An advanced singing voice synthesis system with high fidelity, expressiveness, controllability and flexibility based on DiffSinger: Singing Voice Synthesis via Shallow Diffusion Mechanism
One small question though, I find small pieces related to nfsHiFi here and there in /modules and /scripts. But does it have training code available somewhere in the repo?
Hi,
First of all, thank you for the streamlined code.
One small question though, I find small pieces related to nfsHiFi here and there in /modules and /scripts. But does it have training code available somewhere in the repo?
Thank you.