Non-parallel training for voice conversion by maximum likelihood constrained adaptation.
Athanasios MouchtarisJan Van der SpiegelPaul MuellerPublished in: ICASSP (1) (2004)
Keyphrases
- maximum likelihood
- discriminative training
- speaker adaptation
- parallel computing
- likelihood function
- em algorithm
- parameter estimation
- training set
- maximum likelihood estimation
- test set
- training examples
- maximum a posteriori
- parallel processing
- training algorithm
- parallel implementation
- online learning
- parallel programming
- covariate shift
- machine learning
- gaussian distribution
- supervised learning
- emotion recognition
- adaptation process
- hidden markov models