Matches in SemOpenAlex for { <https://semopenalex.org/work/W2097560587> ?p ?o ?g. }
- W2097560587 endingPage "1744" @default.
- W2097560587 startingPage "1729" @default.
- W2097560587 abstract "We summarize recent progress in automatic speech-to-text transcription at SRI, ICSI, and the University of Washington. The work encompasses all components of speech modeling found in a state-of-the-art recognition system, from acoustic features, to acoustic modeling and adaptation, to language modeling. In the front end, we experimented with nonstandard features, including various measures of voicing, discriminative phone posterior features estimated by multilayer perceptrons, and a novel phone-level macro-averaging for cepstral normalization. Acoustic modeling was improved with combinations of front ends operating at multiple frame rates, as well as by modifications to the standard methods for discriminative Gaussian estimation. We show that acoustic adaptation can be improved by predicting the optimal regression class complexity for a given speaker. Language modeling innovations include the use of a syntax-motivated almost-parsing language model, as well as principled vocabulary-selection techniques. Finally, we address portability issues, such as the use of imperfect training transcripts, and language-specific adjustments required for recognition of Arabic and Mandarin" @default.
- W2097560587 created "2016-06-24" @default.
- W2097560587 creator A5014229661 @default.
- W2097560587 creator A5023103090 @default.
- W2097560587 creator A5025265147 @default.
- W2097560587 creator A5030660132 @default.
- W2097560587 creator A5035394016 @default.
- W2097560587 creator A5038072013 @default.
- W2097560587 creator A5046717231 @default.
- W2097560587 creator A5049242358 @default.
- W2097560587 creator A5049457213 @default.
- W2097560587 creator A5050325468 @default.
- W2097560587 creator A5059342552 @default.
- W2097560587 creator A5060715364 @default.
- W2097560587 creator A5060979948 @default.
- W2097560587 creator A5065509001 @default.
- W2097560587 creator A5067651004 @default.
- W2097560587 creator A5087215613 @default.
- W2097560587 creator A5087698068 @default.
- W2097560587 creator A5088966231 @default.
- W2097560587 date "2006-09-01" @default.
- W2097560587 modified "2023-10-17" @default.
- W2097560587 title "Recent innovations in speech-to-text transcription at SRI-ICSI-UW" @default.
- W2097560587 cites W1877570817 @default.
- W2097560587 cites W2003123121 @default.
- W2097560587 cites W2050693797 @default.
- W2097560587 cites W2056250865 @default.
- W2097560587 cites W2069976350 @default.
- W2097560587 cites W2090861223 @default.
- W2097560587 cites W2094971681 @default.
- W2097560587 cites W2103621378 @default.
- W2097560587 cites W2104663520 @default.
- W2097560587 cites W2105830342 @default.
- W2097560587 cites W2106232466 @default.
- W2097560587 cites W2110372834 @default.
- W2097560587 cites W2113094193 @default.
- W2097560587 cites W2123876776 @default.
- W2097560587 cites W2126948708 @default.
- W2097560587 cites W2137075158 @default.
- W2097560587 cites W2144810223 @default.
- W2097560587 cites W2146871184 @default.
- W2097560587 cites W2158069733 @default.
- W2097560587 cites W2161513525 @default.
- W2097560587 cites W2167763959 @default.
- W2097560587 cites W2168818382 @default.
- W2097560587 cites W2169998051 @default.
- W2097560587 cites W2594610113 @default.
- W2097560587 doi "https://doi.org/10.1109/tasl.2006.879807" @default.
- W2097560587 hasPublicationYear "2006" @default.
- W2097560587 type Work @default.
- W2097560587 sameAs 2097560587 @default.
- W2097560587 citedByCount "79" @default.
- W2097560587 countsByYear W20975605872012 @default.
- W2097560587 countsByYear W20975605872013 @default.
- W2097560587 countsByYear W20975605872014 @default.
- W2097560587 countsByYear W20975605872015 @default.
- W2097560587 countsByYear W20975605872016 @default.
- W2097560587 countsByYear W20975605872017 @default.
- W2097560587 countsByYear W20975605872018 @default.
- W2097560587 countsByYear W20975605872019 @default.
- W2097560587 countsByYear W20975605872022 @default.
- W2097560587 countsByYear W20975605872023 @default.
- W2097560587 crossrefType "journal-article" @default.
- W2097560587 hasAuthorship W2097560587A5014229661 @default.
- W2097560587 hasAuthorship W2097560587A5023103090 @default.
- W2097560587 hasAuthorship W2097560587A5025265147 @default.
- W2097560587 hasAuthorship W2097560587A5030660132 @default.
- W2097560587 hasAuthorship W2097560587A5035394016 @default.
- W2097560587 hasAuthorship W2097560587A5038072013 @default.
- W2097560587 hasAuthorship W2097560587A5046717231 @default.
- W2097560587 hasAuthorship W2097560587A5049242358 @default.
- W2097560587 hasAuthorship W2097560587A5049457213 @default.
- W2097560587 hasAuthorship W2097560587A5050325468 @default.
- W2097560587 hasAuthorship W2097560587A5059342552 @default.
- W2097560587 hasAuthorship W2097560587A5060715364 @default.
- W2097560587 hasAuthorship W2097560587A5060979948 @default.
- W2097560587 hasAuthorship W2097560587A5065509001 @default.
- W2097560587 hasAuthorship W2097560587A5067651004 @default.
- W2097560587 hasAuthorship W2097560587A5087215613 @default.
- W2097560587 hasAuthorship W2097560587A5087698068 @default.
- W2097560587 hasAuthorship W2097560587A5088966231 @default.
- W2097560587 hasConcept C136886441 @default.
- W2097560587 hasConcept C137293760 @default.
- W2097560587 hasConcept C138885662 @default.
- W2097560587 hasConcept C144024400 @default.
- W2097560587 hasConcept C154945302 @default.
- W2097560587 hasConcept C155635449 @default.
- W2097560587 hasConcept C179926584 @default.
- W2097560587 hasConcept C19165224 @default.
- W2097560587 hasConcept C199360897 @default.
- W2097560587 hasConcept C204321447 @default.
- W2097560587 hasConcept C23224414 @default.
- W2097560587 hasConcept C2778707766 @default.
- W2097560587 hasConcept C2781213101 @default.
- W2097560587 hasConcept C28490314 @default.
- W2097560587 hasConcept C40969351 @default.
- W2097560587 hasConcept C41008148 @default.
- W2097560587 hasConcept C41895202 @default.