<h1 align="center"> CaptiPrep - Your AI-Powered Language Learning Assistant </h1> <a href="https://github.com/jeanchristophe13v/CaptiPrep"> <img src="icon.png?raw=true" alt="CaptiPrep Icon" title="CaptiPrep Icon" width="250"> </a> English | <a href="README.zh_CN.md">简体中文</a> | <a href="README.es.md">Español</a> | <a href="README.fr.md">Français</a> | <a href="README.de.md">Deutsch</a> | <a href="README.ja.md">日本語</a> | <a href="README.ko.md">한국어</a> | <a href="README.ru.md">Русский</a> Preview key vocabulary from YouTube subtitles before you watch – then watch with confidence and context.

CaptiPrep = Caption + Preparation. It helps you learn languages efficiently by turning any YouTube video with subtitles into a targeted pre‑study session. First, preview and internalize key words & expressions. Then watch the video; the built‑in context cements memory naturally. Contextual learning becomes lighter, faster, and more fun.

🤔 What can it do?

1｜Auto subtitle capture: pull raw subtitles & extract candidate words/phrases.

2｜Smart filtering & flashcard generation: AI does an initial sift; you fine‑tune the selection; flashcards are generated automatically—start studying instantly.

3｜Wordbook: a persistent library of videos & learned vocabulary for spaced review and export.

4｜Multi‑language gloss settings: learn any language, explained in the language you prefer.

<table> <tr> <td align="center"> <a href="https://github.com/user-attachments/assets/d14eb772-8679-44e6-9c90-9376567bea4b"><img src="https://github.com/user-attachments/assets/d14eb772-8679-44e6-9c90-9376567bea4b" alt="Chinese (Simplified)" width="300" /></a> Simplified Chinese </td> <td align="center"> <a href="https://github.com/user-attachments/assets/543fa9d7-024c-4619-bb55-cec191606b34"><img src="https://github.com/user-attachments/assets/543fa9d7-024c-4619-bb55-cec191606b34" alt="Chinese (Traditional)" width="300" /></a> Traditional Chinese </td> <td align="center"> <a href="https://github.com/user-attachments/assets/074c2f62-1c93-43d6-b9ee-ceb86a307e3c"><img src="https://github.com/user-attachments/assets/074c2f62-1c93-43d6-b9ee-ceb86a307e3c" alt="English" width="300" /></a> English </td> </tr> <tr> <td align="center"> <a href="https://github.com/user-attachments/assets/a259136a-9f58-4c0f-9474-c34a7d557940"><img src="https://github.com/user-attachments/assets/a259136a-9f58-4c0f-9474-c34a7d557940" alt="Spanish" width="300" /></a> Español </td> <td align="center"> <a href="https://github.com/user-attachments/assets/4d028322-914a-4171-8a2a-e3d1193cfac7"><img src="https://github.com/user-attachments/assets/4d028322-914a-4171-8a2a-e3d1193cfac7" alt="Japanese" width="300" /></a> 日本語 </td> <td align="center"> <a href="https://github.com/user-attachments/assets/3c51cafe-34c7-4243-8670-5f87be6407fc"><img src="https://github.com/user-attachments/assets/3c51cafe-34c7-4243-8670-5f87be6407fc" alt="Korean" width="300" /></a> 한국어 </td> </tr> <tr> <td align="center"> <a href="https://github.com/user-attachments/assets/acc893fc-b00c-4d16-ac1a-18e22f5bba96"><img src="https://github.com/user-attachments/assets/acc893fc-b00c-4d16-ac1a-18e22f5bba96" alt="French" width="300" /></a> Français </td> <td align="center"> <a href="https://github.com/user-attachments/assets/dff198e4-8a04-4282-ad61-7cb0e29a519b"><img src="https://github.com/user-attachments/assets/dff198e4-8a04-4282-ad61-7cb0e29a519b" alt="Russian" width="300" /></a> Русский </td> <td align="center"> <a href="https://github.com/user-attachments/assets/9ab5b61e-a4c5-4cc8-93ce-a8da446a43c8"><img src="https://github.com/user-attachments/assets/9ab5b61e-a4c5-4cc8-93ce-a8da446a43c8" alt="German" width="300" /></a> Deutsch </td> </tr> </table>

⚙️ How to use

Install:

Chrome Web Store: https://chromewebstore.google.com/detail/captiprep/jgbcfnmpjaflngdajjjnlehfkfohlmfl?authuser=0
Edge Add-ons: https://microsoftedge.microsoft.com/addons/detail/captiprep/fbfhhgelelmeopkjdihhklkcncbikjem
Dev (local): Chrome → chrome://extensions → enable Developer Mode → Load unpacked → select this repo (git clone https://github.com/jeanchristophe13v/CaptiPrep.git first).

Open any YouTube video with (English) subtitles and click the extension icon.
In the floating panel: extract subtitles → select words/phrases → generate cards → study. Arrow keys switch cards; Space favorites a word.
Open the Wordbook from the right‑side entry to revisit learned items anytime.
Export to download your collected words.
Model tips: set filtering model to gemini-2.5-flash-lite and card generation model to gemini-2.5-flash for a good balance.

❓ Why it works

Pre‑exposure: you remove friction before watching.
Real context: examples come from the exact video—retention sticks.
Accumulation: everything funnels into one Wordbook for spaced review.

🧾 Roadmap

[x] Multi‑language UI & i18n.
[ ] Custom prompt templates.
[ ] Richer card data (collocations, patterns).

Note: API keys & model vendor (OpenAI, Claude, Gemini, or OpenAI‑compatible) are configured locally on the Options page and stored only on your machine.

❤️ Acknowledgments

Subtitle extraction groundwork inspired by / thanks to: https://github.com/devhims/youtube-caption-extractor

☕️ Support CaptiPrep

Buy me a coffee if you enjoy CaptiPrep ~ ☕️ 🩵❤️🧡🩷💛💚

<table> <tr> <td align="center"> <img src="https://github.com/user-attachments/assets/cb4fc3c9-ef68-4acb-8c32-232876364a62" alt="WeChat" width="300" /> WeChat </td> <td align="center"> <img src="https://github.com/user-attachments/assets/e9111d96-3fa4-4b79-bcf0-0dfa6a67705f" alt="Alipay" width="300" /> Alipay </td> </tr> </table>

CaptiPrep

Install / Use

README