Start with a script
Choose this when you have copy but no recording. Vimorph creates the speech first, then uses it to drive the portrait.
AI talking photo
Turn one clear portrait into a talking video
Write a script and choose a voice, or upload audio you have already recorded. It works well for product explainers, course intros, announcements, and short-form videos.
Clear portraitText or audio
Use text or bring your own audio
Upload recorded audio
Use a finished recording when its pace, pauses, and delivery should carry into the video.
From portrait to finished clip in three steps
Choose a suitable portrait
One person, a clear face, and a front or slightly angled pose work best.
Add what they should say
Enter a script and choose a voice, or upload an audio file.
Review and download
The task runs in the background. Check the mouth movement and frame before saving the result.
Check these four things before uploading
- There is one main person in frame
- The face is large enough to see clearly
- Nothing covers the mouth or chin
- Lighting is even, without harsh shadow or blown highlights
Some portraits are simply a poor fit
Group photos, strong side profiles, low-resolution screenshots, and covered mouths are more likely to fail or produce awkward movement. A better source photo usually helps more than submitting the same one again.
Only use images and voices you have permission to process.
Frequently asked questions
Do I need to upload audio?
No. Enter a script and choose a voice, or upload your own audio when timing and delivery already matter.
Which photos work best?
Use one clearly visible person, facing forward or at a slight angle, with even light and no object covering the mouth.
What determines video length?
The generated video follows the length of the synthesized or uploaded audio.
Can I use someone else's photo?
Only use a person's image and voice when you have permission to create and publish the result.