AnkiSnap is an AI-powered desktop application designed to streamline the creation of Anki cards from images. Whether it's a photo of a textbook page, a screenshot of a video, or any other visual material, AnkiSnap uses multimodal Large Language Models (LLMs) to analyze the content and generate high-quality study cards with audio.
- AI Image-to-Card: Transform visual content into structured Anki cards (English/Chinese/Grammar) using models like OpenAI GPT-4o or Aliyun DashScope (Qwen-VL).
- Multimodal Support: Seamlessly processes images and extracts linguistic information, examples, and grammatical explanations.
- Integrated TTS: Enhance your cards with high-quality audio using multiple providers:
- OpenAI TTS
- Azure Cognitive Services
- Volcengine (ByteDance)
- One-Click Sync: Directly add cards to your local Anki collections via AnkiConnect.
- Built with Modern Tech: Powered by Electron, React, Vite, and Tailwind CSS.
- Anki: Must be installed and running.
- AnkiConnect: The Anki add-on (ID: 2055492159) must be installed and configured to allow connections.
- API Keys: You will need API keys for the LLM and TTS providers you wish to use (OpenAI, Aliyun, Azure, or Volcengine).
-
Clone the repository:
git clone https://github.com/changyoung/AnkiSnap.git cd AnkiSnap -
Install dependencies:
npm install
-
Start development mode:
npm run dev
To package the application for production (currently configured for a portable Windows build):
npm run dist
The output will be located in the dist-release directory.
- Open AnkiSnap and navigate to the Settings page.
- Configure your LLM Provider (OpenAI-compatible or DashScope).
- Configure your TTS Provider for audio generation.
- Set your Anki Connection settings (ensure AnkiConnect is active).
Private.
Developed with β€οΈ by AI