Optimization of Temporal Filters in the Modulation Frequency Domain via Constrained Linear Discriminant Analysis (C-LDA) for Constructing Robust Features in Speech Recognition.
Jeih-weih HungPublished in: ICASSP (4) (2007)
Keyphrases
- linear discriminant analysis
- frequency domain
- feature extraction
- speech recognition
- discriminant analysis
- fisher criterion
- face recognition
- pattern recognition
- spatial domain
- dimension reduction
- feature space
- speaker identification
- principal component analysis
- dimensionality reduction
- noisy environments
- feature vectors
- power spectrum
- feature analysis
- language model
- hidden markov models
- gabor filters
- null space
- feature set
- support vector machine svm
- scatter matrix
- cross correlation
- high dimensional data
- subspace methods
- image processing
- image classification
- denoising
- speech signal
- support vector
- feature selection
- edge detection
- single channel
- singular value decomposition
- facial expressions
- wavelet transform
- high dimensional
- scatter matrices
- discriminant information
- qr decomposition