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
2.62k
stars
275
forks
source link
Support using `e` as the base number of the log mel-spectrogram #175