Matches in SemOpenAlex for { <https://semopenalex.org/work/W2005232502> ?p ?o ?g. }
- W2005232502 abstract "For low resource languages, collecting sufficient training data to build acoustic and language models is time consuming and often expensive. But large amounts of text data, such as online newspapers, web forums or online encyclopedias, usually exist for languages that have a large population of native speakers. This text data can be easily collected from the web and then used to both expand the recognizer's vocabulary and improve the language model. One challenge, however, is normalizing and filtering the web data for a specific task. In this paper, we investigate the use of online text resources to improve the performance of speech recognition specifically for the task of keyword spotting. For the five languages provided in the base period of the IARPA BABEL project, we automatically collected text data from the web using only Limited LP resources. We then compared two methods for filtering the web data, one based on perplexity ranking and the other based on out-of-vocabulary (OOV) word detection. By integrating the web text into our systems, we observed significant improvements in keyword spotting accuracy for four out of the five languages. The best approach obtained an improvement in actual term weighted value (ATWV) of 0.0424 compared to a baseline system trained only on LimitedLP resources. On average, ATWV was improved by 0.0243 across five languages." @default.
- W2005232502 created "2016-06-24" @default.
- W2005232502 creator A5014054572 @default.
- W2005232502 creator A5028182466 @default.
- W2005232502 creator A5036273624 @default.
- W2005232502 creator A5040668817 @default.
- W2005232502 creator A5064290493 @default.
- W2005232502 creator A5085262529 @default.
- W2005232502 date "2013-12-01" @default.
- W2005232502 modified "2023-09-23" @default.
- W2005232502 title "Using web text to improve keyword spotting in speech" @default.
- W2005232502 cites W1480662687 @default.
- W2005232502 cites W161909597 @default.
- W2005232502 cites W1631260214 @default.
- W2005232502 cites W1733633583 @default.
- W2005232502 cites W1970026646 @default.
- W2005232502 cites W2018959378 @default.
- W2005232502 cites W2090755665 @default.
- W2005232502 cites W2094971681 @default.
- W2005232502 cites W2105830342 @default.
- W2005232502 cites W2117257534 @default.
- W2005232502 cites W2132780103 @default.
- W2005232502 cites W2133882397 @default.
- W2005232502 cites W2136780519 @default.
- W2005232502 cites W2154304724 @default.
- W2005232502 cites W2155388323 @default.
- W2005232502 cites W2164666834 @default.
- W2005232502 cites W3143499733 @default.
- W2005232502 cites W626798658 @default.
- W2005232502 doi "https://doi.org/10.1109/asru.2013.6707768" @default.
- W2005232502 hasPublicationYear "2013" @default.
- W2005232502 type Work @default.
- W2005232502 sameAs 2005232502 @default.
- W2005232502 citedByCount "12" @default.
- W2005232502 countsByYear W20052325022014 @default.
- W2005232502 countsByYear W20052325022015 @default.
- W2005232502 countsByYear W20052325022016 @default.
- W2005232502 countsByYear W20052325022017 @default.
- W2005232502 countsByYear W20052325022018 @default.
- W2005232502 countsByYear W20052325022019 @default.
- W2005232502 countsByYear W20052325022020 @default.
- W2005232502 crossrefType "proceedings-article" @default.
- W2005232502 hasAuthorship W2005232502A5014054572 @default.
- W2005232502 hasAuthorship W2005232502A5028182466 @default.
- W2005232502 hasAuthorship W2005232502A5036273624 @default.
- W2005232502 hasAuthorship W2005232502A5040668817 @default.
- W2005232502 hasAuthorship W2005232502A5064290493 @default.
- W2005232502 hasAuthorship W2005232502A5085262529 @default.
- W2005232502 hasBestOaLocation W20052325022 @default.
- W2005232502 hasConcept C100279451 @default.
- W2005232502 hasConcept C137293760 @default.
- W2005232502 hasConcept C138885662 @default.
- W2005232502 hasConcept C144024400 @default.
- W2005232502 hasConcept C148863701 @default.
- W2005232502 hasConcept C149923435 @default.
- W2005232502 hasConcept C154945302 @default.
- W2005232502 hasConcept C161191863 @default.
- W2005232502 hasConcept C162324750 @default.
- W2005232502 hasConcept C187736073 @default.
- W2005232502 hasConcept C204321447 @default.
- W2005232502 hasConcept C23123220 @default.
- W2005232502 hasConcept C2777601683 @default.
- W2005232502 hasConcept C2780451532 @default.
- W2005232502 hasConcept C2781213101 @default.
- W2005232502 hasConcept C2908647359 @default.
- W2005232502 hasConcept C41008148 @default.
- W2005232502 hasConcept C41895202 @default.
- W2005232502 hasConcept C90805587 @default.
- W2005232502 hasConceptScore W2005232502C100279451 @default.
- W2005232502 hasConceptScore W2005232502C137293760 @default.
- W2005232502 hasConceptScore W2005232502C138885662 @default.
- W2005232502 hasConceptScore W2005232502C144024400 @default.
- W2005232502 hasConceptScore W2005232502C148863701 @default.
- W2005232502 hasConceptScore W2005232502C149923435 @default.
- W2005232502 hasConceptScore W2005232502C154945302 @default.
- W2005232502 hasConceptScore W2005232502C161191863 @default.
- W2005232502 hasConceptScore W2005232502C162324750 @default.
- W2005232502 hasConceptScore W2005232502C187736073 @default.
- W2005232502 hasConceptScore W2005232502C204321447 @default.
- W2005232502 hasConceptScore W2005232502C23123220 @default.
- W2005232502 hasConceptScore W2005232502C2777601683 @default.
- W2005232502 hasConceptScore W2005232502C2780451532 @default.
- W2005232502 hasConceptScore W2005232502C2781213101 @default.
- W2005232502 hasConceptScore W2005232502C2908647359 @default.
- W2005232502 hasConceptScore W2005232502C41008148 @default.
- W2005232502 hasConceptScore W2005232502C41895202 @default.
- W2005232502 hasConceptScore W2005232502C90805587 @default.
- W2005232502 hasLocation W20052325021 @default.
- W2005232502 hasLocation W20052325022 @default.
- W2005232502 hasOpenAccess W2005232502 @default.
- W2005232502 hasPrimaryLocation W20052325021 @default.
- W2005232502 hasRelatedWork W2000654171 @default.
- W2005232502 hasRelatedWork W2005708814 @default.
- W2005232502 hasRelatedWork W2018959378 @default.
- W2005232502 hasRelatedWork W2034243835 @default.
- W2005232502 hasRelatedWork W2090755665 @default.
- W2005232502 hasRelatedWork W2136803790 @default.
- W2005232502 hasRelatedWork W2146688305 @default.
- W2005232502 hasRelatedWork W2167338739 @default.
- W2005232502 hasRelatedWork W2288712388 @default.