Turn a photo into an AI influencer that talks.

One photo of the person, a voice recording or the words you want said, a line about the scene — and you have a lip-synced, gesturing video up to 30 seconds, ready for TikTok, Reels and Shorts. Same face every time, any script, no camera, no reshoots.

Try it free

60 free credits on signup. No card required.

What is the AI Influencer Generator?

The Plurel AI Influencer Generator makes talking-head videos of a person who never has to be filmed. Upload one or more photos to fix who they are — face, hair, clothes, the room — add a voice recording (they say it, lip-synced) or just write the words (a voice is generated), and describe the scene and the performance: where they sit, how they act, what they hold up. The model, Alibaba's Wan 3.0 or the faster Wan 3.0 Prime, produces up to 30 seconds at up to 1080p with speech, gestures and expressions matched to the words. Because the same photos go in every time, the same character comes out every time — which is what makes an AI influencer possible: a consistent face across a hundred videos, each with a different script. Product demos, daily tips, UGC-style ads, a character for a brand, a presenter for a channel. Nothing to install, no API key, free credits to start.

What you can do with it

Same person, every video

Your photos are the reference in every generation, so the face, hair and style hold across a whole feed — the one thing an AI influencer needs.

Say it your way

Upload a recording and they say exactly that, lip-synced. Or write the words and the voice is generated. Or upload a recording as the voice and let the model write the line.

Directed, not random

Describe the scene and the acting — "at her kitchen counter, playful, big hand gestures, leans in on the punchline" — and the performance follows.

Products in hand

Add a photo of the product and it appears as pictured — held up, turned, put down — for UGC-style ads and demos.

Up to 30 seconds, vertical

One generation covers a whole TikTok or Reel: 9:16, up to 1080p, with sound. No stitching.

Copy real motion

Add a short clip and the avatar follows its movement and timing — a walk, a dance, a specific gesture.

Made with the AI Influencer Generator

One photo, one voice recording, Wan 3.0 Prime, 8 seconds — the person speaks the recording to camera with synced lips and natural gestures.

AI Influencer Generator key features

Photo + voice → video

Up to 10 photos, 5 recordings (15 s combined), 5 clips to follow (15 s combined).

2–30 seconds

Length follows your recording, or set it yourself.

480p, 720p or 1080p

Drafts cheap at 480p; deliver at 1080p.

Vertical, square or wide

9:16 for TikTok, Reels and Shorts; 1:1, 16:9, 4:3, 3:4 when you need them.

Two speeds

Wan 3.0 for the lowest price per second; Wan 3.0 Prime for the same result in a fraction of the time.

Pay per second

From 2 credits a second. A 10-second 720p clip is 35 credits on Wan 3.0, 55 on Prime — new accounts start with 60 free.

How it works

  1. 1

    Open Video → Avatar

    In the Plurel app choose Video, then Avatar — or start from this page.

  2. 2

    Add the person

    One to ten photos: the face, the look, optionally the product and the place.

  3. 3

    Give them something to say

    Upload a recording — as the words, or just the voice — or write the script and the voice is generated.

  4. 4

    Describe the scene, generate

    Where they are, how they act. Pick 9:16 and a length, tap Make the video. It lands in Media History with sound.

Frequently asked questions

Make your first AI influencer video

60 free credits. No card. Your character, your script.

Try it free