A Talking 3D Avatar from Photos
From face photos, a 3D head can be assembled that talks, blinks, and moves its lips in time with speech.
From photo to model
A few sharp face shots become a three-dimensional head model. The better the photos, the more accurate the likeness.
Expression and visemes
The model is equipped with facial expression and visemes — mouth shapes for the sounds of speech. That's why the lips move in sync with the voice and the face conveys emotion.
2D and 3D
The avatar is available both as a simple 2D view and as a 3D one. Both versions use the same track of emotion and lip movement.
Which photos work
Best of all are sharp front-on and three-quarter shots in even light, with no glaring glasses or hats. Several angles make the model more accurate. One good portrait is more useful than ten blurry ones.
- A few sharp photos → a 3D head.
- Expression and visemes — lips in time with speech.
- Available in 2D and 3D.
- Front-on and three-quarter, even light.
Frequently asked questions
Save the story while it is with you
Create a memorial page in a few minutes — gently, beautifully and with respect for your loved ones. Free forever for the text version.
Create a memorialRead also
AI copy
Visemes and Expression: How a Face Comes Alive
What visemes are, how lips and speech are synchronized, and why facial expression makes an avatar feel alive.
AI copy
Three ways to talk with an AI memory copy
Text, text-to-voice, and voice-to-voice: three ways to talk with an AI memory copy in Pomni, and how they differ.
AI copy
Farewell mode: a gentle way to let go
Why an AI copy needs a farewell mode, and how it helps you move through loss rather than get stuck in dependency.