Matches in SemOpenAlex for { <https://semopenalex.org/work/W323718150> ?p ?o ?g. }
Showing items 1 to 86 of
86
with 100 items per page.
- W323718150 abstract "State-of-the-art large vocabulary continuous speech recognition (LVCSR) systems often combine outputs from multiple sub-systems that may even be developed at different sites. Cross system adaptation, in which model adaptation is performed using the outputs from another sub-system, can be used as an alternative to hypothesis level combination schemes such as ROVER. Normally cross adaptation is only performed on the acoustic models. However, there are many other levels in LVCSR systems' modelling hierarchy where complimentary features may be exploited, for example, the sub-word and the word level, to further improve cross adaptation based system combination. It is thus interesting to also cross adapt language models (LMs) to capture these additional useful features. In this paper cross adaptation is applied to three forms of language models, a multi-level LM that models both syllable and word sequences, a word level neural network LM, and the linear combination of the two. Significant error rate reductions of 4.0-7.1% relative were obtained over ROVER and acoustic model only cross adaptation when combining a range of Chinese LVCSR sub-systems used in the 2010 and 2011 DARPA GALE evaluations." @default.
- W323718150 created "2016-06-24" @default.
- W323718150 creator A5002191410 @default.
- W323718150 creator A5037109470 @default.
- W323718150 creator A5050766679 @default.
- W323718150 date "2010-09-26" @default.
- W323718150 modified "2023-09-26" @default.
- W323718150 title "Language model cross adaptation for LVCSR system combination" @default.
- W323718150 doi "https://doi.org/10.21437/interspeech.2010-126" @default.
- W323718150 hasPublicationYear "2010" @default.
- W323718150 type Work @default.
- W323718150 sameAs 323718150 @default.
- W323718150 citedByCount "3" @default.
- W323718150 countsByYear W3237181502012 @default.
- W323718150 countsByYear W3237181502013 @default.
- W323718150 crossrefType "proceedings-article" @default.
- W323718150 hasAuthorship W323718150A5002191410 @default.
- W323718150 hasAuthorship W323718150A5037109470 @default.
- W323718150 hasAuthorship W323718150A5050766679 @default.
- W323718150 hasConcept C109089402 @default.
- W323718150 hasConcept C120665830 @default.
- W323718150 hasConcept C121332964 @default.
- W323718150 hasConcept C137293760 @default.
- W323718150 hasConcept C138885662 @default.
- W323718150 hasConcept C139807058 @default.
- W323718150 hasConcept C154945302 @default.
- W323718150 hasConcept C159985019 @default.
- W323718150 hasConcept C162324750 @default.
- W323718150 hasConcept C192562407 @default.
- W323718150 hasConcept C204321447 @default.
- W323718150 hasConcept C204323151 @default.
- W323718150 hasConcept C2777601683 @default.
- W323718150 hasConcept C28490314 @default.
- W323718150 hasConcept C31170391 @default.
- W323718150 hasConcept C34447519 @default.
- W323718150 hasConcept C40969351 @default.
- W323718150 hasConcept C41008148 @default.
- W323718150 hasConcept C41895202 @default.
- W323718150 hasConcept C90805587 @default.
- W323718150 hasConceptScore W323718150C109089402 @default.
- W323718150 hasConceptScore W323718150C120665830 @default.
- W323718150 hasConceptScore W323718150C121332964 @default.
- W323718150 hasConceptScore W323718150C137293760 @default.
- W323718150 hasConceptScore W323718150C138885662 @default.
- W323718150 hasConceptScore W323718150C139807058 @default.
- W323718150 hasConceptScore W323718150C154945302 @default.
- W323718150 hasConceptScore W323718150C159985019 @default.
- W323718150 hasConceptScore W323718150C162324750 @default.
- W323718150 hasConceptScore W323718150C192562407 @default.
- W323718150 hasConceptScore W323718150C204321447 @default.
- W323718150 hasConceptScore W323718150C204323151 @default.
- W323718150 hasConceptScore W323718150C2777601683 @default.
- W323718150 hasConceptScore W323718150C28490314 @default.
- W323718150 hasConceptScore W323718150C31170391 @default.
- W323718150 hasConceptScore W323718150C34447519 @default.
- W323718150 hasConceptScore W323718150C40969351 @default.
- W323718150 hasConceptScore W323718150C41008148 @default.
- W323718150 hasConceptScore W323718150C41895202 @default.
- W323718150 hasConceptScore W323718150C90805587 @default.
- W323718150 hasLocation W3237181501 @default.
- W323718150 hasOpenAccess W323718150 @default.
- W323718150 hasPrimaryLocation W3237181501 @default.
- W323718150 hasRelatedWork W1493433087 @default.
- W323718150 hasRelatedWork W1543083551 @default.
- W323718150 hasRelatedWork W1544673079 @default.
- W323718150 hasRelatedWork W1566337287 @default.
- W323718150 hasRelatedWork W1700124960 @default.
- W323718150 hasRelatedWork W1974455634 @default.
- W323718150 hasRelatedWork W2028908713 @default.
- W323718150 hasRelatedWork W2128176022 @default.
- W323718150 hasRelatedWork W2239847623 @default.
- W323718150 hasRelatedWork W2403731734 @default.
- W323718150 hasRelatedWork W2520176975 @default.
- W323718150 hasRelatedWork W2548256801 @default.
- W323718150 hasRelatedWork W2587675110 @default.
- W323718150 hasRelatedWork W2890197052 @default.
- W323718150 hasRelatedWork W2953937146 @default.
- W323718150 hasRelatedWork W2963057973 @default.
- W323718150 hasRelatedWork W2972692349 @default.
- W323718150 hasRelatedWork W2992448548 @default.
- W323718150 hasRelatedWork W3163339532 @default.
- W323718150 hasRelatedWork W3205404577 @default.
- W323718150 isParatext "false" @default.
- W323718150 isRetracted "false" @default.
- W323718150 magId "323718150" @default.
- W323718150 workType "article" @default.