Pruna AI built P-Video-Avatar in the end of April 2026 as a tool that turns a single portrait into a video of a person talking. It works by taking your photo and matching it up with either a typed script or a recording you upload. Strikingly, the goal here is to make these videos fast and cheap-ish, which is great for marketing or customer support stuff. You can pick between 720p or 1080p for the final video (depending on how sharp you need it to be).
This model is a bit more focused than your average video generator because it only cares about animating the face and head. It adds mouth movements and expressions to make the person look like they're actually speaking. Teams use it for things like product demos, training videos, or social posts. And that's helpful, because you don't have to set up a camera or a studio every time you need a quick clip.
Pruna AI is a company from Germany and France that specializes in making AI run more efficiently. They claim this model is way faster and cheaper than others, though you should probably treat those specific numbers as sales talk. To be clear, you just upload a picture, give it some audio, and it spits out an MP4.
The software doesn't just move the lips, it handles the whole talking-head vibe with natural movements, actually. Which is a pretty big deal, and it's why it works so well for personalized sales or announcements, I guess.
AI Avatar Videos. The system creates talking videos from a single portrait image using lip-sync tech.
Multiple Input Options. You can generate speech by typing a script or by uploading your own audio file.
High Resolution Support. The API allows you to choose between 720p and 1080p for the final video output.
Efficient Commercial Use. It is designed for businesses needing fast, low-cost videos for training or social media.
Flexible Controls. Users can adjust the speaking style, use specific voices, and set seeds for consistent results.
