Deep Variational Generative Models for Audio-visual Speech Separation.
Viet-Nhat NguyenMostafa SadeghiElisa RicciXavier Alameda-PinedaPublished in: CoRR (2020)
Keyphrases
- generative model
- visual speech
- hidden markov models
- speaker identification
- probabilistic model
- noisy environments
- mixture model
- discriminative models
- acoustic features
- audio signals
- prior knowledge
- deep belief networks
- speech signal
- discriminative learning
- image segmentation
- gaussian mixture model
- object categories
- topic models
- optical flow
- audio signal
- video signals
- em algorithm
- semi supervised
- text to speech
- broadcast news
- expectation maximization
- multiscale
- semi supervised learning
- machine learning
- training data
- bayesian networks
- denoising
- information extraction
- pairwise
- multimedia