cd /news/machine-learning/solving-the-ground-truth-bottleneck-… · home › topics › machine-learning › article
[ARTICLE · art-170311] src=dev.to ↗ pub= topic=machine-learning verified=true sentiment=↑ positive

Solving the Ground-Truth Bottleneck in Music Source Separation: Announcing mixforge-audio on PyPI

A developer released mixforge-audio v0.1.0 on PyPI, an open-source Python engine that builds reproducible test mixes with an exact answer key and calibrated acoustic bleed for evaluating music source separation models. Validated on 20 multi-track songs (40 mixes) from MUSDB18-HQ, the tool reported a 40/40 verification pass rate, a maximum mixture-to-stem residual of 6.52 × 10⁻⁸ in 64-bit float, and a -69.3 dB median residual versus original master mixes, running on NumPy, SciPy and SoundFile without GPU compute.

by read1 min views1 publishedNov 16, 2026

🚀 Excited to announce the official release of mixforge-audio (v0.1.0) on PyPI!

If you work in Audio AI, Music Source Separation (Demucs, Spleeter, Open-Unmix), or Bleed Detection, you've hit this core evaluation bottleneck: 👉 On commercial tracks, the true ground truth is completely unknown.

Commercial recordings contain unmeasured mic leakage, mastering compression, and stereo imaging that bias metrics (SDR, SIR, SAR). When models fail, it's hard to tell whether separation failed or if source stems had bleed originally.

mixforge-audio solves this: an open-source Python engine engineered to build reproducible, mathematically exact test mixes with an unalterable answer key and calibrated acoustic bleed.

recipe.json recording every sample offset, gain factor, and clipping ratio.mixforge verify) verifies |Mixture - ∑ Stems| ≤ ε and checks that leaky stems match recipe bleed.

Validated across 20 multi-track songs (40 mixes) from MUSDB18-HQ:

✅ Verification Pass Rate: 40 / 40 mixes OK (0 failures)

✅ 64-Bit Float Precision: Max |Mixture - ∑ Stems| = 6.52 × 10⁻⁸

✅ Flat Bleed Precision: 0.0000 dB gap between requested and measured leakage

✅ Low-Pass Bleed: -14.45 dB high-frequency damping above 4 kHz ✅ Official Mix Residual: -69.3 dB median vs original master mixes (well below the human auditory threshold)

No PyTorch, TensorFlow, or GPU compute needed. Runs purely on NumPy, SciPy, and SoundFile for instant, sub-second execution on any CPU or CI/CD runner.

Install directly from PyPI:

pip install mixforge-audio

Generate a test dataset with bleed in 1 command:

mixforge make stems/ --out mixes/ --count 3 --seconds 10 \

    --bleed "drums>vocals:-26@4000" --bleed "bass>other:-30" \

    --seed 42

Verify mix integrity:

mixforge verify mixes/ --html report.html

🔗 PyPI: [https://pypi.org/project/mixforge-audio/](https://pypi.org/project/mixforge-audio/)

💻 GitHub: [https://github.com/BoobalanArjunan/Mix-Forge](https://github.com/BoobalanArjunan/Mix-Forge)

If you're benchmarking separation models or training audio bleed detectors, give it a spin! ⭐


── more in #machine-learning 4 stories · sorted by recency
── more on @mixforge-audio 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
→ Live at https://your-agent.zahid.host ✓
Get free account → Pricing
from €0/mo · no card required
LIVE [news/solving-the-ground-t…] indexed:0 read:1min 2026-11-16 · —