About StoryPixie

Project Vision

StoryPixie is an AI-powered multimodal storytelling agent designed to bring your imagination to life. By simply entering a text prompt, StoryPixie orchestrates a complex AI pipeline to generate a rich, interleaved experience consisting of story text, AI-generated images, professional-sounding voice narration, and an animated video.

Technology Stack

Frontend

  • Next.js & React
  • Tailwind CSS
  • Axios for API calls
  • Responsive Design & Dark/Light Mode

Backend

  • FastAPI (Python)
  • Gemini AI (Google GenAI SDK)
  • Vertex AI Imagen (Images)
  • Google TTS (Audio)
  • MoviePy (Video)

The AI Pipeline

When you submit a prompt, the following steps occur:

  1. Gemini AI processes your prompt and generates multiple storyboard scenes, including narration text and image prompts.
  2. Imagen (Vertex AI) generates high-quality images based on the specific scene descriptions.
  3. Google Text-to-Speech converts the story text into a natural-sounding audio narration.
  4. MoviePy compiles the images and audio into a final animated story video.
  5. Everything is served via a unified interface, creating a seamless storytelling experience.