Wrapper class for speaker specific GMMs for voice conversion This is the dual of WeightedCodebook class in codebook mapping
Implements joint-GMM based speaker feature transformation from source acoustic space to target acoustic space
Joint-GMM output for voice conversion
Joint-GMM voice conversion training using parallel source and target databases Reference: A.
A collection of JointGMMs, i.e.
Joint-GMM voice conversion training Both parallel and non-parallel training should derive from this class
Parameters for joint-GMM based voice conversion training
Voice conversion transformation using Joint-GMM approach.
Parameters for joint-GMM based voice conversion transformation stage
Copyright © 2000–2016 DFKI GmbH. All rights reserved.