Matches in SemOpenAlex for { <https://semopenalex.org/work/W207489108> ?p ?o ?g. }
Showing items 1 to 72 of
72
with 100 items per page.
- W207489108 abstract "Discriminative training applied to hidden Markov model (HMM) design can yield significant benefits in recognition accuracy and model compactness. However, compared to Maximum Likelihood based methods, discriminative training typically requires much more computation, as all competing candidates must be considered, not just the correct one. The choice of the algorithm used to optimize the discriminative criterion function is thus a key issue. We investigated several algorithms and used them for discriminative training based on the Minimum Classification Error (MCE) framework. In particular, we examined on-line, batch, and semi-batch Probabilistic Descent (PD), as well as Quickprop, Rprop and BFGS.We describe each algorithm and present comparative results on the TIMIT phone classification task and on the 230 hour Corpus of Spontaneous Japanese (CSJ) 30K word continuous speech recognition task." @default.
- W207489108 created "2016-06-24" @default.
- W207489108 creator A5030109648 @default.
- W207489108 creator A5076453358 @default.
- W207489108 date "2005-09-04" @default.
- W207489108 modified "2023-10-14" @default.
- W207489108 title "Optimization methods for discriminative training" @default.
- W207489108 cites W1533051419 @default.
- W207489108 cites W1541007220 @default.
- W207489108 cites W1972472467 @default.
- W207489108 cites W2003123121 @default.
- W207489108 cites W2006544565 @default.
- W207489108 cites W2010581677 @default.
- W207489108 cites W2051669046 @default.
- W207489108 cites W2084236346 @default.
- W207489108 cites W2110871230 @default.
- W207489108 cites W2115602126 @default.
- W207489108 cites W2116196769 @default.
- W207489108 cites W2134950286 @default.
- W207489108 cites W2143908786 @default.
- W207489108 cites W2162432783 @default.
- W207489108 cites W87954838 @default.
- W207489108 doi "https://doi.org/10.21437/interspeech.2005-858" @default.
- W207489108 hasPublicationYear "2005" @default.
- W207489108 type Work @default.
- W207489108 sameAs 207489108 @default.
- W207489108 citedByCount "25" @default.
- W207489108 countsByYear W2074891082012 @default.
- W207489108 countsByYear W2074891082013 @default.
- W207489108 countsByYear W2074891082014 @default.
- W207489108 crossrefType "proceedings-article" @default.
- W207489108 hasAuthorship W207489108A5030109648 @default.
- W207489108 hasAuthorship W207489108A5076453358 @default.
- W207489108 hasConcept C119857082 @default.
- W207489108 hasConcept C127413603 @default.
- W207489108 hasConcept C153180895 @default.
- W207489108 hasConcept C154945302 @default.
- W207489108 hasConcept C201995342 @default.
- W207489108 hasConcept C23224414 @default.
- W207489108 hasConcept C2778724510 @default.
- W207489108 hasConcept C2780451532 @default.
- W207489108 hasConcept C28490314 @default.
- W207489108 hasConcept C41008148 @default.
- W207489108 hasConcept C97931131 @default.
- W207489108 hasConceptScore W207489108C119857082 @default.
- W207489108 hasConceptScore W207489108C127413603 @default.
- W207489108 hasConceptScore W207489108C153180895 @default.
- W207489108 hasConceptScore W207489108C154945302 @default.
- W207489108 hasConceptScore W207489108C201995342 @default.
- W207489108 hasConceptScore W207489108C23224414 @default.
- W207489108 hasConceptScore W207489108C2778724510 @default.
- W207489108 hasConceptScore W207489108C2780451532 @default.
- W207489108 hasConceptScore W207489108C28490314 @default.
- W207489108 hasConceptScore W207489108C41008148 @default.
- W207489108 hasConceptScore W207489108C97931131 @default.
- W207489108 hasLocation W2074891081 @default.
- W207489108 hasOpenAccess W207489108 @default.
- W207489108 hasPrimaryLocation W2074891081 @default.
- W207489108 hasRelatedWork W1528058524 @default.
- W207489108 hasRelatedWork W1652783584 @default.
- W207489108 hasRelatedWork W2024160000 @default.
- W207489108 hasRelatedWork W2104535716 @default.
- W207489108 hasRelatedWork W2136516428 @default.
- W207489108 hasRelatedWork W2285052147 @default.
- W207489108 hasRelatedWork W2404514746 @default.
- W207489108 hasRelatedWork W2729514902 @default.
- W207489108 hasRelatedWork W4287995534 @default.
- W207489108 hasRelatedWork W74886973 @default.
- W207489108 isParatext "false" @default.
- W207489108 isRetracted "false" @default.
- W207489108 magId "207489108" @default.
- W207489108 workType "article" @default.