Audio Classification and Retrieval Using Wavelets and Gaussian Mixture Models.
Ching-Hua ChuanPublished in: Int. J. Multim. Data Eng. Manag. (2013)
Keyphrases
- gaussian mixture model
- feature vectors
- speaker identification
- speech music discrimination
- mixture model
- signal processing
- density estimation
- mel frequency cepstral coefficients
- feature extraction
- em algorithm
- maximum likelihood criterion
- speaker recognition
- classification accuracy
- information retrieval
- support vector
- pattern recognition
- expectation maximization
- image classification
- image compression
- pattern classification
- maximum likelihood
- training set
- feature set
- decision trees
- probabilistic neural network
- feature space
- image retrieval
- text classification
- information retrieval systems
- wavelet transform
- unsupervised learning
- visual information
- distance measure
- low level
- image processing
- finite mixtures
- feature selection
- machine learning