End-to-End Text-to-Speech using Latent Duration based on VQ-VAE

Re-assign community