Matches in SemOpenAlex for { <https://semopenalex.org/work/W2893932660> ?p ?o ?g. }
Showing items 1 to 87 of
87
with 100 items per page.
- W2893932660 abstract "Voice Dialogue Applications(VDAs) increase popularity nowadays. As the same sentence expressed with different emotion may convey different meanings, inferring emotion from users' queries can help give a more humanized response for VDAs. However, the large-scale Internet voice data involving a tremendous amount of users, bring in a great diversity of users' dialects and expression preferences. Therefore, the traditional speech emotion recognition methods mainly targeting at acted corpora cannot handle the massive and diverse data effectively. In this paper, we propose a semi-supervised Emotion-oriented Bimodal Deep Autoencoder (EBDA) to infer emotion from large-scale Internet voice data. Specifically, as the previous research mainly focuses on acoustic features only, we utilize EBDA to fully integrate both acoustic and textual features. Meanwhile, to employ large-scale unlabeled data to enhance the classification performance, we adopt a semi-supervised strategy. The experimental results on 6 emotion categories based on a dataset collected from Sogou Voice Assistant <sup xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink>1</sup> containing 7.5 million utterances outperform several alternative baselines (+0.18% in terms of F1 on average). Finally, we show some interesting case studies to further demonstrate the practicability of our model." @default.
- W2893932660 created "2018-10-05" @default.
- W2893932660 creator A5012385922 @default.
- W2893932660 creator A5017193282 @default.
- W2893932660 creator A5017541508 @default.
- W2893932660 creator A5022832205 @default.
- W2893932660 creator A5051489725 @default.
- W2893932660 creator A5054899321 @default.
- W2893932660 creator A5071074705 @default.
- W2893932660 date "2018-05-01" @default.
- W2893932660 modified "2023-09-26" @default.
- W2893932660 title "Emotion Inferring from Large-scale Internet Voice Data: A Multimodal Deep Learning Approach" @default.
- W2893932660 cites W114517082 @default.
- W2893932660 cites W175750906 @default.
- W2893932660 cites W1895199447 @default.
- W2893932660 cites W2137409775 @default.
- W2893932660 cites W2146334809 @default.
- W2893932660 cites W2152349600 @default.
- W2893932660 cites W2153635508 @default.
- W2893932660 cites W2161073241 @default.
- W2893932660 cites W2327261103 @default.
- W2893932660 cites W4236796448 @default.
- W2893932660 cites W4249972823 @default.
- W2893932660 doi "https://doi.org/10.1109/aciiasia.2018.8470311" @default.
- W2893932660 hasPublicationYear "2018" @default.
- W2893932660 type Work @default.
- W2893932660 sameAs 2893932660 @default.
- W2893932660 citedByCount "5" @default.
- W2893932660 countsByYear W28939326602020 @default.
- W2893932660 countsByYear W28939326602021 @default.
- W2893932660 countsByYear W28939326602022 @default.
- W2893932660 countsByYear W28939326602023 @default.
- W2893932660 crossrefType "proceedings-article" @default.
- W2893932660 hasAuthorship W2893932660A5012385922 @default.
- W2893932660 hasAuthorship W2893932660A5017193282 @default.
- W2893932660 hasAuthorship W2893932660A5017541508 @default.
- W2893932660 hasAuthorship W2893932660A5022832205 @default.
- W2893932660 hasAuthorship W2893932660A5051489725 @default.
- W2893932660 hasAuthorship W2893932660A5054899321 @default.
- W2893932660 hasAuthorship W2893932660A5071074705 @default.
- W2893932660 hasConcept C101738243 @default.
- W2893932660 hasConcept C108583219 @default.
- W2893932660 hasConcept C110875604 @default.
- W2893932660 hasConcept C121332964 @default.
- W2893932660 hasConcept C136764020 @default.
- W2893932660 hasConcept C154945302 @default.
- W2893932660 hasConcept C15744967 @default.
- W2893932660 hasConcept C204321447 @default.
- W2893932660 hasConcept C2777530160 @default.
- W2893932660 hasConcept C2778755073 @default.
- W2893932660 hasConcept C2780586970 @default.
- W2893932660 hasConcept C28490314 @default.
- W2893932660 hasConcept C41008148 @default.
- W2893932660 hasConcept C62520636 @default.
- W2893932660 hasConcept C77805123 @default.
- W2893932660 hasConceptScore W2893932660C101738243 @default.
- W2893932660 hasConceptScore W2893932660C108583219 @default.
- W2893932660 hasConceptScore W2893932660C110875604 @default.
- W2893932660 hasConceptScore W2893932660C121332964 @default.
- W2893932660 hasConceptScore W2893932660C136764020 @default.
- W2893932660 hasConceptScore W2893932660C154945302 @default.
- W2893932660 hasConceptScore W2893932660C15744967 @default.
- W2893932660 hasConceptScore W2893932660C204321447 @default.
- W2893932660 hasConceptScore W2893932660C2777530160 @default.
- W2893932660 hasConceptScore W2893932660C2778755073 @default.
- W2893932660 hasConceptScore W2893932660C2780586970 @default.
- W2893932660 hasConceptScore W2893932660C28490314 @default.
- W2893932660 hasConceptScore W2893932660C41008148 @default.
- W2893932660 hasConceptScore W2893932660C62520636 @default.
- W2893932660 hasConceptScore W2893932660C77805123 @default.
- W2893932660 hasLocation W28939326601 @default.
- W2893932660 hasOpenAccess W2893932660 @default.
- W2893932660 hasPrimaryLocation W28939326601 @default.
- W2893932660 hasRelatedWork W2669956259 @default.
- W2893932660 hasRelatedWork W2939353110 @default.
- W2893932660 hasRelatedWork W3165097609 @default.
- W2893932660 hasRelatedWork W3165463024 @default.
- W2893932660 hasRelatedWork W3209662401 @default.
- W2893932660 hasRelatedWork W4220775285 @default.
- W2893932660 hasRelatedWork W4287178339 @default.
- W2893932660 hasRelatedWork W4292874285 @default.
- W2893932660 hasRelatedWork W4310034804 @default.
- W2893932660 hasRelatedWork W4327774331 @default.
- W2893932660 isParatext "false" @default.
- W2893932660 isRetracted "false" @default.
- W2893932660 magId "2893932660" @default.
- W2893932660 workType "article" @default.