PixelPanel - Multimodal Comic Generator

PreviousNext

Create comics panel-by-panel with AI. Sketch modifications, add voice narration. Built at AI Tinkerers hackathon.

PixelPanel - Multimodal Comic Generator

PixelPanel

Overview

PixelPanel is a multimodal comic generator that transforms simple prompts into complete illustrated stories with narration (AI Tinkerers Ultimate Agents hackathon winning project). The project is open source and available on GitHub.

Key Features

  • Panel-by-Panel Creation: Build your comic incrementally with simple text prompts for each panel
  • Iterative Editing: Sketch modifications directly on panels or refine with additional prompts
  • Voice Narration: Add professional voice-over narration using ElevenLabs text-to-speech

How It Works

Creating Your Comic

The workflow is simple:

Start with a prompt: Describe your first panel. "A superhero overlooking the city at sunset" or "a cozy coffee shop on a rainy morning."

Generate and refine: PixelPanel creates the image. If it's not quite right, sketch modifications or add a follow-up prompt like "make the sunset more dramatic."

Build the story: Move to the next panel. Each new prompt builds on your narrative, creating visual continuity across panels.

Add narration: Once your panels are ready, write the narrative text and generate voice-over. The audio syncs with each panel automatically.

Export: Download your comic as individual panels, or export as a video with narration included.

Real Examples

Mystery Story

  • Panel 1: "Detective examining clues in a dimly lit office"
  • Panel 2: "Close-up of a mysterious letter with cryptic symbols"
  • Panel 3: "Detective rushing out into the night"
  • Narration: "Detective Morris knew time was running out..."

Fantasy Adventure

  • Panel 1: "Warrior standing at the edge of an enchanted forest"
  • Panel 2: "Magical creatures watching from behind ancient trees"
  • Panel 3: "The warrior discovering a glowing artifact"
  • Narration: "The prophecy had spoken of this moment..."

Technologies Used

  • ElevenLabs Text-to-Speech: Professional voice narration generation
  • Nanobanana: AI-powered image generation with iterative editing

Getting Started

Using the Hosted Platform

The easiest way to get started with PixelPanel is to use our hosted platform:

  1. Visit pixelpanel.co
  2. Sign up for a free account
  3. Start creating your first comic with simple text prompts
  4. Add voice narration using ElevenLabs integration
  5. Export and share your completed comics

Local Development

For developers who want to run PixelPanel locally or contribute to the project, detailed setup instructions are available in our GitHub repository.

The project includes:

  • Frontend: Next.js 15 with TypeScript and Tailwind CSS
  • Backend: FastAPI with Python
  • Database: Supabase (PostgreSQL)
  • AI: Google Gemini 2.5 Flash for image generation
  • Voice: ElevenLabs for text-to-speech narration

Demo

What's Next?

The next major feature is character voice dialogue. Right now, PixelPanel handles narrator voice-over, but we want to add speech bubbles with actual character voices. Imagine creating a comic where each character has their own distinct voice, generated through ElevenLabs' voice cloning or character presets.

We're also exploring:

  • Collaborative editing where multiple creators work on the same comic
  • Integration with popular comic platforms for direct publishing
  • Advanced panel layouts and transitions

Create comic with just your imagination.

Try it yourself and let us know what you create.

Acknowledgments

Team Members:

  • Jamie Ogundiran
  • Natalie Chan
  • Maks Sekuła
  • David Szalai