Phonetic classification and recognition using HMM representation of overlapping articulatory features for all classes of English sounds.
Li DengDon X. SunPublished in: ICASSP (1) (1994)
Keyphrases
- multi stream
- hidden markov models
- class labels
- feature representation
- classification accuracy
- feature extraction
- feature vectors
- feature construction
- classification method
- feature set
- classification process
- classification models
- spectral features
- pattern representation
- visual speech recognition
- pattern recognition
- extracted features
- feature values
- decision trees
- visual speech
- feature space
- speech recognition
- svm classifier
- benchmark datasets
- image features
- classification algorithm
- text classification
- feature selection algorithms
- image representation
- training data
- image classification
- support vector
- supervised learning
- multi class
- word level
- machine learning
- musical instrument
- structural features
- eeg signals
- feature subset