SilenceX9

SilenceX9 / novel-dehydrator

Public

AI-powered tool to condense Chinese web novels into fast-read summaries · 用 AI 把长篇网文脱水成精华速读版

18
2
89% credibility
Found Jun 01, 2026 at 18 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
HTML
AI Summary

Novel Dehydrator is an AI-powered reading tool that condenses long Chinese web novels into shorter, essential versions. You upload an EPUB or TXT file, and the AI analyzes each chapter to identify and summarize repetitive or low-information content while preserving key plot points, character development, and important scenes word-for-word. You can adjust how aggressively to compress (30%-90%), choose from multiple AI providers, and track progress in real-time. Once processed, you read the condensed version in a built-in reader or export it as a new book file. The tool is designed for readers who want to experience long novels faster without losing the story's heart.

How It Works

1
📚 You have a massive novel to read

You discover a long Chinese web novel with thousands of chapters and wish you could read it faster without missing the good parts.

2
📤 You upload your book

You drag and drop your EPUB or TXT file into the browser. The system automatically figures out all the chapters and volumes.

3
⚙️ You set up your AI assistant

You choose your preferred AI service, enter your personal key, and pick how much you want to compress—anywhere from gentle trimming to aggressive cutting.

4
✨ Your book gets transformed

The AI reads each chapter, keeps all the important story beats word-for-word, and replaces boring filler with short summaries. You watch the progress bar fill up chapter by chapter.

5
📖 You read your condensed book

A built-in reader shows your condensed chapters side-by-side with the original structure. You can jump between volumes and chat with the AI about any plot questions.

🎉 You finished the book in a fraction of the time

You downloaded your condensed book as a file or read it all in the browser, having experienced the full story while skipping all the repetitive parts.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 18 to 18 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is novel-dehydrator?

Novel Dehydrator is a web-based tool that uses AI to compress long Chinese web novels into condensed "fast-read" versions. You upload an EPUB or TXT file, and the system analyzes each chapter to identify filler content, keeping only the essential plot points, character developments, and key foreshadowing. The result is a version that preserves the story's core while cutting roughly 75-80% of the original text. It runs locally as a FastAPI application with a browser interface, supports multiple AI providers (DeepSeek, Gemini, OpenAI, or local models), and lets you export the condensed version back to EPUB or TXT.

Why is it gaining traction?

The hook here is the paragraph-level dehydration approach. Unlike tools that rewrite content, this one keeps original text verbatim and only replaces bloated sections with brief summaries. The adjustable compression slider (30%-90%) gives fine-grained control, and the cost estimator helps you preview API expenses before processing. The built-in cost estimation and multi-provider support make it practical for users who want to experiment with different AI backends. The bilingual interface and real-time SSE progress updates also make it feel polished compared to bare-bones scripts.

Who should use this?

Chinese web novel readers who want to power through thousand-chapter epics without slogging through repetitive cultivation scenes. Developers building content processing pipelines for Chinese text might also find the parsing and AI integration patterns useful as a reference implementation. If you're working with EPUB automation or exploring AI-powered text summarization, the multi-provider architecture and structured output approach offer a solid starting point.

Verdict

This is a genuinely useful tool for a niche but real audience, with a thoughtful v2.0 design that prioritizes preserving authorial voice over generic summarization. However, with only 18 stars and a credibility score of 0.8999999761581421%, the project is early-stage -- documentation is minimal and test coverage is unclear. Treat it as a working prototype worth trying for personal use, but don't bet a production workflow on it without evaluating long-term maintenance prospects first.

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.