DVDGCN: Modeling Both Context-Static and Speaker-Dynamic Graph for Emotion Recognition in Multi-speaker Conversations.
Shuofeng ZhaoPengyuan LiuPublished in: NLPCC (1) (2020)
Keyphrases
- emotion recognition
- speaker verification
- audio visual
- dynamic graph
- multi modal
- emotional speech
- human computer interaction
- speech recognition
- visual information
- visual data
- facial expressions
- contextual information
- speaker identification
- facial images
- affective states
- information fusion
- spatio temporal
- context aware
- hidden markov models