On the Distribution of Speaker Verification Scores: Generative Models for Unsupervised Calibration.
Sandro CumaniPublished in: IEEE ACM Trans. Audio Speech Lang. Process. (2021)
Keyphrases
- generative model
- speaker verification
- semi supervised
- probabilistic model
- noisy environments
- discriminative learning
- discriminative models
- prior knowledge
- mixture model
- supervised learning
- em algorithm
- conditional random fields
- audio visual
- deep belief networks
- multilayer perceptron
- semi supervised learning
- topic models
- labeled data
- unsupervised learning
- emotion recognition
- language identification
- neural network
- visual information
- multi modal
- active learning
- high resolution
- training set
- image processing