Zero-shot Voice Conversion via Self-supervised Prosody Representation Learning | IEEE Conference Publication | IEEE Xplore