2024 ICML ICML 2024

NaturalSpeech 3: Zero-Shot Speech Synthesis with Factorized Codec and Diffusion Models