About SceneWoven

What is SceneWoven?

SceneWoven is a reading companion that brings your books to life. Upload any EPUB and SceneWoven will analyze the chapters, identify key narrative moments, and generate original imagery for the scenes, characters, and locations that make the story memorable.

The result is an illustrated reading experience that stays true to the book — no summaries, no spoilers, just the visual context that helps you stay immersed.

How It Works

When you upload a book, SceneWoven reads the full text to understand the cast of characters: their appearance, personality, and role in the story. It then works through each chapter, finding the moments worth visualizing — a character's first appearance, a pivotal location, a turning point in the plot.

Those moments are turned into AI-generated images using a combination of large language models and image generation providers, then anchored inline to the exact passage they came from. As you read, the visuals surface at the right moment.

The Technology

SceneWoven is built on a modern web stack: Next.js, TypeScript, and PostgreSQL for the application layer; OpenAI and Google Gemini for language and image generation; and AWS S3 for asset storage. A lightweight background job system handles the processing pipeline asynchronously so the app stays responsive while your book is being prepared.

Why We Built It

Reading is one of the most personal experiences there is, yet the tools for readers haven't changed much. SceneWoven started as an experiment: what if the gap between text and imagination could be narrowed? What if the characters you picture in your head could be reflected back to you on the page?

We're still exploring that question. SceneWoven is an early-stage project and we're actively developing new features, improving image quality, and refining how visual moments are selected and presented.

Get in Touch

Have feedback, ideas, or a book you think SceneWoven should know about? We'd love to hear from you via the contact page.