Constrained discriminative PLDA training for speaker verification.
Johan RohdinSangeeta BiswasKoichi ShinodaPublished in: ICASSP (2014)
Keyphrases
- speaker verification
- noisy environments
- speaker recognition
- prosodic features
- audio visual
- training set
- multilayer perceptron
- multiscale
- image processing
- feature selection
- contextual information
- neural network
- feature extraction
- face verification
- emotion recognition
- language identification
- computer vision
- information retrieval