BDDM: Bilateral Denoising Diffusion Models for Fast and High-Quality Speech Synthesis.

Max W. Y. Lam, Jun Wang, Dan Su, Dong Yu

Browse the full ICLR paper archive.