AutoPersonify

Digital twin API

A digital twin API you can call from your own app

Send photos or a short video from your software, the account owner approves, and a realistic video twin is ready to speak any script you send.

AutoPersonify's digital twin API lets your app upload photos or a short video, request a twin with a likeness statement, and make videos with it once the account owner approves. One API key, plain REST, no studio.

How it works

  1. 1Upload the sourcePOST /uploads returns a signed upload link. Send original, full-resolution photos (up to 50 MB each) and, ideally, one short video (up to 500 MB).
  2. 2Request the twinPOST /twins with the upload paths, a name, whose likeness it is, a voice_id or a voice_personality description, and optional context about the brand.
  3. 3Owner approvesThe account owner gets an email and an in-app card showing every photo, the chosen voice and your context. Nothing is built until they tap Approve.
  4. 4Use it in videosPoll GET /twins/{id} every 30 seconds, or wait for the twin.ready webhook, then POST /videos with twin_id and a script.

The API calls

curl -X POST https://autopersonify.com/api/v1/twins \
  -H "Authorization: Bearer $AUTOPERSONIFY_API_KEY" -H "Content-Type: application/json" \
  -d '{"name":"Max","photos":["<upload id from POST /uploads>"],"voice_personality":"warm, confident British male, 30s",
       "likeness":{"type":"original_character","consent_by":"Jane Smith, founder","statement":"Max is a fictional character we own."},
       "context":"Welcome video for the homepage"}'
# -> status "awaiting_approval" until the account owner taps Approve

Full reference: developer docs, OpenAPI spec, markdown guide.

What you can build

  • A 'make me a twin' flow inside your own product
  • A twin for your app's AI character, so it has a real face
  • Personal video messages from a founder or coach, written by your AI

Honest limits

  • A twin needs at least one real photo or a short video. It cannot be made from a written description alone.
  • Every twin requested through the API waits for the account owner to tap Approve before it is built.
  • Several photos do not create a 360° model. The sharpest front-facing photo sets the face; a short video adds the most realism.

Questions people ask

How long does a twin take to build?

Training starts when the owner approves and usually takes a few hours. Your app should show a clear 'building' state and poll GET /twins/{id}.

Who owns the twin?

The account owner. It lives in their AutoPersonify account and they can withdraw it at any time.

Can I pick the voice?

Yes. Send an exact voice_id from GET /voices, or a voice_personality description that is matched to a real voice in the catalogue. The owner sees and approves the match.

Related

Last updated 2 October 2026.

Start free, no card required

3 free video minutes and one twin. Create an API key in Integrations and paste the prompt into your AI tool.

Start freeRead the API docs