Noise-Robust Zero-Shot Text-to-Speech Synthesis Conditioned on Self-Supervised Speech-Representation Model with Adapters | IEEE Conference Publication | IEEE Xplore