DMs

引入

原理

正向扩散

反向过程

优化（推导略）

条件

音频

NUWAVE

NU-Wave: A Diffusion Probabilistic Model for Neural Audio Upsampling

WSRGlow

WSRGlow: A Glow-based Waveform Generative Model for Audio Super-Resolution

NUWAVE2

NU-Wave 2: A General Neural Audio Upsampling Model for Various Sampling Rates

UDM

Conditioning and Sampling in Variational Diffusion Models for Speech Super-Resolution

相关

Dhariwal & Nichol, 2021
GLIDE: Nichol et al., 2021
Palette: Saharia et al., 2021
DALL-E 2: Ramesh et al., 2022