Matches in SemOpenAlex for { <https://semopenalex.org/work/W45838861> ?p ?o ?g. }
Showing items 1 to 69 of
69
with 100 items per page.
- W45838861 abstract "In this work we compare two parameter optimization techniques for discriminative training using the MMI criterion: the extended Baum-Welch (EBW) algorithm and the generalized probabilistic descent (GPD) method. Using Gaussian emission densities we found special expressions for the step sizes in GPD, leading to reestimation formula very similar to those derived for the EBW algorithm. Results were produced for both the TI digitstring and the SieTill corpus for continuously spoken American English and German digitstrings. The results for both techniques do not show significant differences. This experimental results support the strong link between EBW and GPD as expected from the analytic comparison." @default.
- W45838861 created "2016-06-24" @default.
- W45838861 creator A5018646392 @default.
- W45838861 creator A5037867732 @default.
- W45838861 creator A5074275406 @default.
- W45838861 creator A5087367411 @default.
- W45838861 creator A5088968292 @default.
- W45838861 date "1997-09-22" @default.
- W45838861 modified "2023-09-26" @default.
- W45838861 title "Comparison of optimization methods for discriminative training criteria" @default.
- W45838861 cites W1565453839 @default.
- W45838861 cites W1574530145 @default.
- W45838861 cites W1590403649 @default.
- W45838861 cites W1920923476 @default.
- W45838861 cites W2077574412 @default.
- W45838861 cites W2098601596 @default.
- W45838861 cites W2115439799 @default.
- W45838861 cites W2116658213 @default.
- W45838861 cites W2152131029 @default.
- W45838861 cites W2167307239 @default.
- W45838861 cites W2171733396 @default.
- W45838861 cites W2395140404 @default.
- W45838861 cites W251778855 @default.
- W45838861 cites W92824239 @default.
- W45838861 doi "https://doi.org/10.21437/eurospeech.1997-10" @default.
- W45838861 hasPublicationYear "1997" @default.
- W45838861 type Work @default.
- W45838861 sameAs 45838861 @default.
- W45838861 citedByCount "17" @default.
- W45838861 countsByYear W458388612012 @default.
- W45838861 crossrefType "proceedings-article" @default.
- W45838861 hasAuthorship W45838861A5018646392 @default.
- W45838861 hasAuthorship W45838861A5037867732 @default.
- W45838861 hasAuthorship W45838861A5074275406 @default.
- W45838861 hasAuthorship W45838861A5087367411 @default.
- W45838861 hasAuthorship W45838861A5088968292 @default.
- W45838861 hasBestOaLocation W458388612 @default.
- W45838861 hasConcept C119857082 @default.
- W45838861 hasConcept C121332964 @default.
- W45838861 hasConcept C153294291 @default.
- W45838861 hasConcept C154945302 @default.
- W45838861 hasConcept C2777211547 @default.
- W45838861 hasConcept C41008148 @default.
- W45838861 hasConcept C97931131 @default.
- W45838861 hasConceptScore W45838861C119857082 @default.
- W45838861 hasConceptScore W45838861C121332964 @default.
- W45838861 hasConceptScore W45838861C153294291 @default.
- W45838861 hasConceptScore W45838861C154945302 @default.
- W45838861 hasConceptScore W45838861C2777211547 @default.
- W45838861 hasConceptScore W45838861C41008148 @default.
- W45838861 hasConceptScore W45838861C97931131 @default.
- W45838861 hasLocation W458388611 @default.
- W45838861 hasLocation W458388612 @default.
- W45838861 hasOpenAccess W45838861 @default.
- W45838861 hasPrimaryLocation W458388611 @default.
- W45838861 hasRelatedWork W1960490432 @default.
- W45838861 hasRelatedWork W2026121273 @default.
- W45838861 hasRelatedWork W2102106825 @default.
- W45838861 hasRelatedWork W2152579687 @default.
- W45838861 hasRelatedWork W2353457699 @default.
- W45838861 hasRelatedWork W2801772698 @default.
- W45838861 hasRelatedWork W2961085424 @default.
- W45838861 hasRelatedWork W2983744209 @default.
- W45838861 hasRelatedWork W3103844505 @default.
- W45838861 hasRelatedWork W66955737 @default.
- W45838861 isParatext "false" @default.
- W45838861 isRetracted "false" @default.
- W45838861 magId "45838861" @default.
- W45838861 workType "article" @default.