Adversarial manifold learning for speaker recognition.
Jen-Tzung ChienKang-Ting PengPublished in: ASRU (2017)
Keyphrases
- manifold learning
- speaker recognition
- gaussian mixture model
- speaker verification
- vector quantization
- dimensionality reduction
- speaker identification
- dimension reduction
- low dimensional
- semi supervised
- probabilistic neural network
- high dimensional
- high dimensional data
- feature extraction
- feature space
- speech recognition
- principal component analysis
- image compression
- mixture model
- sparse representation
- noisy environments
- maximum likelihood
- speech signal
- data sets
- image classification
- least squares
- nearest neighbor
- image segmentation
- image processing
- neural network