Stand-in

Not an avatar

An AI avatar that copies your movements, because it is your video

Most AI avatars read a script: a generated head, a generated voice, generated hand waves. They look like avatars. Stand-in works the other way around. You film a real clip. A fictional Stand-in replaces you whole-person, face and body, while your camera, your lighting, your background and your real movements stay. Nothing is acted by a machine.

Try it on your next videoSee the Stand-ins

How it works

  1. 1. Film yourself

    Any phone clip with one person in frame. Your movements are the performance; the Stand-in follows them.

  2. 2. Pick the person

    A ready-made Stand-in, or your own. By default they wear what you wear; you can pick an outfit instead.

  3. 3. Get the video

    Your voice stays, or swap it in one tap. Usually ready in about 10 minutes.

What you need

  • • A real clip of you
  • • Face visible, steady camera, one person
  • • Vertical works best

What it costs

from $3.99 per 30-second video with a $10 credit pack (from $3.63 on Pro); from $2.00 up to 15 seconds. No free video; credit packs from $10. Up to 10 minutes on any account; each started 15 seconds uses credits. Credit packs never expire, plan credits roll over, and a failed video gives its credits back. Ready-made Stand-ins are free on every video; HD is included.

Questions

Is Stand-in an AI avatar?
Not in the usual sense. An avatar is generated from a script or a photo. Stand-in edits the video you filmed: it replaces you with a fictional Stand-in and keeps everything else, so it is closer to a body double than an avatar.
Does the Stand-in copy my hand gestures and head turns?
Yes. The Stand-in is placed into your footage, so your gestures, head turns, pauses and energy carry over. Only your face and body change.
Does it make a digital twin of me?
No. Stand-in never builds a twin of your face. Every Stand-in is a generated fictional person who belongs to no one.
Can it lip-sync a script I did not say?
No. Stand-in keeps the words and timing you recorded. If you want different words, record them. Swapping the voice keeps your timing.

Related