A Comparative Study of Linear and Nonlinear Dimensionality Reduction for Speaker Identification.
Andrew ErrityJohn McKennaPublished in: DSP (2007)
Keyphrases
- speaker identification
- dimensionality reduction
- feature extraction
- speech recognition
- gaussian mixture model
- speech signal
- noisy environments
- high dimensional
- principal component analysis
- high dimensional data
- pattern recognition
- feature transformation
- data points
- feature space
- low dimensional
- feature selection
- unsupervised learning
- principal components
- broadcast news
- image processing
- face recognition
- video sequences
- automatic speech recognition
- image classification
- mixture model
- neural network
- information retrieval