Cartoon Narrator

An original cartoon guy narrates your topic in a deadpan brainrot voice over satisfying gameplay (Minecraft-style parkour, Subway-style runner, GTA-style driving and more), with reference images popping in and viral captions.

See a finished video

Recipe facts

Output
Video
Category
Brainrot
Visuals
Video
Best model
Wan 2.7

What it automates

1. Plan the video
Turn a topic or script into scenes, narration beats, visual directions, and caption timing.
2. Generate the assets
Run the recommended models for voice, scene media, optional music, and caption transcription.
3. Render the final
Assemble the scenes into a vertical short with captions, audio mixing, and a downloadable MP4.
Open in Compose

Use it via the API

Plan the storyboard first, then render the final video:

Compose plan
curl https://api.empiriolabs.ai/v1/videos/compose \
  -H "Authorization: Bearer $EMPIRIOLABS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "recipe": "cartoon-narrator",
    "mode": "plan",
    "brief": "Your topic or script.",
    "aspect_ratio": "9:16"
  }'
Open the Compose API docs

Common questions

What is this Compose recipe?

A full-production workflow that turns a brief or script into a finished short-form video: it plans the scenes, creates the visuals, adds voiceover and captions, and renders the final MP4.

How is it different from an effect template?

Effect templates apply one curated look to a single generation request. Compose recipes run the whole production and return an assembled video. Browse single-shot effects at Generation Templates.

Can I use it through the API?

Yes. Call POST /v1/videos/compose with the recipe slug. Use mode: "plan" to get an editable storyboard, then mode: "render" to produce the final video.

How does billing work?

Each production step is billed at the normal pay-as-you-go rate of the model it runs, and the steps are itemized in your usage logs. There is no extra recipe fee.

Ready to use better endpoints?

Check out our pricing or reach out if you want your own model deployed on our stack.