Subspace-constrained supervector PLDA for speaker verification.
Daniel Garcia-RomeroAlan McCreePublished in: INTERSPEECH (2013)
Keyphrases
- speaker verification
- speaker recognition
- noisy environments
- gaussian mixture model
- audio visual
- multilayer perceptron
- prosodic features
- emotion recognition
- low dimensional
- linear discriminant analysis
- multi modal
- neural network
- language identification
- principal component analysis
- visual information
- using artificial neural networks
- image classification
- dimensionality reduction
- high dimensional