AI talking avatar generator: text to video

Six original presenters. Type a line, pick who says it, get the clip. Nothing of you on file.

Pick a character Free to try · no card

What it does

DeepMade is a text-to-video AI avatar generator with six original presenters. Pick a character, type a line, and get a talking video back with the mouth matched to the words. No camera, no upload, no signup.

Each character has its own voice and personality, from a measured narrator to an upbeat host, so the same script reads differently depending on who delivers it. Meet the six characters.

How to make a talking avatar video

  1. Pick a character. Six to choose from, each with a sample so you can hear the voice first.
  2. Type the line. Say exactly this, or give the character something to reply to and let it answer in its own words.
  3. Generate. Voice first, then video. Download it or share the link.

What people use AI presenters for

  • Explainer and tutorial videos where a face on screen keeps attention better than a voice-over.
  • Product and update announcements without anyone having to be on camera.
  • Character-driven social content: the same presenter across a whole series.
  • Localised versions of one script, delivered by the same face each time.

Want the presenter to be you?

That is the deepfake video maker: upload a short clip of yourself, clone your voice, and type the line. Same workflow, your face.

Questions people ask

What is an AI talking avatar?

A ready-made on-screen presenter that speaks whatever you type, with matching lip movement. Ours are six original characters, each with a distinct voice, so you can make a talking video with no camera, no upload and nothing of you on file.

Do I need an account?

No. DeepMade characters work without signing up. An account only adds a Library to keep your clips in.

Can I use the characters commercially?

Yes, on any plan. The characters are original, not based on real people, so there is no likeness to clear. Free clips carry a small watermark; Creator and Pro do not.

Can the avatar reply in character rather than read a script?

Yes. Choose 'reply to this in character', paste the message or question, and the character answers in its own voice and personality.

How long can a clip be?

Up to 15 seconds on the free plan, 60 seconds on Creator and Pro.

How is this different from a deepfake?

The characters are synthetic. Nobody's face or voice was cloned to make them. If you want a video of yourself, that is the deepfake video maker.

Pick a character

See also: Deepfake video maker · Voice cloning · Pricing