FableFlow

AI-powered multimedia story production system

FableFlow is an open-source platform that converts story manuscripts into multimedia content with narration, illustrations, and video production using AI agents.

---

Overview

FableFlow automates the multimedia book production pipeline using specialized AI agents that handle story refinement, narration, illustration, music, and video assembly.

Core Components:

• Production Pipeline (producer/fable_flow/) - CLI tools for AI-powered content generation
• FableFlow Studio (studio/) - Web workspace for managing projects and running production workflows
• Website (docs/) - MkDocs website for FableFlow users with guides and tutorials

Getting Started

Installation

# Clone and install
git clone https://github.com/suneeta-mall/fable-flow.git
cd fable-flow
python3.11 -m venv .venv
source .venv/bin/activate
make install

Requirements: Python 3.11+, Node.js 18+ (for Studio)

Usage

Option 1: Development Environment (All Services)

make studio-install  # First time only
make dev-all         # Start Studio + Website
# Studio: http://localhost:3000
# Backend API: http://localhost:8000
# Website: http://localhost:8080

make stop-all        # Stop all services

Option 2: FableFlow Studio (Web Interface Only)

make studio-start    # Start Studio only
# Access at http://localhost:3000

make studio-stop     # Stop Studio

Features: Project browser, Monaco editor, version comparison, media gallery, real-time production monitoring.

Option 3: CLI (Command Line)

# Run production pipeline
fable-flow publisher process

# Individual steps
fable-flow illustrator draw
fable-flow narration produce
fable-flow music produce
fable-flow story process

Option 4: Website (Local Development Only)

make serve  # Serve MkDocs website
# Access at http://localhost:8080

Production Pipeline

FableFlow uses specialized AI agents for each production stage:

1. Story Processing - Manuscript refinement and optimization
2. Narration - Text-to-speech audio generation
3. Illustration - AI-generated contextual images
4. Music - Background score composition
5. Video Assembly - Combine all elements into final output

See the complete workflow documentation for more details.

Configuration

Set environment variables for AI model access:

export MODEL_SERVER_URL="https://your_api.com/v1"
export MODEL_API_KEY="your_key"
export DEFAULT_MODEL="your_model"

See documentation for custom model configuration.

Example

The "Curious Cassie" children's book series uses FableFlow to produce educational story books that teach scientific concepts to readers aged 5-10.

Website

FableFlow website at: https://suneeta-mall.github.io/fable-flow

Built with MkDocs (docs/), includes:

• Getting started guides
• CLI command reference
• Production workflow tutorials
• Configuration examples
• Best practices

Local Development

# Serve website locally
make serve
# Access at http://localhost:8080

Contributing

Contributions welcome via:

• Bug reports and feature requests in Issues
• Pull requests for code improvements
• Documentation enhancements

License

Elastic License 2.0 - see LICENSE

Links

• Website: https://suneeta-mall.github.io/fable-flow
• Repository: https://github.com/suneeta-mall/fable-flow
• Issues: https://github.com/suneeta-mall/fable-flow/issues
• Contact: suneetamall@gmail.com

---

Built by Suneeta Mall, and Claude Code.