Detect AI-generated music
Analyze a song or music track and classify it as AI-generated or human-produced.
Supply audio via one of three intake methods:
- Direct file upload:
multipart/form-datawith the file attached asfile. Files must be 150 MB or smaller. - Public URL:
application/jsonwith aurlfield. Resemble downloads and stores the file before analysis. - Secure upload token:
application/jsonwith amedia_tokenfield obtained fromPOST /secure_uploads.
Exactly one of file, url, or media_token must be provided per request. Only audio files are
accepted.
Processing is asynchronous by default: the response returns immediately with status: processing.
Music detections are standard detections, so you can poll
GET /detect/{uuid} or supply a callback_url to
be notified. Send the Prefer: wait header to block until the verdict is available and receive the
completed item in the same response. Allow a generous client timeout when waiting, since the first
request after a period of inactivity can take several minutes while the model warms up.
Each successful analysis is billed as audio Deepfake Detection usage. Failed analyses are not billed.
Authentication
API key from https://app.resemble.ai/hub/api
Headers
Request
POST destination for the result once processing finishes. The body matches the response of GET /detect/{uuid}.
Return a base64-encoded heatmap image in metrics.image.
Response
Music detection created (or completed when Prefer: wait was sent)
