A Class-dependent Background Model for Speech Signal Feature Extraction.
Yuechi JiangFrank H. F. LeungPublished in: DSL (2018)
Keyphrases
- speech signal
- background model
- feature extraction
- background subtraction
- speech recognition
- moving objects
- automatic speech recognition
- foreground objects
- gaussian mixture model
- non stationary
- feature selection
- feature vectors
- object detection
- preprocessing
- face recognition
- hidden markov models
- noisy environments
- feature space
- image processing
- image classification
- pattern classification
- pattern recognition
- image sequences
- principal component analysis
- segmentation algorithm
- manifold learning
- video sequences
- mixture model
- denoising
- cost sensitive
- extracted features
- bayesian networks
- information retrieval