# ImmersiveReader（沉浸阅读）

> A local Windows reading system that unifies Zhihu archiving, podcast transcription, and Markdown deep-reading in one library. Spotlight focus mode is inherited from MMbook.

ImmersiveReader is the successor to **MMbook**, **PodcastTranscriber**, and **Zhihu Packer**. Those repositories are deprecated/archived; all new development continues here.

## What ImmersiveReader does

- **Unified local library**: books from Zhihu, manual Markdown imports, and other sources live under one Documents library with per-book `manifest.json` and reading progress.
- **Markdown deep reading**: spotlight focus mode, themes, search, TOC, inline paragraph editing (from MMbook).
- **Podcast transcription**: audio/video → polished bilingual Markdown notes (from PodcastTranscriber).
- **Zhihu archiving**: pack long-form Zhihu content into the same library (from Zhihu Packer).
- **Deep read + continuous read**: desktop immersive reading; browser continuous reading shares progress with the desktop app.
- **Privacy-first**: content, progress, credentials, and models stay on the local machine.

## Links

- Product site (this page): https://immersivereader.vercel.app/
- Source code: https://github.com/lingcang728/ImmersiveReader
- Issues / feedback: https://github.com/lingcang728/ImmersiveReader/issues

## Deprecated predecessors

- MMbook: https://github.com/lingcang728/MMbook (archived)
- PodcastTranscriber: https://github.com/lingcang728/PodcastTranscriber (archived)

## Stack

- Desktop: Tauri 2 + Svelte 5 + TypeScript + Rust
- Podcast runtime: Python + faster-whisper (+ optional DeepSeek / Ollama)
- Zhihu runtime: Node + local Chromium profile
- Storage: local filesystem library under Documents

## Author

凌苍 (LingCang) — independent developer.