Matches in SemOpenAlex for { <https://semopenalex.org/work/W2982608933> ?p ?o ?g. }
Showing items 1 to 96 of
96
with 100 items per page.
- W2982608933 abstract "Sequence-to-Sequence (S2S) models recently started to show state-of-the-art performance for automatic speech recognition (ASR). With these large and deep models overfitting remains the largest problem, outweighing performance improvements that can be obtained from better architectures. One solution to the overfitting problem is increasing the amount of available training data and the variety exhibited by the training data with the help of data augmentation. In this paper we examine the influence of three data augmentation methods on the performance of two S2S model architectures. One of the data augmentation method comes from literature, while two other methods are our own development - a time perturbation in the frequency domain and sub-sequence sampling. Our experiments on Switchboard and Fisher data show state-of-the-art performance for S2S models that are trained solely on the speech training data and do not use additional text data." @default.
- W2982608933 created "2019-11-08" @default.
- W2982608933 creator A5004861130 @default.
- W2982608933 creator A5007526409 @default.
- W2982608933 creator A5046084081 @default.
- W2982608933 creator A5082064946 @default.
- W2982608933 date "2019-10-29" @default.
- W2982608933 modified "2023-09-27" @default.
- W2982608933 title "Improving sequence-to-sequence speech recognition training with on-the-fly data augmentation" @default.
- W2982608933 cites W1855892484 @default.
- W2982608933 cites W1922655562 @default.
- W2982608933 cites W1989674786 @default.
- W2982608933 cites W2095705004 @default.
- W2982608933 cites W2099621636 @default.
- W2982608933 cites W2184343439 @default.
- W2982608933 cites W2327501763 @default.
- W2982608933 cites W2407080277 @default.
- W2982608933 cites W2514741789 @default.
- W2982608933 cites W2696967604 @default.
- W2982608933 cites W2782451907 @default.
- W2982608933 cites W2799800213 @default.
- W2982608933 cites W2889163603 @default.
- W2982608933 cites W2936774411 @default.
- W2982608933 cites W2962824709 @default.
- W2982608933 cites W2963070863 @default.
- W2982608933 cites W2963266252 @default.
- W2982608933 cites W2963403868 @default.
- W2982608933 cites W2964121744 @default.
- W2982608933 cites W2972451902 @default.
- W2982608933 cites W2973040747 @default.
- W2982608933 cites W854541894 @default.
- W2982608933 hasPublicationYear "2019" @default.
- W2982608933 type Work @default.
- W2982608933 sameAs 2982608933 @default.
- W2982608933 citedByCount "1" @default.
- W2982608933 countsByYear W29826089332019 @default.
- W2982608933 crossrefType "posted-content" @default.
- W2982608933 hasAuthorship W2982608933A5004861130 @default.
- W2982608933 hasAuthorship W2982608933A5007526409 @default.
- W2982608933 hasAuthorship W2982608933A5046084081 @default.
- W2982608933 hasAuthorship W2982608933A5082064946 @default.
- W2982608933 hasConcept C11413529 @default.
- W2982608933 hasConcept C119857082 @default.
- W2982608933 hasConcept C136197465 @default.
- W2982608933 hasConcept C153180895 @default.
- W2982608933 hasConcept C154945302 @default.
- W2982608933 hasConcept C22019652 @default.
- W2982608933 hasConcept C2778112365 @default.
- W2982608933 hasConcept C28490314 @default.
- W2982608933 hasConcept C41008148 @default.
- W2982608933 hasConcept C48103436 @default.
- W2982608933 hasConcept C50644808 @default.
- W2982608933 hasConcept C51632099 @default.
- W2982608933 hasConcept C54355233 @default.
- W2982608933 hasConcept C86803240 @default.
- W2982608933 hasConceptScore W2982608933C11413529 @default.
- W2982608933 hasConceptScore W2982608933C119857082 @default.
- W2982608933 hasConceptScore W2982608933C136197465 @default.
- W2982608933 hasConceptScore W2982608933C153180895 @default.
- W2982608933 hasConceptScore W2982608933C154945302 @default.
- W2982608933 hasConceptScore W2982608933C22019652 @default.
- W2982608933 hasConceptScore W2982608933C2778112365 @default.
- W2982608933 hasConceptScore W2982608933C28490314 @default.
- W2982608933 hasConceptScore W2982608933C41008148 @default.
- W2982608933 hasConceptScore W2982608933C48103436 @default.
- W2982608933 hasConceptScore W2982608933C50644808 @default.
- W2982608933 hasConceptScore W2982608933C51632099 @default.
- W2982608933 hasConceptScore W2982608933C54355233 @default.
- W2982608933 hasConceptScore W2982608933C86803240 @default.
- W2982608933 hasLocation W29826089331 @default.
- W2982608933 hasOpenAccess W2982608933 @default.
- W2982608933 hasPrimaryLocation W29826089331 @default.
- W2982608933 hasRelatedWork W1495019370 @default.
- W2982608933 hasRelatedWork W2572023240 @default.
- W2982608933 hasRelatedWork W2737108017 @default.
- W2982608933 hasRelatedWork W2766572290 @default.
- W2982608933 hasRelatedWork W2775304348 @default.
- W2982608933 hasRelatedWork W2776984718 @default.
- W2982608933 hasRelatedWork W2793863914 @default.
- W2982608933 hasRelatedWork W2886093827 @default.
- W2982608933 hasRelatedWork W2891541684 @default.
- W2982608933 hasRelatedWork W2900312178 @default.
- W2982608933 hasRelatedWork W2922358226 @default.
- W2982608933 hasRelatedWork W2953598317 @default.
- W2982608933 hasRelatedWork W2972981541 @default.
- W2982608933 hasRelatedWork W3097573669 @default.
- W2982608933 hasRelatedWork W3118753411 @default.
- W2982608933 hasRelatedWork W3144810982 @default.
- W2982608933 hasRelatedWork W3155343771 @default.
- W2982608933 hasRelatedWork W3165957370 @default.
- W2982608933 hasRelatedWork W3168872064 @default.
- W2982608933 hasRelatedWork W3182245104 @default.
- W2982608933 isParatext "false" @default.
- W2982608933 isRetracted "false" @default.
- W2982608933 magId "2982608933" @default.
- W2982608933 workType "article" @default.