Improving deep neural networks using state projection vectors of subspace Gaussian mixture model as features.
Murali Karthick BSrinivasan UmeshPublished in: SLT (2014)
Keyphrases
- gaussian mixture model
- feature vectors
- feature space
- neural network
- covariance matrices
- feature extraction
- mixture model
- principal components
- gaussian distribution
- background subtraction
- speaker recognition
- expectation maximization
- high dimensional
- gaussian mixture
- pattern recognition
- speaker identification
- similarity measure
- maximum likelihood
- low dimensional
- vector space
- feature set
- object recognition