasRelatedWork W2293457016 @default.
- W52100036 hasRelatedWork W2368651715 @default.
- W52100036 hasRelatedWork W2611614995 @default.
- W52100036 hasRelatedWork W2789919619 @default.
- W52100036 hasRelatedWork W3107474891 @default.
- W52100036 hasRelatedWork W3169305685 @default.
- W52100036 isParatext "false" @default.
- W52100036 isRetracted "false" @default.
- W52100036 magId "52100036" @default.
- W52100036 workType "article" @default.