CCA based feature selection with application to continuous depression recognition from acoustic speech features.
Heysem KayaFlorian EybenAlbert Ali SalahBjörn W. SchullerPublished in: ICASSP (2014)
Keyphrases
- feature extraction
- feature selection
- feature set
- speech recognition systems
- feature space
- feature selection algorithms
- recognition engine
- speech recognition
- classification accuracy
- irrelevant features
- canonical correlation analysis
- selected features
- feature vectors
- feature subset
- emotional speech
- information gain
- mel frequency cepstral coefficients
- feature weighting
- recognition rate
- informative features
- speech sounds
- text categorization
- support vector
- visual speech
- redundant features
- personal identification
- discriminative features
- machine learning
- classification models
- dimensionality reduction
- selecting features
- text classification
- object recognition
- bird species
- speech corpus
- svm classifier