Voice conversion using speaker-dependent conditional restricted Boltzmann machine.
Toru NakashikaTetsuya TakiguchiYasuo ArikiPublished in: EURASIP J. Audio Speech Music. Process. (2015)
Keyphrases
- restricted boltzmann machine
- speaker dependent
- deep learning
- speech recognition
- speaker identification
- mel frequency cepstral coefficients
- conditional random fields
- probabilistic graphical models
- speaker independent
- learning algorithm
- gaussian mixture model
- hidden layer
- hidden markov models
- semi supervised
- speaker recognition
- image processing
- loopy belief propagation
- speech recognizer
- machine learning
- text classification
- speaker adaptation
- feature selection
- spectral features
- training data
- active learning
- transfer learning
- markov random field
- graphical models
- mixture model