StemSplit API: Integrate AI Stem Separation into Your Apps

The StemSplit API is publicly available. If you're building a karaoke app, automating stem extraction for a music library, or adding vocal removal to a production pipeline, you can do it programmatically now.

The API provides the same AI separation quality as the web app — same HTDemucs FT model, same processing infrastructure — through simple REST endpoints with webhook notifications and direct YouTube URL support.

What You Can Build

Stem Separation from Any Audio

Upload MP3, WAV, FLAC, or M4A files and extract:

Process YouTube Videos Directly

Pass a YouTube URL and get separated stems back — no need to download first:

curl -X POST https://stemsplit.io/api/v1/youtube-jobs \
  -H "Authorization: Bearer sk_live_xxx" \
  -H "Content-Type: application/json" \
  -d '{"youtubeUrl": "https://youtube.com/watch?v=dQw4w9WgXcQ"}'

Webhook Notifications

Register a webhook endpoint and receive a POST when jobs complete, rather than polling:

{
  "event": "job.completed",
  "timestamp": "2026-01-05T12:30:00Z",
  "data": {
    "jobId": "clxxx123...",
    "status": "COMPLETED",
    "outputs": {
      "vocals": { "url": "https://...", "expiresAt": "..." },
      "instrumental": { "url": "https://...", "expiresAt": "..." }
    }
  }
}

Webhook payloads include HMAC-SHA256 signatures for verification.

---

Ready to start? Get your API key — new accounts get 5 free minutes to test.

---

How It Works

Step 1: Upload audio directly via presigned URL, or pass a sourceUrl pointing to an existing file.

Step 2: Create a job:

curl -X POST https://stemsplit.io/api/v1/jobs \
  -H "Authorization: Bearer sk_live_xxx" \
  -H "Content-Type: application/json" \
  -d '{
    "sourceUrl": "https://example.com/song.mp3",
    "outputType": "BOTH",
    "quality": "BEST"
  }'

Step 3: Receive results via webhook or poll the job status endpoint. Download stems from the provided URLs once the job completes. Typical processing time is 1–3 minutes depending on audio length and quality setting.

Example: Karaoke Generator in Python

import requests
import time

API_KEY = "sk_live_xxx"
HEADERS = {"Authorization": f"Bearer {API_KEY}"}

response = requests.post(
    "https://stemsplit.io/api/v1/jobs",
    headers=HEADERS,
    json={
        "sourceUrl": "https://example.com/song.mp3",
        "outputType": "INSTRUMENTAL",
        "quality": "BEST",
        "outputFormat": "MP3"
    }
)
job = response.json()
print(f"Created job: {job['id']}")

while True:
    status = requests.get(
        f"https://stemsplit.io/api/v1/jobs/{job['id']}",
        headers=HEADERS
    ).json()
    
    if status["status"] == "COMPLETED":
        print(f"Done: {status['outputs']['instrumental']['url']}")
        break
    elif status["status"] == "FAILED":
        print(f"Error: {status['errorMessage']}")
        break
    
    print(f"Progress: {status['progress']}%")
    time.sleep(5)

That's the complete logic for a karaoke generator — about 30 lines from API call to download URL.

Integration Guides

Documentation covers the most common use cases:

Technical Reference

Authentication

Authorization: Bearer sk_live_xxxxxxxxxxxxxxxxxxxxx

Generate keys in Settings → API Keys. Each account supports up to 5 active keys.

Rate Limits and Constraints

Supported Formats

Input: MP3, WAV, FLAC, M4A, OGG, WebM

Output: MP3, WAV, FLAC (specify in job request)

Quality Settings

Pricing

The API uses the same credit system as the web app — no separate API pricing, no subscription required. Credits are purchased as needed and never expire. New accounts receive 5 free minutes to test. Current rates are on the pricing page.

Documentation

• Getting Started — Quick start with code examples in cURL, Python, and JavaScript
• API Reference — Complete endpoint documentation with request and response schemas
• Integration Guides — Step-by-step tutorials for specific platforms and use cases

Frequently Asked Questions

How is this different from the web app? Same AI model, same quality, same credits — the API provides programmatic access for automation and integration. The web app is for individual one-off processing.

Do I need a subscription? No. Buy credits when you need them. They don't expire.

Can I use this in commercial products? Yes. The API is intended for integration into products and workflows. Your credits pay for processing.

How do I handle webhooks securely? Verify the HMAC-SHA256 signature on every webhook payload using your API key as the secret. Reject any requests that don't validate. Details in the API reference.

Can I process videos, not just audio? The API processes audio. For YouTube URLs, we extract the audio track automatically. For video files, extract the audio first and submit that.

---

Start Building

1. Create an account
2. Generate an API key
3. Read the documentation

New accounts include 5 free minutes. No credit card required to test.

Get Your API Key →