AI Creative OS
01Use Omnia, not Character 302Feed it all three inputs03Direct the shot, not the face04Build around the 8 seconds

A Talking Video Of You, Free

One photo, one audio track, one prompt, and you have a talking video of yourself. The free omni model changed since everyone posted about it, so here is what is actually free today. Verified live on hedra.com, July 2026.
Build it step by stepInteractive · ~2 min · with Z’s build guideStart →
01This is the flip
Use Omnia, not Character 3
Everyone's viral clip said Character 3. On Hedra's own model page today that one reads Free tier: No at 8 credits a second. Hedra Omnia reads Free tier: Yes. Same company, and the free one is now the better one.
02Omnimodal
Feed it all three inputs
Hedra describes Omnia as generating video “from a single reference image, an audio track, and a text prompt.” Most people skip the prompt and wonder why it looks stiff. The prompt is what buys you the performance.
03The upgrade
Direct the shot, not the face
Omnia gives “direct control over camera movement, subtle facial expressions, and backgrounds while maintaining consistent lip-sync.” So write the prompt like a director: the move, the light, the room. Not just the emotion.
04The real limit
Build around the 8 seconds
Max duration is 8s per clip at 15 credits / second, so a clip costs about 120 credits. Do not fight it. Script in 8-second beats and cut between them, which is what a good short does anyway.
The prompt shape that stops it looking like AI
[shot] slow push in, subject centred
[light] warm key from camera left, soft falloff
[action] he leans toward the camera on the first line
[room] out-of-focus desk, no moving objects
# describe the ACTING and the SCENE, the audio handles the mouth
Want this running for you?
Build it step by step inside AI Creative OS, with me in the thread.
Implement this today →
Zdeno Mucina
Not ready for that?
Drop your email and 3 more guides open in the Library, free.
Download as PDF ↓