Matches in SemOpenAlex for { <https://semopenalex.org/work/W4224903891> ?p ?o ?g. }
- W4224903891 abstract "Turn-taking, aiming to decide when the next speaker can start talking, is an essential component in building human-robot spoken dialogue systems. Previous studies indicate that multi-modal cues can facilitate this challenging task. However, due to the paucity of public multimodal datasets, current methods are mostly limited to either utilizing unimodal features or simplistic multimodal ensemble models. Besides, the inherent class imbalance in real scenario, e.g. sentence ending with short pause will be mostly regarded as the end of turn, also poses great challenge to the turn-taking decision. In this paper, we first collect a large-scale annotated corpus for turn-taking with over 5,000 real human-robot dialogues in speech and text modalities. Then, a novel gated multimodal fusion mechanism is devised to utilize various information seamlessly for turn-taking prediction. More importantly, to tackle the data imbalance issue, we design a simple yet effective data augmentation method to construct negative instances without supervision and apply contrastive learning to obtain better feature representations. Extensive experiments are conducted and the results demonstrate the superiority and competitiveness of our model over several state-of-the-art baselines." @default.
- W4224903891 created "2022-04-27" @default.
- W4224903891 creator A5003185739 @default.
- W4224903891 creator A5024006251 @default.
- W4224903891 creator A5024845093 @default.
- W4224903891 creator A5039641921 @default.
- W4224903891 creator A5048159038 @default.
- W4224903891 creator A5076923788 @default.
- W4224903891 date "2022-05-23" @default.
- W4224903891 modified "2023-10-18" @default.
- W4224903891 title "Gated Multimodal Fusion with Contrastive Learning for Turn-Taking Prediction in Human-Robot Dialogue" @default.
- W4224903891 cites W1964725106 @default.
- W4224903891 cites W2008741806 @default.
- W4224903891 cites W2085662862 @default.
- W4224903891 cites W2194775991 @default.
- W4224903891 cites W2250891614 @default.
- W4224903891 cites W2748406667 @default.
- W4224903891 cites W2786387151 @default.
- W4224903891 cites W2889445100 @default.
- W4224903891 cites W2963093689 @default.
- W4224903891 cites W2963150162 @default.
- W4224903891 cites W2963747517 @default.
- W4224903891 cites W2973109987 @default.
- W4224903891 cites W2973135621 @default.
- W4224903891 cites W2973137193 @default.
- W4224903891 cites W3102393842 @default.
- W4224903891 cites W316215934 @default.
- W4224903891 cites W2972438655 @default.
- W4224903891 cites W2973172454 @default.
- W4224903891 doi "https://doi.org/10.1109/icassp43922.2022.9747056" @default.
- W4224903891 hasPublicationYear "2022" @default.
- W4224903891 type Work @default.
- W4224903891 citedByCount "3" @default.
- W4224903891 countsByYear W42249038912023 @default.
- W4224903891 crossrefType "proceedings-article" @default.
- W4224903891 hasAuthorship W4224903891A5003185739 @default.
- W4224903891 hasAuthorship W4224903891A5024006251 @default.
- W4224903891 hasAuthorship W4224903891A5024845093 @default.
- W4224903891 hasAuthorship W4224903891A5039641921 @default.
- W4224903891 hasAuthorship W4224903891A5048159038 @default.
- W4224903891 hasAuthorship W4224903891A5076923788 @default.
- W4224903891 hasBestOaLocation W42249038912 @default.
- W4224903891 hasConcept C119857082 @default.
- W4224903891 hasConcept C138885662 @default.
- W4224903891 hasConcept C144024400 @default.
- W4224903891 hasConcept C154945302 @default.
- W4224903891 hasConcept C162324750 @default.
- W4224903891 hasConcept C187736073 @default.
- W4224903891 hasConcept C199360897 @default.
- W4224903891 hasConcept C204321447 @default.
- W4224903891 hasConcept C2776352735 @default.
- W4224903891 hasConcept C2776401178 @default.
- W4224903891 hasConcept C2777200299 @default.
- W4224903891 hasConcept C2777212361 @default.
- W4224903891 hasConcept C2777530160 @default.
- W4224903891 hasConcept C2779903281 @default.
- W4224903891 hasConcept C2780451532 @default.
- W4224903891 hasConcept C2780660688 @default.
- W4224903891 hasConcept C2780801425 @default.
- W4224903891 hasConcept C36289849 @default.
- W4224903891 hasConcept C41008148 @default.
- W4224903891 hasConcept C41895202 @default.
- W4224903891 hasConcept C90509273 @default.
- W4224903891 hasConceptScore W4224903891C119857082 @default.
- W4224903891 hasConceptScore W4224903891C138885662 @default.
- W4224903891 hasConceptScore W4224903891C144024400 @default.
- W4224903891 hasConceptScore W4224903891C154945302 @default.
- W4224903891 hasConceptScore W4224903891C162324750 @default.
- W4224903891 hasConceptScore W4224903891C187736073 @default.
- W4224903891 hasConceptScore W4224903891C199360897 @default.
- W4224903891 hasConceptScore W4224903891C204321447 @default.
- W4224903891 hasConceptScore W4224903891C2776352735 @default.
- W4224903891 hasConceptScore W4224903891C2776401178 @default.
- W4224903891 hasConceptScore W4224903891C2777200299 @default.
- W4224903891 hasConceptScore W4224903891C2777212361 @default.
- W4224903891 hasConceptScore W4224903891C2777530160 @default.
- W4224903891 hasConceptScore W4224903891C2779903281 @default.
- W4224903891 hasConceptScore W4224903891C2780451532 @default.
- W4224903891 hasConceptScore W4224903891C2780660688 @default.
- W4224903891 hasConceptScore W4224903891C2780801425 @default.
- W4224903891 hasConceptScore W4224903891C36289849 @default.
- W4224903891 hasConceptScore W4224903891C41008148 @default.
- W4224903891 hasConceptScore W4224903891C41895202 @default.
- W4224903891 hasConceptScore W4224903891C90509273 @default.
- W4224903891 hasLocation W42249038911 @default.
- W4224903891 hasLocation W42249038912 @default.
- W4224903891 hasOpenAccess W4224903891 @default.
- W4224903891 hasPrimaryLocation W42249038911 @default.
- W4224903891 hasRelatedWork W1573537589 @default.
- W4224903891 hasRelatedWork W159132833 @default.
- W4224903891 hasRelatedWork W2048614195 @default.
- W4224903891 hasRelatedWork W2354687001 @default.
- W4224903891 hasRelatedWork W2356267126 @default.
- W4224903891 hasRelatedWork W2526529698 @default.
- W4224903891 hasRelatedWork W3082447286 @default.
- W4224903891 hasRelatedWork W4283702175 @default.
- W4224903891 hasRelatedWork W4285066727 @default.
- W4224903891 hasRelatedWork W2612663087 @default.
- W4224903891 isParatext "false" @default.
- W4224903891 isRetracted "false" @default.