Fig. 5From: Speaker-adaptive-trainable Boltzmann machine and its application to non-parallel voice conversionComparison of VC performance of non-parallel approachesBack to article page