计算机应用研究2024，Vol.41Issue(1)：51-58,8.DOI:10.19734/j.issn.1001-3695.2023.04.0183
基于时序感知DAG的多模态对话情绪识别模型
Multi-modal temporal-aware DAG for emotion recognition in conversation
摘要
Abstract
Aiming at the issue of insufficient utilization of temporal information,speaker information,and multi-modal infor-mation in existing conversational emotion recognition methods,this paper proposed a multi-modal temporal-aware DAG model(MTDAG).The designed temporal-aware unit optimized the discourse weight setting in chronological order and collected his-torical emotional cues to achieve more effective utilization of temporal and historical information based on recency effect.The context and speaker information fusion module achieved the full utilization of discourse information by extracting the deep joint information of contextual context and speaker self-context.By setting the DAG subgraphs to capture multi-modal information and constrain the interaction direction,the model achieved full utilization of multi-modal information while reducing the intro-duction of noise.Extensive experiments conducted on two benchmark datasets,IEMOCAP and MELD,demonstrate that the model exhibits excellent performance in emotion recognition.关键词
对话情绪识别/有向无环图/近因效应/特征提取/多模态交互Key words
emotion recognition in conversation(ERC)/directed acyclic graph/recency effect/feature extraction/multi-modal interaction分类
信息技术与安全科学引用本文复制引用
沈旭东,黄贤英,邹世豪..基于时序感知DAG的多模态对话情绪识别模型[J].计算机应用研究,2024,41(1):51-58,8.基金项目
国家自然科学基金资助项目(62141201) （62141201）
重庆市自然科学基金资助项目(CSTB2022NSCQ-MSX1672) （CSTB2022NSCQ-MSX1672）
重庆理工大学研究生教育高质量发展行动计划资助项目(gzlcx20223190,gzlcx20232067) （gzlcx20223190,gzlcx20232067）
