Portrait image
Start from a clear, rights-cleared portrait with a stable subject, even lighting, and a face that reads well at small sizes. A clean frontal or near-frontal shot gives the avatar workflow the most to work with.
Portrait + audio → short talking video drafts
Create avatar-style short videos from a portrait image and an audio track — built for social clips, talking concepts, product updates, announcements, and lightweight presenter drafts you can put together right in the browser.
✨ 60 free signup credits · Credits do not expire
Your result appears here
P Video Avatar · 720p · 10s
Avatar workflow
P Video keeps avatar creation deliberately lightweight. Instead of writing a scene from scratch, you start from a face and a voice: upload a rights-cleared portrait, attach the driving audio, keep the video prompt to a short phrase, and generate a brief talking-style draft you can preview and download in a few moments.
That makes Avatar the fastest path from a written update to something that looks and feels spoken. It is tuned for the moment before a real recording — when you want to see how a talking concept reads on screen, test the pacing of a delivery, or share a presenter-style draft in a thread or a deck without booking a shoot or briefing talent.
Avatar is one of three focused workflows. For clips that start from a sentence or a reference image with no audio at all, use the P Video AI Video Generator. When the goal is to drive a still image with reference motion rather than a voice, switch to P Video Animate. Every workflow keeps the same light console rhythm, so moving between them feels like one product rather than three separate tools.
What you provide
Avatar exposes only the inputs that matter for a talking-portrait clip — a face, a voice, and a light touch of direction.
Start from a clear, rights-cleared portrait with a stable subject, even lighting, and a face that reads well at small sizes. A clean frontal or near-frontal shot gives the avatar workflow the most to work with.
Add the audio track that should drive the clip. Keep it concise while you are testing so each generation stays quick to review and easy to compare against the take before it.
Describe the expression, energy, or subtle motion in a short phrase rather than a full script. A compact prompt — calm delivery, slight head movement, steady framing — keeps the result anchored on the portrait.
Workflow
Pick a rights-cleared image with a readable face and stable framing. This is the subject the avatar draft is built around, so a clean, well-lit shot pays off across every later iteration.
Attach the audio track that should lead the clip. Shorter audio keeps tests fast and cost predictable while you settle on the portrait, the tone, and the framing you want.
Use a short phrase for expression and motion instead of a long description. One clear direction reads more naturally in a brief avatar clip than several competing instructions.
Run the draft, preview the result, and keep the takes that land. Saved generations stay in your library so you can compare versions before committing to one direction.
Inside the console
Avatar pairs your portrait with the driving audio and a compact prompt, then renders a brief talking-style clip. Per-second rates are listed on P Video Pricing.
Prompting
A few habits keep avatar clips steady and readable while you iterate on the portrait and audio.
Lead with the tone you want — calm, upbeat, conversational — so the expression matches the audio rather than fighting it. A single clear delivery note keeps short avatar clips readable.
Call out gentle, intentional movement like a slight head tilt or natural blink instead of large gestures. Subtle motion holds the likeness steadier across the length of a short draft.
Trim the audio to the segment you actually want to test. Because avatar generation is billed per second, shorter audio means faster, cheaper passes while you iterate on the look.
Adjust a single variable per take — the portrait, the audio, or the expression phrase — so you can read what changed. Focused iterations converge faster than rewriting everything at once.
Use cases
Lightweight talking-style clips for the updates, explainers, and concepts you make many of, quickly.
Pricing
P Video Avatar costs 6 credits per second at 720p and 12 credits per second at 1080p, so a trimmed audio track keeps each test inexpensive. New accounts start with 60 free credits, credits do not expire, and you can compare credit packs on P Video Pricing.
FAQ
P Video Avatar works from a portrait image plus a driving audio track, with an optional short video prompt for expression or subtle motion. You upload the portrait, attach the audio, add a brief direction if you want one, and generate a short talking-style draft in the browser.
Avatar generation is billed per second of output: 6 credits per second at 720p and 12 credits per second at 1080p. Because cost scales with duration, short audio keeps each test inexpensive while you refine the portrait and delivery. Full credit packs are listed on the pricing page.
Use a clear, rights-cleared image with a stable subject, even lighting, and a face that stays readable at small sizes. A clean frontal or near-frontal shot gives the avatar workflow the most consistent base to build a short draft around.
No. Only upload faces, portraits, audio, and likenesses you have the rights and consent to use. Uploading someone else's image or voice without permission is not allowed, and rights-infringing or impersonating content is prohibited under the Terms of Service.
Use 720p while you experiment with the portrait, audio, and expression so passes stay fast and affordable. Switch to 1080p once the direction is locked and a draft is worth a higher-quality render. You only pay the higher per-second rate when a take is ready.
No. P Video Avatar runs entirely in the browser as part of the online product. There is no local model download and no GPU setup for the online workflow — you upload your inputs, choose your settings, and generate.
Generated works are retained in your library for 6 months, so you can revisit, compare, and download earlier avatar drafts while you iterate. New accounts also start with 60 free signup credits, and credits do not expire.
Avatar is the audio-driven, talking-portrait workflow. If you only have an idea or a still image and no audio, the P Video AI Video Generator handles text and image to video. If you want to drive a still image with reference motion instead of audio, P Video Animate is the better fit.