> ## Documentation Index
> Fetch the complete documentation index at: https://docs.origami.techmitten.com/llms.txt
> Use this file to discover all available pages before exploring further.

# AI Shorts Generator — Topic to Short-Form Video in Minutes

> Turn any topic into a short-form video: AI scripts, Pollinations visuals, Kokoro TTS voiceover, animated captions, and MP4 export — all in your browser.

Describe a topic and Origami AI builds a complete short-form video for you from scratch. It writes a scene-by-scene script, generates matching visuals via Pollinations, synthesizes a voiceover with Kokoro TTS, burns in animated captions, and exports a finished vertical (or landscape, or square) MP4 — without leaving your browser.

## The Shorts Workflow

<Steps>
  <Step title="Set up your short">
    Enter your topic and configure the shot: choose a target duration (15–90 seconds), a tone (Punchy & fast, Documentary, Storytelling, Educational, or Hype), your preferred aspect ratio (9:16, 16:9, or 1:1), and a visual style such as Cinematic, Photorealistic, 3D Render, Anime, Retro Film, and more.
  </Step>

  <Step title="Review scripts and visual prompts">
    Before anything is generated, Origami AI presents the full scene-by-scene narration scripts and image prompts for your review. Edit any scene's copy or image description to steer the final output exactly where you want it.
  </Step>

  <Step title="Generate visuals and voiceover">
    Click **Generate Images** (or **Generate Video Clips** if you've selected video mode). Origami AI requests visuals from Pollinations for each scene while simultaneously synthesizing Kokoro TTS voiceover audio in parallel — so both tracks are ready at the same time.
  </Step>

  <Step title="Choose a caption style">
    Pick one of six animated caption styles to display your narration on-screen: **Bold Pop**, **Karaoke Fill**, **Clean Lower Third**, **Highlighter**, **Neon Glow**, or **Classic Cinema**. You can also adjust caption size (Small, Medium, Large) and position (Top, Middle, Bottom).
  </Step>

  <Step title="Add music and export">
    Optionally add background music from the built-in library or upload your own track. When you're happy, click **Export MP4** to render the complete short directly on your device.
  </Step>
</Steps>

## Visual Generation with Pollinations

Origami AI generates all images and video clips through [Pollinations](https://pollinations.ai), an external AI visual generation service. Each scene's image prompt is sent to Pollinations and the result is returned directly to your browser.

To get the best experience:

* **Connect your Pollinations account** in **⚙️ Settings → API → Pollinations Account** by clicking **Connect with Pollinations**. Authenticated requests get higher rate limits and priority in the generation queue.
* **Use the Free (slow) model** if you don't have or don't want to connect an account. The free model requires no API key but is queued behind authenticated requests, so generation may take longer per scene.
* If a scene fails to generate, use the **regenerate** button on that scene card rather than restarting the whole project.

<Note>
  Visual generation requires an active internet connection. Pollinations is an external service — your image and video prompts are sent to Pollinations servers to generate the visuals. Audio (TTS voiceover) is always synthesized locally on your device.
</Note>

## Caption Styles

Origami AI offers six caption styles, each with a distinct look suited to different content and platforms:

| Style                 | Description                                                                                                               |
| --------------------- | ------------------------------------------------------------------------------------------------------------------------- |
| **Bold Pop**          | Large, high-contrast bold text that snaps word-by-word — great for attention-grabbing social shorts.                      |
| **Karaoke Fill**      | Words highlight progressively as they're spoken, mimicking classic karaoke timing.                                        |
| **Clean Lower Third** | Understated subtitle-style text anchored near the bottom of the frame. Works well for educational or documentary content. |
| **Highlighter**       | A colored highlight box appears behind each active word, making copy easy to follow at a glance.                          |
| **Neon Glow**         | Glowing neon-style text with a vibrant edge — ideal for high-energy or gaming content.                                    |
| **Classic Cinema**    | Elegant centered titles with a timeless cinematic feel, suited to storytelling and narrative shorts.                      |

## Supported Aspect Ratios

<CardGroup cols={3}>
  <Card title="9:16 — Vertical" icon="mobile">
    The default for Shorts, Reels, and TikTok. Renders at 1080 × 1920 px for full-resolution portrait playback.
  </Card>

  <Card title="16:9 — Landscape" icon="monitor">
    Standard widescreen for YouTube and presentations. Renders at 1920 × 1080 px.
  </Card>

  <Card title="1:1 — Square" icon="square">
    Fits Instagram feed posts and multi-platform repurposing. Renders at 1080 × 1080 px.
  </Card>
</CardGroup>