Conversational Emotion Recognition Using Self-Attention Mechanisms and Graph Neural Networks

Zheng Lian, Jianhua Tao, Bin Liu, Jian Huang, Zhanlei Yang, Rongjun Li

Different from the emotion estimation in individual utterances, context-sensitive and speaker-sensitive dependences are vitally pivotal for conversational emotion analysis. In this paper, we propose a graph-based neural network to model these dependences. Specifically, our approach represents each utterance and each speaker as a node. To bridge the context-sensitive dependence, each utterance node has edges between immediate utterances from the same conversation. Meanwhile, the directed edges between each utterance node and its speaker node bridge the speaker-sensitive dependence. To verify the effectiveness of our strategy, we conduct experiments on the MELD dataset. Experimental results demonstrate that our method shows an absolute improvement of 1%~2% over state-of-the-art strategies.

doi: 10.21437/Interspeech.2020-1703

Cite as: Lian, Z., Tao, J., Liu, B., Huang, J., Yang, Z., Li, R. (2020) Conversational Emotion Recognition Using Self-Attention Mechanisms and Graph Neural Networks. Proc. Interspeech 2020, 2347-2351, doi: 10.21437/Interspeech.2020-1703

@inproceedings{lian20c_interspeech,
  author={Zheng Lian and Jianhua Tao and Bin Liu and Jian Huang and Zhanlei Yang and Rongjun Li},
  title={{Conversational Emotion Recognition Using Self-Attention Mechanisms and Graph Neural Networks}},
  year=2020,
  booktitle={Proc. Interspeech 2020},
  pages={2347--2351},
  doi={10.21437/Interspeech.2020-1703}
}