Matches in SemOpenAlex for { <https://semopenalex.org/work/W2170406711> ?p ?o ?g. }
- W2170406711 abstract "We present our work on semi-supervised learning of discriminative language models where the negative examples for sentences in a text corpus are generated using confusion models for Turkish at various granularities, specifically, word, sub-word, syllable and phone levels. We experiment with different language models and various sampling strategies to select competing hypotheses for training with a variant of the perceptron algorithm. We find that morph-based confusion models with a sample selection strategy aiming to match the error distribution of the baseline ASR system gives the best performance. We also observe that substituting half of the supervised training examples with those obtained in a semi-supervised manner gives similar results." @default.
- W2170406711 created "2016-06-24" @default.
- W2170406711 creator A5007425606 @default.
- W2170406711 creator A5010122901 @default.
- W2170406711 creator A5013958949 @default.
- W2170406711 creator A5020068498 @default.
- W2170406711 creator A5025419994 @default.
- W2170406711 creator A5027532022 @default.
- W2170406711 creator A5047997552 @default.
- W2170406711 creator A5051484770 @default.
- W2170406711 creator A5054618933 @default.
- W2170406711 creator A5054865609 @default.
- W2170406711 creator A5055086464 @default.
- W2170406711 creator A5062345833 @default.
- W2170406711 creator A5068508539 @default.
- W2170406711 creator A5069654424 @default.
- W2170406711 creator A5070742827 @default.
- W2170406711 creator A5071170549 @default.
- W2170406711 creator A5072223342 @default.
- W2170406711 creator A5072978012 @default.
- W2170406711 creator A5074890965 @default.
- W2170406711 creator A5082446768 @default.
- W2170406711 creator A5085014772 @default.
- W2170406711 creator A5089260563 @default.
- W2170406711 date "2012-03-01" @default.
- W2170406711 modified "2023-10-11" @default.
- W2170406711 title "Semi-supervised discriminative language modeling for Turkish ASR" @default.
- W2170406711 cites W168280322 @default.
- W2170406711 cites W2078349113 @default.
- W2170406711 cites W2112752961 @default.
- W2170406711 cites W2137675115 @default.
- W2170406711 cites W2148182949 @default.
- W2170406711 cites W2152753389 @default.
- W2170406711 cites W2158025800 @default.
- W2170406711 cites W4245685776 @default.
- W2170406711 cites W99561634 @default.
- W2170406711 doi "https://doi.org/10.1109/icassp.2012.6289049" @default.
- W2170406711 hasPublicationYear "2012" @default.
- W2170406711 type Work @default.
- W2170406711 sameAs 2170406711 @default.
- W2170406711 citedByCount "13" @default.
- W2170406711 countsByYear W21704067112012 @default.
- W2170406711 countsByYear W21704067112013 @default.
- W2170406711 countsByYear W21704067112014 @default.
- W2170406711 countsByYear W21704067112015 @default.
- W2170406711 countsByYear W21704067112019 @default.
- W2170406711 countsByYear W21704067112021 @default.
- W2170406711 countsByYear W21704067112023 @default.
- W2170406711 crossrefType "proceedings-article" @default.
- W2170406711 hasAuthorship W2170406711A5007425606 @default.
- W2170406711 hasAuthorship W2170406711A5010122901 @default.
- W2170406711 hasAuthorship W2170406711A5013958949 @default.
- W2170406711 hasAuthorship W2170406711A5020068498 @default.
- W2170406711 hasAuthorship W2170406711A5025419994 @default.
- W2170406711 hasAuthorship W2170406711A5027532022 @default.
- W2170406711 hasAuthorship W2170406711A5047997552 @default.
- W2170406711 hasAuthorship W2170406711A5051484770 @default.
- W2170406711 hasAuthorship W2170406711A5054618933 @default.
- W2170406711 hasAuthorship W2170406711A5054865609 @default.
- W2170406711 hasAuthorship W2170406711A5055086464 @default.
- W2170406711 hasAuthorship W2170406711A5062345833 @default.
- W2170406711 hasAuthorship W2170406711A5068508539 @default.
- W2170406711 hasAuthorship W2170406711A5069654424 @default.
- W2170406711 hasAuthorship W2170406711A5070742827 @default.
- W2170406711 hasAuthorship W2170406711A5071170549 @default.
- W2170406711 hasAuthorship W2170406711A5072223342 @default.
- W2170406711 hasAuthorship W2170406711A5072978012 @default.
- W2170406711 hasAuthorship W2170406711A5074890965 @default.
- W2170406711 hasAuthorship W2170406711A5082446768 @default.
- W2170406711 hasAuthorship W2170406711A5085014772 @default.
- W2170406711 hasAuthorship W2170406711A5089260563 @default.
- W2170406711 hasBestOaLocation W21704067112 @default.
- W2170406711 hasConcept C109089402 @default.
- W2170406711 hasConcept C11171543 @default.
- W2170406711 hasConcept C119857082 @default.
- W2170406711 hasConcept C136389625 @default.
- W2170406711 hasConcept C137293760 @default.
- W2170406711 hasConcept C138885662 @default.
- W2170406711 hasConcept C154945302 @default.
- W2170406711 hasConcept C15744967 @default.
- W2170406711 hasConcept C204321447 @default.
- W2170406711 hasConcept C2778707766 @default.
- W2170406711 hasConcept C2781121862 @default.
- W2170406711 hasConcept C2781140086 @default.
- W2170406711 hasConcept C28490314 @default.
- W2170406711 hasConcept C41008148 @default.
- W2170406711 hasConcept C41895202 @default.
- W2170406711 hasConcept C50644808 @default.
- W2170406711 hasConcept C60908668 @default.
- W2170406711 hasConcept C81917197 @default.
- W2170406711 hasConcept C90805587 @default.
- W2170406711 hasConcept C97931131 @default.
- W2170406711 hasConceptScore W2170406711C109089402 @default.
- W2170406711 hasConceptScore W2170406711C11171543 @default.
- W2170406711 hasConceptScore W2170406711C119857082 @default.
- W2170406711 hasConceptScore W2170406711C136389625 @default.
- W2170406711 hasConceptScore W2170406711C137293760 @default.
- W2170406711 hasConceptScore W2170406711C138885662 @default.
- W2170406711 hasConceptScore W2170406711C154945302 @default.
- W2170406711 hasConceptScore W2170406711C15744967 @default.