Matches in SemOpenAlex for { <https://semopenalex.org/work/W1569402938> ?p ?o ?g. }
Showing items 1 to 93 of
93
with 100 items per page.
- W1569402938 endingPage "2161" @default.
- W1569402938 startingPage "2148" @default.
- W1569402938 abstract "The rise of mobile devices and online learning brings into sharp focus the importance of speech recognition not only for the many languages of the world but also for code-mixed speech, especially where English is the second language. The recognition of code-mixed speech, where the speaker mixes languages within a single utterance, is a challenge for both computers and humans, not least because of the limited training data. We conduct research on a Mandarin-English code-mixed lecture corpus, where Mandarin is the host language and English the guest language, and attempt to find complex features for the recovery of English segments that were misrecognized in the initial recognition pass. We propose a multi-level framework wherein both low-level and high-level cues are jointly considered; we use phonotactic, prosodic, and linguistic cues in addition to acoustic-phonetic cues to discriminate at the frame level between English- and Chinese-language segments. We develop a simple and exact method for CRF feature induction, and improved methods for using cascaded features derived from the training corpus. By additionally tuning the data imbalance ratio between English and Chinese, we demonstrate highly significant improvements over previous work in the recovery of English-language segments, and demonstrate performance superior to DNN-based methods. We demonstrate considerable performance improvements not only with the traditional GMM-HMM recognition paradigm but also with a state-of-the-art hybrid CD-HMM-DNN recognition framework." @default.
- W1569402938 created "2016-06-24" @default.
- W1569402938 creator A5000293651 @default.
- W1569402938 creator A5044010123 @default.
- W1569402938 creator A5080149033 @default.
- W1569402938 date "2015-12-01" @default.
- W1569402938 modified "2023-09-25" @default.
- W1569402938 title "Finding Complex Features for Guest Language Fragment Recovery in Resource-Limited Code-Mixed Speech Recognition" @default.
- W1569402938 cites W1606054028 @default.
- W1569402938 cites W179875071 @default.
- W1569402938 cites W1815792 @default.
- W1569402938 cites W1965154800 @default.
- W1569402938 cites W1993482042 @default.
- W1569402938 cites W1996034123 @default.
- W1569402938 cites W1999965501 @default.
- W1569402938 cites W2026782514 @default.
- W1569402938 cites W2037127751 @default.
- W1569402938 cites W2062164080 @default.
- W1569402938 cites W2076710288 @default.
- W1569402938 cites W2081209971 @default.
- W1569402938 cites W2085365725 @default.
- W1569402938 cites W2087112010 @default.
- W1569402938 cites W2114569717 @default.
- W1569402938 cites W2119340951 @default.
- W1569402938 cites W2121141940 @default.
- W1569402938 cites W2123162330 @default.
- W1569402938 cites W2135704565 @default.
- W1569402938 cites W2136848157 @default.
- W1569402938 cites W2148931622 @default.
- W1569402938 cites W2160815625 @default.
- W1569402938 cites W2171928131 @default.
- W1569402938 cites W2172287020 @default.
- W1569402938 cites W2264970638 @default.
- W1569402938 cites W2398092513 @default.
- W1569402938 cites W2402268235 @default.
- W1569402938 cites W4214739450 @default.
- W1569402938 cites W58887963 @default.
- W1569402938 doi "https://doi.org/10.1109/taslp.2015.2469634" @default.
- W1569402938 hasPublicationYear "2015" @default.
- W1569402938 type Work @default.
- W1569402938 sameAs 1569402938 @default.
- W1569402938 citedByCount "2" @default.
- W1569402938 countsByYear W15694029382022 @default.
- W1569402938 countsByYear W15694029382023 @default.
- W1569402938 crossrefType "journal-article" @default.
- W1569402938 hasAuthorship W1569402938A5000293651 @default.
- W1569402938 hasAuthorship W1569402938A5044010123 @default.
- W1569402938 hasAuthorship W1569402938A5080149033 @default.
- W1569402938 hasConcept C138885662 @default.
- W1569402938 hasConcept C177264268 @default.
- W1569402938 hasConcept C18552078 @default.
- W1569402938 hasConcept C199360897 @default.
- W1569402938 hasConcept C204321447 @default.
- W1569402938 hasConcept C206345919 @default.
- W1569402938 hasConcept C2776235265 @default.
- W1569402938 hasConcept C2776760102 @default.
- W1569402938 hasConcept C28490314 @default.
- W1569402938 hasConcept C31258907 @default.
- W1569402938 hasConcept C41008148 @default.
- W1569402938 hasConcept C41895202 @default.
- W1569402938 hasConceptScore W1569402938C138885662 @default.
- W1569402938 hasConceptScore W1569402938C177264268 @default.
- W1569402938 hasConceptScore W1569402938C18552078 @default.
- W1569402938 hasConceptScore W1569402938C199360897 @default.
- W1569402938 hasConceptScore W1569402938C204321447 @default.
- W1569402938 hasConceptScore W1569402938C206345919 @default.
- W1569402938 hasConceptScore W1569402938C2776235265 @default.
- W1569402938 hasConceptScore W1569402938C2776760102 @default.
- W1569402938 hasConceptScore W1569402938C28490314 @default.
- W1569402938 hasConceptScore W1569402938C31258907 @default.
- W1569402938 hasConceptScore W1569402938C41008148 @default.
- W1569402938 hasConceptScore W1569402938C41895202 @default.
- W1569402938 hasIssue "12" @default.
- W1569402938 hasLocation W15694029381 @default.
- W1569402938 hasOpenAccess W1569402938 @default.
- W1569402938 hasPrimaryLocation W15694029381 @default.
- W1569402938 hasRelatedWork W2252156092 @default.
- W1569402938 hasRelatedWork W2611614995 @default.
- W1569402938 hasRelatedWork W2952018704 @default.
- W1569402938 hasRelatedWork W2964084554 @default.
- W1569402938 hasRelatedWork W2972943990 @default.
- W1569402938 hasRelatedWork W2998485758 @default.
- W1569402938 hasRelatedWork W3031778297 @default.
- W1569402938 hasRelatedWork W3104716243 @default.
- W1569402938 hasRelatedWork W3215337221 @default.
- W1569402938 hasRelatedWork W4288329983 @default.
- W1569402938 hasVolume "23" @default.
- W1569402938 isParatext "false" @default.
- W1569402938 isRetracted "false" @default.
- W1569402938 magId "1569402938" @default.
- W1569402938 workType "article" @default.