H3 Max Lip Sync
H3 Max Lip Sync is now the short-clip image model in Talking Head. It animates a still portrait to match spoken audio through fal.
Start from a portrait and speech
Open Talking Head with an image. Write a script and choose a DayGen voice, or upload your own audio. H3 Max Lip Sync needs about 5 to 14.8 seconds of audio. Choose 480p, 768p, 1080p, or 2K; 768p is the default.
- Use Voice when you want DayGen to speak the script.
- Use Custom Audio when you already have the soundtrack.
Choose Auto, H3 Max, or OmniHuman
Auto measures the finished audio and uses H3 Max Lip Sync when the duration fits. Longer clips stay on OmniHuman. Explicit H3 Max keeps that model and reports an error if the audio is too short or too long. Video sources still use Sync.
Use it in Talking Head
The same image Talking Head path is available from the Talking Head route, gallery and full-size actions, Character sources, Brand, and Team. This is not a Video Create model, and it does not replace H3 Max or H3 Max Turbo for text-to-video.