Maya product presenter
A friendly product presenter built from a generated portrait and an Eleven V3 voice.
View prompt
Natural head movement, subtle facial expression, stable identity, clean speaking performance, realistic motion.
Pruna AI avatar video model
Use it when a portrait and voice track need to become a short talking presenter with controllable body motion.
Your generated avatar will appear here.
Upload a portrait image, add voice audio, and click 'Generate Avatar' to start.
Examples
Hover a card to preview, then open the player to watch with sound and review the full prompt.
A friendly product presenter built from a generated portrait and an Eleven V3 voice.
Natural head movement, subtle facial expression, stable identity, clean speaking performance, realistic motion.
Model context
P Video Avatar is a direct image-and-audio avatar model. Give it a portrait, a speech track, and a short motion prompt, then review the talking result.
It is useful for product explainers, onboarding snippets, social announcements, and character clips where speed and controllable presentation matter.
Model strengths
Use it when a portrait and voice track need to become a short talking presenter with controllable body motion.
Prompt guide
A clear face, even lighting, and simple background give the model fewer identity problems to solve.
Trim silence and fix pronunciation before generation. The video follows the audio.
Prompt head movement, expression, and framing. Avoid exaggerated gestures unless the source image supports them.
Review mouth shape, hairline, eyes, and face shape across the full clip.
It needs a portrait image and speech audio. The motion prompt guides how the avatar behaves.
Use a sharp front-facing or slight three-quarter portrait with visible facial features and simple lighting.
Yes, through the video prompt. Keep movement modest for the most stable output.
Use Omni Human 1.5 when high-fidelity portrait performance matters more than quick presenter iteration.