AI Music Detection in Python — Complete SDK Tutorial

Published May 22, 2026 · 10 min read · AI Song Checker team

Integrating AI music detection into Python apps takes ~10 minutes with our SDK. This tutorial covers install, auth, file/URL analysis, batch processing, error handling, and production patterns.

Installation

pip install aisongchecker

Or directly from source:

pip install git+https://github.com/aisongchecker/python-sdk.git

Authentication

Get your API key from the dashboard (requires Pro tier — 4,99 €/month).

from aisongchecker import Client

client = Client(api_key="ASC_LIVE_xxx")  # or use env: os.environ["ASC_API_KEY"]

Analyze a single file

result = client.analyze_file("track.mp3")

print(f"AI probability: {result.ai_probability:.1%}")
print(f"Verdict: {result.verdict}")  # "ai" | "human" | "uncertain"
print(f"Engine: {result.platform_attribution.top}")
print(f"Confidence: {result.confidence:.2f}")

Analyze a URL (YouTube/Spotify/SoundCloud)

result = client.analyze_url("https://www.youtube.com/watch?v=xyz")
if result.verdict == "ai":
    print(f"AI detected: {result.platform_attribution.top}")
    print(f"Watermarks: c2pa={result.watermarks.c2pa_detected}, synthid={result.watermarks.synthid_detected}")

Batch analysis

tracks = [
    {"id": "t1", "url": "https://soundcloud.com/.../track1"},
    {"id": "t2", "url": "https://open.spotify.com/track/abc"},
    {"id": "t3", "file_path": "/local/track3.mp3"},
]

batch = client.analyze_batch(tracks, webhook_url="https://your-app.com/asc-webhook")
print(f"Job ID: {batch.job_id}, status: {batch.status}")

# Or poll synchronously
result = client.wait_for_batch(batch.job_id, timeout=600)
for r in result.results:
    print(f"{r.id}: {r.ai_probability:.1%} ({r.verdict})")

Error handling with retry

from aisongchecker import Client, RateLimitError, APIError
import time

client = Client(api_key="...", max_retries=5, retry_backoff=2.0)

try:
    result = client.analyze_file("track.mp3")
except RateLimitError as e:
    print(f"Rate limited, retry after {e.retry_after}s")
except APIError as e:
    print(f"API error: {e.code} {e.message}")

Async (asyncio)

import asyncio
from aisongchecker.asyncio import AsyncClient

async def main():
    async with AsyncClient(api_key="...") as client:
        results = await asyncio.gather(*[
            client.analyze_url(url) for url in urls
        ])
        for r in results:
            print(r.verdict)

asyncio.run(main())

Production checklist

Full SDK docs on GitHub · API reference

Related