arxiv
PublishedJune 1, 2026 at 4:00 AM
Targeted Speaker Poisoning Framework in Zero-Shot Text-to-Speech
Publisher summary· verbatim
arXiv:2603.07551v2 Announce Type: replace-cross Abstract: Zero-shot Text-to-Speech (TTS) voice cloning poses severe privacy risks, demanding the removal of specific speaker identities from trained TTS models. Conventional machine unlearning is insufficient in this context, as zero-shot TTS can dynam
Stay posted· Newsletter
A 5-min weekly brief — top movers, price watch, story of the week.
Discussion
No replies yet. Be first.
Related coverage
More from ARXIV
arxivWhen LLMs Agree, Are They Right? Auditing Self-Consistency and Cross-Model Agreement as Confidence Signals9harxivPLURAL: A Global Dataset for Value Alignment9harxivDoes Dimensionality Reduction via Random Projections Preserve Landscape Features?9harxivInfinity-Parser2 Technical Report9hThe Bubble Brief
WEEKLYRead AI insights every Tuesday — top movers, new releases, story of the week.
Originally published on arxiv ↗