AI Full-Body Avatar
A Photo That Speaks and Moves
Upload a photo of a person and a voice track, and get a full-body video of that person speaking it. Cost: from 34 credits for 10 seconds of audio.
Trusted by 10,000+ creators worldwide
More Than a Moving Mouth
The whole person performs the voice track: face, hands and posture, not only the lips.
Full-Body Performance
The person in your photo speaks and gestures, so the result reads as a performance rather than a face on a still.
Your Own Voice Track
Bring a recording, a voice-over or a text-to-speech file. The video follows its timing and delivery.
Choose Who Speaks
Photo with several people? Add a mask image and only the person in its white area speaks.
720p or 1080p, Plus Fast Mode
Pick the resolution, and switch on fast mode when you want the clip sooner at a small cost in quality.
Priced by the Audio
From 34 credits for 10 seconds of audio. You see the exact price before you run.
Checked Before You Pay
Your photo and your prompt are checked before any credits are charged, and a refused request costs nothing.
Three Steps to a Speaking Avatar
Photo, voice, download.
Add a Photo and a Voice Track
Voice tracks up to 60 s at 720p or 30 s at 1080p. Trim a longer track first.
Pick the Details
Choose 720p or 1080p, add a mask if the photo has several people, and optionally describe how the person should move.
Generate and Download
Check the price, run it, and download the video from your History.
AI Full-Body Avatar FAQ
What does the full-body avatar make?+
You give it a photo of one person and a voice track. It returns a video of that person speaking the track, with natural head, hand and body movement instead of a lip-sync overlay on a still face.
How is a full-body avatar different from a talking photo?+
A talking photo animates the face to match the words. The full-body avatar also moves the body, so it suits a presenter in frame, a spokesperson or a character who gestures while talking.
How much does the full-body avatar cost?+
From 34 credits for 10 seconds of audio. Priced per second of audio, the same at 720p and 1080p: 100 credits for 30 seconds, 199 credits for 60 seconds at 720p. You see the exact price before you run.
How long can the voice track be?+
Voice tracks up to 60 s at 720p or 30 s at 1080p. Trim a longer track first.
Which resolution should I pick for an avatar video?+
Pick 1080p for the sharpest result and 720p when you need a longer track, since 720p takes audio up to twice as long. The price follows the audio length, not the resolution.
What is the mask for?+
When your photo shows more than one person, a mask image tells the avatar who speaks: only the person inside the white area of the mask talks, and everyone else stays quiet.
Whose photo and voice can I use?+
Only your own, or those of people who agreed. Do not use the avatar to pass someone off as a person they are not. Running it takes an account with a verified email and credits; there is no anonymous generation. If a run fails on the provider side, the credits go back to your balance.
Do I need a subscription?+
No. You can pay as you go: credit packs are one-time purchases and their credits never expire. If you would rather have a steady supply, the monthly Starter ($7.99), Creator ($19) and Pro ($49) plans refill your credits each month, and you can cancel anytime. Unused monthly credits do not roll over.