Forensic AI Model

Audio forensics
for AI-generated music

ArtifactNet is a lightweight forensic framework for detecting AI-generated music. Rather than learning generator-specific fingerprints, it targets residual artifacts introduced by neural audio codec and RVQ-based generation pipelines.

Try it free → View pricing
Powered by ArtifactNet

VST3 / AU / CLAP plug-ins for producers and engineers working with AI music, neural codecs, and separated stems.

Launch promo through July 31, 2026  ·  Start the 14-day trial →

$179 $129
de-artifact plug-in interface — HPSS spectrum view, attenuation parameters, de-esser controls
de-artifact
A master bus built for AI music.
$149 $99
de-leak-rt plug-in interface — 3-band leak gate visualization with low/mid/high frequency bands
de-leak-rt
Real-Time Leak Gate for Stem Separations

Start your 14-day trial

See both plug-ins side-by-side first — the video below puts de-artifact and de-leak-rt on the same drum bus, A/B against the dry source. Then download either installer for a free 14-day full trial — every preset, every parameter, no gaps. After 14 days the plug-in inserts a brief silent gap every 60 seconds until you activate.

de-artifact

14-day free trial

A master bus built for AI music. Adaptive HPSS + UNet on the 2-bus, no per-track setup. Full quality for 14 days with no gaps; brief silent gap every 60 s after that until activated.

de-artifact-0.1.8.pkg Universal · 200 MB

de-leak-rt

14-day free trial

Real-time leak gate for Demucs / htdemucs stems. F1 = 0.993 classifier, multiband HF preservation. Same 14-day full trial, same .pkg shape.

de-leak-0.1.6.pkg Universal · 31 MB

macOS 11+ · Apple Silicon + Intel · Notarized by Apple  ·  Windows 10/11 x86_64  ·  Linux on the roadmap  ·  VST3 / AU / CLAP / Standalone. Activate with a license key emailed to you after Paddle checkout.

From the blog

Guide · May 2026
How to Fix AI Music Artifacts (Suno, Udio, Stable Audio)
Remove metallic, watery artifacts from AI-generated tracks — causes and fixes.
Guide · May 2026
How to Detect AI-Generated Music — Methods & Tools
Listening cues, spectrogram analysis, and ArtifactNet benchmark comparison.