Cross-covariance-based features for speech classification in film audio.
Matthew BenatanKia NgPublished in: J. Vis. Lang. Comput. (2015)
Keyphrases
- classification accuracy
- feature vectors
- feature extraction
- feature set
- feature space
- classification process
- classification method
- cepstral features
- classification models
- mel frequency cepstral coefficients
- audio visual
- audio features
- genre classification
- extracted features
- spectral features
- feature analysis
- acoustic signals
- benchmark datasets
- extracting features
- support vector machine
- pattern recognition
- feature representation
- svm classifier
- class labels
- image features
- emotion classification
- multi modal
- linear predictive coding
- decision trees
- digital audio
- feature selection
- acoustic features
- audio stream
- audio signals
- textural features
- speaker identification
- high dimensionality
- text classification
- training set
- text to speech
- audio signal
- music information retrieval
- automatic speech recognition
- speech synthesis
- speech signal
- audio recordings
- classification algorithm
- low level
- machine learning