AhmadHassan-BTed / NaSuno
git clone https://github.com/AhmadHassan-BTed/NaSuno.git
Python DSP toolkit that attempts to strip Suno, Udio, and other AI audio watermarks and fingerprints from generated audio files.
Why you might care
Aimed at producers who incorporate AI-generated stems into their own tracks and want to remove embedded provenance markers. It uses notch filtering and psychoacoustic masking to target watermark carriers, with a claimed goal of keeping audible quality intact. Effectiveness claims are the author's and unverified here.
+230
stars, 24h · #141
+~2.6k
stars, 7 days · #67
+~11k
stars, 30 days · #58
3.8k
★ total · +219% this week
4
forks · 0 open issues
Star history (as recorded here)
LicenseMIT
Created2023-07-25 (3y ago)
Last push2d ago
Watchers0
OwnerUser
GitHub topics
ai-audioanti-attributionaudio-forensicsaudio-processingaudio-watermarkingdspfl-studiomusic-productionneural-vocoderperceptual-maskingpsychoacousticspythonsteganographysuno-aiwatermark-removal
Links
- GitHub: github.com/AhmadHassan-BTed/NaSuno
- Homepage: https://ahmadhassan-bted.github.io/NaSuno/
- Releases · Issues · Hacker News mentions
README
Loading README…
Blurb generated by claude-haiku-5-5 today. Star deltas marked ~ are extrapolated from partial history.