Audio-based emotion recognition using GMM supervector an SVM linear kernel.
Dinh-Son TranHyung-Jeong YangSoo-Hyung KimGueesang LeeLuu Ngoc DoNgoc-Huynh HoVan Quan NguyenPublished in: ICMLSC (2018)
Keyphrases
- gaussian mixture model
- speaker identification
- feature vectors
- emotion recognition
- support vector machine svm
- support vector
- feature space
- mixture model
- mouth region
- support vector machine
- speaker recognition
- multimedia
- em algorithm
- facial expressions
- gaussian mixture modeling
- multimodal fusion
- audio visual
- mel frequency cepstral coefficients
- expectation maximization
- gaussian mixture
- background subtraction
- svm classification
- svm classifier
- knn
- audio signals
- generalization ability
- kernel function
- training data
- visual information
- training set
- feature selection
- multi class classification
- multi class
- maximum likelihood
- hyperplane
- kernel methods
- classification algorithm
- voice activity detection
- audio signal
- music information retrieval
- signal processing
- support vectors
- multi modal
- noisy environments
- feature extraction
- face recognition
- speech signal
- audio files
- visual data
- pattern recognition
- visual features
- generative model