About
P Video 2 Pro Avatar is Pruna AI's talking-head model on the P-Video-2-Pro line. One portrait plus speech in, a lip-synced talking video out. Supply an audio file and the video follows it exactly, or type a script and one of 30 built-in voices in 10 languages reads it. Output is 24 fps MP4 from 1 second to 3 minutes at 480p or 768p.
What to expect
- •Synced lips and expressions driven by the soundtrack; the portrait sets the aspect ratio.
- •Uploaded audio always wins over a script. Audio under 1 second is rejected.
- •Generation runs at roughly 1 to 2.2 seconds per second of output, depending on resolution.
Specifications
- •Input: one portrait (jpg, jpeg, png, webp, up to 20 MiB) and either audio (up to 64 MiB, 3 min) or a script.
- •Output: MP4, 24 fps, 480p or 768p, length equal to the speech.
- •Price: per second of returned video; 768p costs twice 480p.
Availability
Launching Thursday 22 October 2026, 18:00 CEST, with a launch offer of 50% off list for the first two weeks.
