Automatic Severity Classification of Korean Dysarthric Speech Using Phoneme-Level Pronunciation Features.
Eun Jung YeoSunhee KimMinhwa ChungPublished in: Interspeech (2021)
Keyphrases
- speech recognition
- classification accuracy
- feature vectors
- feature extraction
- feature set
- pattern recognition
- classification process
- classification models
- feature space
- features extraction
- speech synthesis
- classification method
- automatic speech recognition
- speech signal
- benchmark datasets
- svm classifier
- speech recognition systems
- feature selection
- spectral features
- extracted features
- machine learning
- support vector
- support vector machine
- hidden markov models
- mel frequency cepstral coefficients
- visual speech
- spontaneous speech
- emotion classification
- feature selection algorithms
- feature subset
- support vector machine svm
- class labels
- extracting features
- training set
- decision trees
- automatic speech recognition systems
- speech sounds
- speech music discrimination
- acoustic models
- speech recognizer
- emotional state
- low level
- image classification
- text classification
- classification algorithm