RANSAC-Based Training Data Selection on Spectral Features for Emotion Recognition from Spontaneous Speech.
Elif BozkurtEngin ErzinÇigdem Eroglu ErdemA. Tanju ErdemPublished in: COST 2102 Conference (2010)
Keyphrases
- emotion recognition
- spectral features
- spontaneous speech
- training data
- structural features
- audio visual
- human computer interaction
- spoken language
- spectral analysis
- facial expressions
- hyperspectral images
- remote sensing images
- classification accuracy
- information fusion
- sentiment analysis
- speaker verification
- automatic speech recognition
- emotional state
- training set
- machine learning
- multi modal
- natural language processing
- hidden markov models
- face recognition