Fig. 4From: U2-VC: one-shot voice conversion using two-level nested U-structureComparison among source, target and the converted spectrograms. a Source speech. b Target speech. c The converted speech of AGAIN-VC. d The converted speech of U2-VCBack to article page