Upmix a Music Catalog to Spatial Audio

Why this matters

Spatial audio is becoming a delivery requirement on major platforms, but remixing thousands of tracks manually is not feasible. AI upmixing produces a consistent, licensable result at scale.

Batch jobs keep per-track settings and outputs uniform, so track 1 and track 4,000 sound like the same spatial approach.

How it works

Submit the folder with output=atmos-7.1.2; each track is upmixed with the same model settings and returned as ADM/BWF + 7.1.2 renders.

Submit a job with an input URL (S3, GCS or HTTPS), poll the job URL, and download the rendered output. No GPU, no queues, no ffmpeg builds to babysit.

Code

import requests, time

API = "https://api.mlslabs.io/v1/ai-mixing/jobs"
headers = {"X-API-Key": "YOUR_API_KEY"}
payload = {"input": "s3://bucket/catalog/", "stems": ["vocal", "guitar", "bass", "drums"], "output": "atmos-7.1.2", "pattern": "*.wav"}

resp = requests.post(API, json=payload, headers=headers)
job = resp.json()
while job["status"] not in ("succeeded", "failed"):
    time.sleep(3)
    job = requests.get(job["url"], headers=headers).json()

print("Output:", job["output_url"])

Pro tips

• Normalize loudness before upmixing — spatial playback is loudness-sensitive.
• QC every 10th track on real spatial monitors, not headphones.
• Keep stereo masters; spatial is an additional delivery.

Run the AI Mixing collection — spatial preset prefilled.