Visual-to-speech conversion based on maximum likelihood estimation.
Rina RaRyo AiharaTetsuya TakiguchiYasuo ArikiPublished in: MVA (2017)
Keyphrases
- maximum likelihood estimation
- maximum likelihood
- em algorithm
- speech recognition
- multivariate gaussian
- boltzmann machine
- expectation maximization
- parameter estimation
- probability distribution
- mixture of gaussians
- visual information
- probability density
- visual features
- automatic speech recognition
- human body
- speech signal
- graph cuts
- visual speech
- markov random field
- poisson noise
- data mining
- proximal point