ML Parameter Generation with a Reformulated MGE Training Criterion - Participation in the Voice Conversion Challenge 2016.
Daniel ErroAgustín AlonsoLuis SerranoDavid TavarezIgor OdriozolaXabier SarasolaEder del BlancoJon SánchezIbon SaratxagaEva NavasInma HernáezPublished in: INTERSPEECH (2016)
Keyphrases
- maximum likelihood
- training algorithm
- training process
- neural network
- training samples
- online learning
- decision making
- information retrieval
- machine learning
- decision trees
- hidden markov models
- text classification
- genetic algorithm
- expectation maximization
- em algorithm
- parameter space
- parameter values
- data sets
- training phase
- minimum classification error