An advanced singing voice synthesis system with high fidelity, expressiveness, controllability and flexibility based on DiffSinger: Singing Voice Synthesis via Shallow Diffusion Mechanism
Apache License 2.0
2.69k
stars
283
forks
source link
Diffusion-base variance predictors with high fidelity, accuracy and expressiveness #88
yqzhishen
commented
1 year ago
Demo of current progress: https://www.bilibili.com/video/BV12T411t7Qg