AI avatar video generator

Give your script a face.

Write a script, pick a presenter and a voice, and approve an itemized credit quote; Shotmatic then renders a talking presenter video with captions.

Make a presenter video Run it from the API

A Talking Avatar video puts a presenter on screen to deliver your script: an explainer, a product introduction, a lesson or an announcement. You write the words, choose one of Shotmatic's featured presenters, and pick the voice and the layout.

Presenter rendering costs more than a narrated slideshow, so Shotmatic prices your exact script and presenter first. You see the itemized quote, approve it, and only then does the render start.

Examples made in Shotmatic

Real outputs from Shotmatic's own account, with the settings behind each one and the credits it was charged.

Presenter tip: start with one clear idea

A portrait Talking Avatar video: a featured presenter reads a two-sentence script with word-timed Viral captions.

Format: Talking Avatar
Presenter: Kinsley, a featured presenter
Layout: Full presenter
Voice: Bella
Captions: Viral
Length: 9 seconds
Aspect ratio: 9:16
Music: Off
Credits charged: 168

Script

Start with one clear idea. Keep your original visuals, choose a natural voice, and review every scene before you share your finished video.

Made by Shotmatic on its internal QA account from a script Shotmatic wrote. Kinsley is one of Shotmatic's featured presenter profiles, and Bella is a curated catalog voice whose rights scope is commercial video.

Make one like this

Presenter tip: review before you export

A landscape Talking Avatar video: a featured presenter in an office reads a one-line tip, with Viral captions at the top of the frame.

Format: Talking Avatar
Presenter: Dashiell Office 10, a featured presenter
Layout: Full presenter
Voice: Darian
Captions: Viral
Length: 6 seconds
Aspect ratio: 16:9
Music: Off
Credits charged: 167

Script

Make one useful video. Review the wording, check captions, and export when you are happy with it.

Made by Shotmatic on its internal QA account from a script Shotmatic wrote. Dashiell Office 10 is one of Shotmatic's featured presenter profiles, and Darian is a curated catalog voice whose rights scope is commercial video.

Make one like this

AI actors

"AI actor" is what many tools call a generated presenter who reads your script. In Shotmatic that is the Talking Avatar format: the presenter speaks your exact words in the voice you choose, full frame or layered with supporting visuals.

Two things are worth knowing before you compare tools. An AI actor here presents; it does not hold or demonstrate a physical product. For product ads built from your own product footage, use the AI video ad generator, which has no presenter. And a presenter based on a real person needs that person's recorded consent before Shotmatic creates it.

Each featured presenter comes with a matched default voice, and you can choose another before you approve.

Why the quote comes first

The quote prices the presenter for your exact script, the voice and the render, and it is the most you can be charged. The presenter is billed for the seconds the narration really lasts, so the charge can only come in under the quote.

How it works

01
Script and avatar
Your script, a presenter and a voice.
02
Layout and settings
Choose how the presenter shares the screen, the aspect ratio and the caption style.
03
Itemized quote
Shotmatic prices your exact script and presenter. Nothing renders until you approve it.
04
Presenter video
The presenter delivers your script with captions. Download the MP4 and post it on your channels yourself.

What you control

Presenter: A featured presenter, each with a matched default voiceA presenter based on a real person needs that person's consent.
Layout6 options: Full presenter · Intro + outro · Split screen · Floating host · Presenter first · Scene cameosSome presenters suit some layouts only; the creator marks which.
Aspect ratio4 options: Vertical 9:16 · Social portrait 4:5 · Square 1:1 · Horizontal 16:9
Caption style4 options: Viral · Bold · Minimal · Story
Script language8 options: English · Spanish · French · German · Italian · Dutch · Portuguese · Portuguese (Brazil)

Limits and availability

Rendering starts only after you approve the quote for your exact script and presenter.
A presenter based on a real person needs that person's recorded consent before Shotmatic creates it.
Presenters read your script; they do not hold or demonstrate a physical product. The AI video ad generator makes product ads without a presenter.
Talking Avatar videos can't be scheduled as a sequence; each one is quoted and approved on its own.
Shotmatic doesn't post anything for you. Download the finished MP4 and upload it to your channel or ad account yourself.

What it costs

The two examples above cost 167 and 168 credits, from their receipts.

A full-frame presenter video starts at 105 credits, and each second of narration adds about 8 (7.2 of them for a featured presenter), so a 30-second script comes to about 346. Layouts with supporting visuals add their scenes. The itemized quote shows your exact maximum before you approve, and the presenter is charged for the seconds the narration really lasts.

See plan prices and monthly credits

Questions

What is an AI actor?

A generated presenter who reads your script on camera. In Shotmatic you pick one in the Talking Avatar format, and the presenter speaks your exact words in the voice you choose.

How much does an avatar video cost?

From 105 credits for a full-frame presenter; a 30-second script comes to about 346. Each second of narration adds about 8 credits (7.2 of them for a featured presenter), and the quote shows your exact maximum before you approve.

Can I change the voice?

Yes. Each featured presenter starts with a matched voice, and you can choose another before you approve.

Does the presenter have to fill the frame?

No. Besides a full-frame presenter, layouts open and close on the presenter, split the screen, or place the presenter over supporting visuals: Full presenter, Intro + outro, Split screen, Floating host, Presenter first and Scene cameos.

Can a presenter be based on a real person?

Only with that person's recorded consent, which Shotmatic asks for before it creates the presenter.

Run it from code or an agent

The hosted MCP server and the REST API run the same creation steps. This is the plan_video call for this format; angle brackets mark your own values.

{
  "tool": "plan_video",
  "arguments": {
    "goal": "A 30-second presenter video introducing our onboarding checklist",
    "script": "<your reviewed script>",
    "avatar_profile_id": "<id from shotmatic://avatar-profiles>",
    "voice_id": "<the profile's default voice id>"
  }
}

Recommends a Talking Avatar with the avatar_spokesperson archetype. After create_session, quote_avatar_session prices it, and approve_and_render takes the quote it returns.

Read the API and MCP guide