Joint maximum a posteriori adaptation of transformation and HMM parameters.
Olivier SiohanCristina ChestaChin-Hui LeePublished in: IEEE Trans. Speech Audio Process. (2001)
Keyphrases
- maximum a posteriori
- maximum likelihood
- speaker adaptation
- expectation maximization
- hyperparameters
- map estimation
- prior distribution
- markov random field
- generalized gaussian
- image reconstruction
- bayesian framework
- parameter estimation
- em algorithm
- hidden markov models
- model parameters are estimated
- posterior distribution
- prior model
- edge preserving
- likelihood function
- energy function
- image super resolution
- label field
- update equations
- latent variables
- closed form
- mrf model
- speech recognition
- denoising
- ground truth
- computational complexity