Mitigating data replication in text-to-audio generative diffusion models through anti-memorization guidance

DiffMRe-assign community