Video API

Text in, avatar video out

Generate a video from text with a POST request and get back a video URL. Pay for the seconds of video you receive.

Free time to evaluate
No per-seat pricing
Failed generations are not charged

A request from your own code

curl -X POST http://avatartalk.ai/api/inference \
  -H "Authorization: Bearer $KEY" \
  -H "Content-Type: application/json" \
  -d '{"text":"Hello and welcome.",
       "avatar":"european_woman",
       "emotion":"neutral",
       "language":"en"}'

The response includes a playable URL and the credits used.

Generate, play and download from one panel

Try a request in the browser, manage your API key and see generation history in the Video API panel. The same account and credit balance also power your hosted customer assistants.

Try the playground

Integrate with your stack

Start with an example, or connect your existing tools

Our public examples and API reference show the request format. You can also use existing integrations instead of building an HTTP client from scratch.

API reference

Request fields, authentication and responses in the AvatarTalk examples repository.

Read the docs →

GitHub examples

Sample projects and integration patterns you can adapt to your application.

Browse examples →

Make.com

Use AvatarTalk in your Make scenarios to automate video generation.

See the integration →

n8n

Add avatar generation to an automation workflow without writing the whole integration yourself.

See the integration →

LiveKit Agents

Explore the AvatarTalk plugin for real-time agent projects. This is separate from the text-to-video endpoint above.

View the plugin →