GitHub · Python
wavtechyukky/LeapSinger
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 export.
- Stars
- 79
- Forks
- 9
- Stars gained in 24 hours
- +3
- Forks gained in 24 hours
- 0
Topics: speech-synthesis, text-to-speech, tts, vocal-synthesis
Latest data point: Nov 22, 2026 · 79 stars · 9 forks