Audio-Based Emotion Recognition from Natural Conversations Based on Co-Occurrence Matrix and Frequency Domain Energy Distribution Features.
Aya SayedelahlPouria FewzeeMohamed S. KamelFakhri KarrayPublished in: ACII (2) (2011)
Keyphrases
- co occurrence
- frequency domain
- emotion recognition
- energy distribution
- subband
- spatial domain
- audio visual
- feature extraction
- gray level
- low frequency
- wordnet
- facial expressions
- denoising
- human computer interaction
- multiresolution
- filter bank
- feature vectors
- audio features
- multi modal
- bit rate
- high frequency
- sentiment analysis
- image features
- multiscale
- frequency band
- artificial intelligence
- computational complexity
- high resolution
- face detection
- feature set
- knowledge representation