Multi-level Fusion of Wav2vec 2.0 and BERT for Multimodal Emotion Recognition.
Zihan ZhaoYanfeng WangYu WangPublished in: INTERSPEECH (2022)
Keyphrases
- emotion recognition
- audio visual
- emotional speech
- multi modal
- multi stream
- facial expressions
- human computer interaction
- emotion classification
- information fusion
- physiological signals
- facial images
- sentiment analysis
- dimensionality reduction
- hidden markov models
- multimedia
- visual information
- intelligent systems
- emotional state
- sentence level
- natural language processing
- fuzzy logic
- video sequences