Feature
Talking Avatar Videos
Turn one photo, an AI influencer, or your own clone into a presenter who speaks your script.
Talking-head videos build trust faster than any other format, but not everyone wants to be on camera every day. Pick a face — a single photo, one of your AI influencers, or a clone of yourself from Clone Studio — then type a script or upload a recording you already have. Landio voices the script, lip-syncs the presenter to it and saves the finished clip to your media library. Choose the engine that suits the job: Higgsfield for quick 15-second hooks, or D-ID, HeyGen, Sync Labs and Tavus for videos up to two minutes.
- Five engines — Higgsfield, D-ID, HeyGen, Sync Labs and Tavus — each used with your own key
- Up to 15 seconds on Higgsfield and up to 2 minutes on the other engines
- Type a script to have it voiced, or upload any MP3, M4A, WAV or MP4 recording
- Change the voice on a recording you made, so your delivery comes out in a different voice
- Present with a photo, an AI influencer, or a face you cloned in Clone Studio
- Voice styles and speaking speed, in the language you choose
- Renders in the background and saves straight to your media library, ready for captions and b-roll
How it works
Talking Avatar Videos in three steps
Choose the presenter
A clear, front-facing photo, one of your AI influencers, or a video clone of you or a teammate.
Add the words
Type the script and pick a voice, or upload a recording and optionally change the voice it's spoken in.
Get a talking video
The video renders in the background and lands in your media library — finish it with captions and b-roll, or attach it to a post.
Who it's for
Built for how you actually work
Founders who hate being on camera
Put a face and a voice on announcements and tips without filming anything.
Short, frequent video updates
Publish a daily tip in the time it takes to write two sentences.
Giving your AI influencer a voice
Pick an AI influencer as the presenter and they speak directly to your audience, in a consistent voice.
Questions about talking avatar videos
What do I need to create an avatar video?+
A face (photo, AI influencer or video clone) and either a script or a recording. You'll also need the key for the engine you pick, added on the AI Providers page — and an OpenAI key if you want scripts voiced for you.
How long can the video be?+
Up to 15 seconds with Higgsfield, and up to two minutes with D-ID, HeyGen, Sync Labs or Tavus.
Does it count against my plan?+
Yes — each avatar video uses one generation from your plan's monthly video allowance.
Can I use a recording I already made?+
Yes. Upload an MP3, M4A, WAV or even an MP4 — the soundtrack is extracted and the presenter lip-syncs to it. You can also swap the voice while keeping your timing and delivery.
Getting started
Up and running in three steps
Nothing here needs a call, a migration, or a connected account to explore first.
Step 1
Create a workspace
Sign up free. No card, and no requirement to connect a social account before you look around.
Step 2
Set your brand profile
Industry, audience and tone. Most features here read from it, so filling it in once improves all of them.
Step 3
Open talking avatar videos
It's available from the workspace sidebar as soon as your plan includes it — nothing to install or enable.
Where it fits
One stage of a connected pipeline
Each stage hands off to the next, which is why this takes fewer steps here than the same job spread across separate tools.
Stage 1
Plan
Brief, brand kit and calendar decide what goes out and when.
Stage 2
Create
AI drafts copy, images and video per platform from that brief.
Stage 3
Approve
Drafts route to a named reviewer before anything reaches a live account.
Stage 4
Publish & learn
A job queue delivers each variant, then feeds results back in.
Explore more