Unsupervised Vocal Dereverberation with Diffusion-Based Generative Models.
Koichi SaitoNaoki MurataToshimitsu UesakaChieh-Hsin LaiYuhta TakidaTakao FukuiYuki MitsufujiPublished in: ICASSP (2023)
Keyphrases
- generative model
- semi supervised
- probabilistic model
- mixture model
- unsupervised learning
- discriminative learning
- generative and discriminative models
- deep belief networks
- em algorithm
- discriminative models
- conditional random fields
- semi supervised learning
- hierarchical hidden markov models
- prior knowledge
- latent dirichlet allocation
- discriminative methods
- expectation maximization
- pairwise
- representational power
- discriminatively trained
- information retrieval
- topic models
- supervised learning
- object categories
- highly efficient
- hidden variables
- training data
- image segmentation
- computer vision