Your Voice Cloning System is Secretly a Voice Anonymizer
Novel research approach showing voice cloning models can be adapted for privacy-preserving applications without additional training.
AI Summary
Researchers repurpose XTTSv2 voice cloning model for speaker anonymization without retraining, achieving near-optimal privacy while preserving speech quality across seven languages.
Excerpt
Speaker anonymization suppresses speaker-identifying attributes from speech while preserving linguistic content and quality. We propose repurposing XTTSv2, a multilingual voice cloning model trained on 27k hours of speech, for speaker anonymization without retraining. Our key insight is that XTTSv2's voice cloning capabilities preserve prosodic structure independently of speaker identity, enabling voice conversion by conditioning on a pseudo-speaker. We introduce an iterative refinement strategy