Spatial Audio
AI Mixing API — auto-mix stems to immersive audio
Static and dynamic analysis of your stems — tempo, genre, masking, strum patterns — drives automatic mixing to surround and object-based immersive audio, at a junior mixing engineer's quality bar.
Before & after
See the difference
Quiet, uneven dry stems go in — a loud, broadcast-ready master comes out.
Mastered to broadcast loudness (-14 LUFS) — loud, punchy and consistent on every speaker.
What it does
Send stems, get a balanced mix — without gain staging by hand. Static analysis identifies tempo and genre to set the right effect parameters before a single fader moves. Dynamic analysis resolves masking: when stems collide in frequency, priority ordering keeps the lead audible. It even recognizes arpeggio versus strummed accompaniments and treats them differently. The result renders to stereo, 5.1, 7.1.2 or object-based spatial masters.
Full API referenceimport requests
url = "https://api.mlslabs.io/v1/ai-mixing/jobs"
resp = requests.post(
url,
headers={"X-API-Key": "YOUR_API_KEY"},
json={
"stems": [
{"url": ".../vocals.wav", "role": "vocals", "priority": 1},
{"url": ".../drums.wav", "role": "drums", "priority": 2},
{"url": ".../bass.wav", "role": "bass", "priority": 3},
{"url": ".../guitar.wav", "role": "guitar", "priority": 4},
],
"preset": "modern_pop",
"output": {"format": "7.1.2", "objects": True},
},
)
job = resp.json()
print(job["job_id"])Key features
Why teams pick this API
Static intelligence
Rhythm and genre detection set the right effect parameters before a single fader moves.
Masking resolution
When stems collide in frequency, priority ordering (predefined, spectral energy or loudness) keeps the lead audible.
Playing-style awareness
Recognizes arpeggio vs strummed accompaniments and treats them differently — more musical, less auto.
Object-based output
Surround/immersive output with optional audio objects for modern spatial workflows.
Use cases
Built for real workflows
Immersive livestreaming
Call the API (or mount the plugin workflow) and get surround/immersive mixes out — fast, with artistic quality.
Media asset servers
Batch-convert legacy stereo programs to surround/immersive on content production servers, saving manual labor costs.
Car & home-theater playback
Pair with Source Separation: split stereo on playback, mix in real time, and experience immersive audio on any device.
Pricing
Credit-based pricing
Every request draws from your credit pool. This API bills 20 credits per minute of audio.
| Billing unit | Cost in credits | Notes |
|---|---|---|
| Processing time | 20 credits | Per minute of audio |
| Failed jobs | 0 credits | Credits auto-refunded |
| Batch processing | Same rate | One queue, per-file callbacks |
Learn
Guides, tutorials & comparisons
How to Automatically Mix Multitrack Stems in Python
Auto-mix multitrack stems in Python: submit stems with roles and priorities, get a balanced stereo o...
Read TutorialUpmix a Music Catalog to Spatial Audio
Upmix an entire music catalog to spatial audio: batch stereo-to-Atmos conversion with per-track obje...
Read TutorialAuto-Mix Audio in Go
Auto-mix multitrack stems in Go with the mlslabs AI Mixing API: balanced stereo output plus a settin...
Read TutorialAuto-Mix Audio in Node.js
Auto-mix multitrack stems in Node.js with the mlslabs AI Mixing API: balanced stereo output plus a s...
Read TutorialSpatial Audio for Livestreams
Deliver spatial audio on livestreams: real-time stereo-to-5.1 upmixing for concerts, sports and even...
ReadAI Mixing vs Human Mixing: Where Each One Wins
AI mixing vs human mixing engineers in 2026: what AI is genuinely good at, what humans still own, an...
Read GuideWhat Is Frequency Masking — and Why It Ruins Your Mixes
Frequency masking explained: why overlapping instruments bury your lead, what engineers do about it,...
Read GuideChannel-Based vs Object-Based Audio
Channel-based vs object-based audio explained: what objects are, ADM/Atmos containers, how beds work...
Read GuideSpatial Audio vs Surround Sound
Spatial audio vs 5.1/7.1 surround sound explained: channel beds vs object-based audio, Dolby Atmos, ...
ReadAI Mixing API vs DAW Plugins (Neutron, Nectar): Which Workflow Scales?
AI mixing API vs in-DAW AI plugins like iZotope Neutron: batch workflows, no DAW dependency, program...
Read CompareAI Mixing API vs RoEx Automix
RoEx Automix is an AI mixing tool; mlslabs is an AI mixing API for pipelines. Compare mixing quality...
ReadFAQ
Common questions
What input does it need?
Pre-separated stems with roles (vocals, drums, bass, guitar, piano, brass, strings). No stems? Run Source Separation first — one pipeline, one vendor.
What output formats do you support?
Stereo WAV/FLAC, 5.1/7.1.2 channel-bed WAV, and object-based spatial masters (ADM BWF). Atmos-branded encoded output requires a Dolby license — talk to us.
Can I get the FX settings instead of the rendered mix?
Yes — the API returns a settings JSON (gain, pan, EQ carve points, masking decisions) you can apply in your own pipeline.
How does it handle frequency masking?
When stems overlap in frequency, priority-based ducking and EQ carving keep the lead audible.
Do you do mastering?
No — mixing only, and we say so plainly. Pair the mix with a mastering house for final loudness and release delivery.
What happens if a job fails?
Credits are auto-refunded. Failed jobs never bill you.
Related APIs