Storybook turns alternating image descriptions and dialog into narrated videos with music.

Quick start

• Install deps: pip install -r requirements.txt (or sudo apt install python3-pyqt5 python3-aiohttp)
• Add keys to creds.py: OpenAI, ElevenLabs, Stability, optional Jamendo
• Create dirs: mkdir -p out/images out/dialog
• Run: python3 main.py

Branches

• gui: Shorts (9:16 portrait), optimized overlays and text wrapping
• gui-longform: Landscape (16:9), longer videos

Workflow in the app Simple Mode (recommended)

1. Type a quick idea (optional) and click Make Video
   • If only an idea is provided, the app writes the script for you
   • It generates TTS, images, auto-picks music, and compiles the video
   • Toggle “Use AI video” to try Runway-generated clips instead of images

Advanced Mode (optional)

• Prompts, seeds, reference image, Runway overrides/clip settings
• Helper text explains each option; defaults work well for most cases

Script and Runway

• Generate Script: turns an idea into a story (OpenAI)
• Runway: one-click single clip, or multi-clip (seconds per clip + max clips); concatenates and mixes music

Notes

• Set RUNWAY_API_URL and RUNWAY_API_KEY env vars to enable Runway prototype
• For Jamendo auto-music, set jamendo_client_id in creds.py