Vae-Space: Deep Generative Model of Voice Fundamental Frequency Contours.
Kou TanakaHirokazu KameokaKazuho MorikawaPublished in: ICASSP (2018)
Keyphrases
- generative model
- fundamental frequency
- probabilistic model
- semi supervised
- em algorithm
- posterior probability
- discriminative models
- discriminative learning
- prior knowledge
- object categories
- expectation maximization
- markov chain monte carlo
- bayesian framework
- speech signal
- latent space
- hierarchical bayesian model
- generative and discriminative models
- space time
- image denoising
- data sets
- topic models
- maximum likelihood
- learning process
- learning algorithm
- machine learning