3 ms·

Voice2Anki: FOSS tool to turn many audio into flashcards

Ey7NFZ3P0nzAe 2y ago
Hi hn, I am nearly at the end of medical school so it is time I publish and "advertise" my open source scripts/apps for anki! I’m excited to share [Voice2Anki](https://github.com/thiswillbeyourgithub/Voice2Anki https://github.com/thiswillbeyourgithub/Voice2Anki), a tool I’ve been developing that combines the power of LLMs (like ChatGPT) and TTS models (like OpenAI’s Whisper) to create high-quality Anki flashcards from voice recordings. It’s designed to be fast, flexible, and language-agnostic, making it a powerful tool for anyone who uses Anki for learning or memorization. This project is part of a larger suite of tools I’ve been building to enhance Anki workflows. You can explore more in [AnkiAIUtils](https://github.com/thiswillbeyourgithub/AnkiAIUtils https://github.com/thiswillbeyourgithub/AnkiAIUtils). ### Why I Built This As a medical student, I needed a way to create large volumes of high-quality flashcards quickly. Traditional methods were time-consuming, so I built Voice2Anki to automate the process. It’s been a game-changer for me, and I believe it has significant potential for anyone who relies on spaced repetition systems like Anki. ### Key Features - Universal Language Support: Works seamlessly in any language. - Adaptive Learning (Few-Shot Learning): Learns from your corrections to improve future cards. - Customization: Supports multiple LLMs (ChatGPT, GPT-4, Replicate, OpenRouter) and TTS models. - Anki Integration: Automatically imports cards with tags, marks, and deck organization. - Smart OCR: Processes images while preserving formatting. ### How It Works 1. Record your voice (e.g., speaking out flashcards). 2. Voice2Anki transcribes and corrects the text using LLMs. 3. It creates and imports the flashcards directly into Anki. ### Getting Started Check out the [GitHub repository](https://github.com/thiswillbeyourgithub/Voice2Anki https://github.com/thiswillbeyourgithub/Voice2Anki) for installation instructions and detailed documentation. ### Looking for Contributors This is a free and open-source project, and I’d love to see it grow. If you’re a developer interested in helping improve the CLI version, integrate it as an Anki addon, or contribute in any other way, your expertise would be hugely appreciated. ### Explore My Other Projects I’ve built several other tools to enhance Anki workflows: - [Whisper Audio Splitter](https://github.com/thiswillbeyourgithub/whisper_audio_splitter https://github.com/thiswillbeyourgithub/whisper_audio_splitt...): Automatically splits long audio recordings into smaller segments. - [OCR_with_format](https://github.com/thiswillbeyourgithub/OCR_with_format https://github.com/thiswillbeyourgithub/OCR_with_format): OCR for Anki that preserves formatting. - And more! Check out my [GitHub profile](https://github.com/thiswillbeyourgithub https://github.com/thiswillbeyourgithub) for the full list. I’d love to hear your feedback, suggestions, or questions. Let me know what you think!

Hi hn, I am nearly at the end of medical school so it is time I publish and "advertise" my open source scripts/apps for anki! I’m excited to share [Voice2Anki](https://github.com/thiswillbeyourgithub/Voice2Anki https://github.com/thiswillbeyourgithub/Voice2Anki), a tool I’ve been developing that combines the power of LLMs (like ChatGPT) and TTS models (like OpenAI’s Whisper) to create high-quality Anki flashcards from voice recordings. It’s designed to be fast, flexible, and language-agnostic, making it a powerful tool for anyone who uses Anki for learning or memorization. This project is part of a larger suite of tools I’ve been building to enhance Anki workflows. You can explore more in [AnkiAIUtils](https://github.com/thiswillbeyourgithub/AnkiAIUtils https://github.com/thiswillbeyourgithub/AnkiAIUtils). ### Why I Built This As a medical student, I needed a way to create large volumes of high-quality flashcards quickly. Traditional methods were time-consuming, so I built Voice2Anki to automate the process. It’s been a game-changer for me, and I believe it has significant potential for anyone who relies on spaced repetition systems like Anki. ### Key Features - Universal Language Support: Works seamlessly in any language. - Adaptive Learning (Few-Shot Learning): Learns from your corrections to improve future cards. - Customization: Supports multiple LLMs (ChatGPT, GPT-4, Replicate, OpenRouter) and TTS models. - Anki Integration: Automatically imports cards with tags, marks, and deck organization. - Smart OCR: Processes images while preserving formatting. ### How It Works 1. Record your voice (e.g., speaking out flashcards). 2. Voice2Anki transcribes and corrects the text using LLMs. 3. It creates and imports the flashcards directly into Anki. ### Getting Started Check out the [GitHub repository](https://github.com/thiswillbeyourgithub/Voice2Anki https://github.com/thiswillbeyourgithub/Voice2Anki) for installation instructions and detailed documentation. ### Looking for Contributors This is a free and open-source project, and I’d love to see it grow. If you’re a developer interested in helping improve the CLI version, integrate it as an Anki addon, or contribute in any other way, your expertise would be hugely appreciated. ### Explore My Other Projects I’ve built several other tools to enhance Anki workflows: - [Whisper Audio Splitter](https://github.com/thiswillbeyourgithub/whisper_audio_splitter https://github.com/thiswillbeyourgithub/whisper_audio_splitt...): Automatically splits long audio recordings into smaller segments. - [OCR_with_format](https://github.com/thiswillbeyourgithub/OCR_with_format https://github.com/thiswillbeyourgithub/OCR_with_format): OCR for Anki that preserves formatting. - And more! Check out my [GitHub profile](https://github.com/thiswillbeyourgithub https://github.com/thiswillbeyourgithub) for the full list. I’d love to hear your feedback, suggestions, or questions. Let me know what you think!