arXiv - CSCL: "NaturalSpeech 2: Latent Diffusion Models are Natu…" - Qoto Mastodon

arXiv - CSCL @arxiv_cscl@qoto.org

NaturalSpeech 2: Latent Diffusion Models are Natural and Zero-Shot Speech and Singing Synthesizers. (arXiv:2304.09116v3 [eess.AS] UPDATED)

http://arxiv.org/abs/2304.09116 #arXiv #NLProc

May 31, 2023, 03:11 · · arxiv-cscl · · ·

Sign in to participate in the conversation