For macOS, Windows and Linux. Free while the app is in its first release. No account, and nothing is uploaded.
macOS will refuse to open it the first time and say the app
is damaged or from an unidentified developer. It is neither — it has not been
through Apple's notarisation. Right-click the app, choose Open,
then Open again. If it still refuses, run
xattr -dr com.apple.quarantine "/Applications/Audio Editor.app"
in Terminal once.
Windows shows a blue SmartScreen warning. Choose More info, then Run anyway.
The app itself is 14 MB and everything except separation works immediately. The first time you split a mix into vocals and backing, press Download beside the model in the separation dialog. That fetches 75 MB of MIT-licensed Spleeter weights, checks them against a checksum, and keeps them for next time.
macOS 12 (Monterey) or later on an Apple Silicon Mac — M1, M2, M3 or M4. 8 GB RAM, and about 200 MB free once the separation model is installed.
Not supported. The machine-learning runtime the separation engine links against publishes no build for Intel Macs, so it cannot be compiled for them at all.
Audio.Editor_1.2.0_x64_en-US.msi (22 MB) — for deploying across several machines.
Windows 10 or 11, 64-bit. 8 GB RAM. SmartScreen shows a blue warning because the app is not signed yet — choose More info, then Run anyway.
A 64-bit distribution with glibc 2.38 or newer — Ubuntu 24.04, Fedora 39
and later. Audio goes through ALSA or PulseAudio. Make the AppImage
executable with chmod +x before running it.