Fast singing-voice acoustic model: harmonic + noise excitation refined into a mel by a 1-step rectified flow (bundled NHVSing vocoder). Multi-speaker + style control; ONNX/OpenUTAU export.
Python Updated 3 months ago 1 commits 14 stars
prism-dev|Mirror: wavtechyukky/LeapSinger (⭐14)
3 months agoE2B SandboxFirecracker microVMpython
No executable code files found in this repository.
About
Fast singing-voice acoustic model: harmonic + noise excitation refined into a mel by a 1-step rectified flow (bundled NHVSing vocoder). Multi-speaker + style control; ONNX/OpenUTAU export.
Readme
14 stars
1 forks