language-transfer-flashcards

Language Transfer Flashcards (LTF) is a CLI tool designed to enhance your language learning experience by bridging the gap between Language Transfer's audio lessons and Anki's spaced repetition system. This project transforms the rich content of Language Transfer YouTube lessons into easily digestible flashcards, allowing learners to reinforce their knowledge effectively.

About the Project • Installation • Usage • Limitations • Anki: CSV Import • Known Issues • Roadmap • Contact

About the Project

You can also read more about this project in my Medium blog post.

Key Features

• 🤖 Automated Extraction: Extracts words, phrases, and sentences with translations from Language Transfer YouTube lessons.
• 📚 Anki Integration: Generates CSV files ready for direct import into Anki, creating instant flashcard decks ( OpenAI API key required).
• 🔄 Alternative Workflow: Download a text file, containing the prompt and lesson transcript to use with your favorite LLM web interface (e.g., ChatGPT, Claude, etc.).

Why it's Valuable?

• 🚀 Boost Learning Efficiency: Combine Language Transfer's proven method with Anki's powerful spaced repetition.
• ⏱️ Save Time: Automate hours of manual flashcard creation.
• 🛠️ Customizable: Adapt to your preferred workflow, with or without API access.

Build With

Python

LangChain

Typer

Installation

Recommended: Create and activate a virtual environment before installing the tool.

Using pip

pip install language-transfer-flashcards

Using poetry

poetry add language-transfer-flashcards

To verify that everything works and to see all available commands

ltf --help

Configuration with .env file [optional, but recommended]

Create a .env file in your home directory under ~/.ltf/.env with the following variables:

OPENAI_API_KEY=<YOUR_OPENAI_API_KEY>
OPENAI_MODEL_NAME=gpt-4.1-mini
TARGET_LANGUAGE=Swahili

Not sure where to create the .env file in your system?

ltf env-location

To see all available OpenAI model names, check the official docs.

To see all valid values for TARGET_LANGUAGE

ltf csv --help

Note: The OpenAI-related variables are optional if you're not using the API.

Usage

Create flashcards in CSV format - file will be saved in current working directory

ltf csv "https://www.youtube.com/watch?v=VIDEO_ID"  # assumes .env file exists

Without the .env file, specify your target-language, the OpenAI model name and API key

ltf csv "https://www.youtube.com/watch?v=VIDEO_ID" -l Swahili -m "gpt-4.1-mini" -k "YOUR_OPENAI_API_KEY"

Important: Find the YouTube URLs for all Language Transfer lessons here.

Usage without using the OpenAI API

Download the full prompt which is used to extract the content of the language lesson in a txt file.

ltf prompt "https://www.youtube.com/watch?v=VIDEO_ID"  # assumes .env file exists

If there is no .env file, add your target-language to the command, for example: -l Swahili

Manual Processing

• Copy the downloaded content into your favorite LLM web interface - recommended: ChatGPT or Claude.
• The output will be the content of the lesson in CSV format -> English, Translation.
• Copy the output into a text editor and create a CSV file
• Import the CSV file into Anki

Limitations

• AI Imperfections: LLMs may occasionally produce inaccurate translations or misinterpret context.
• Transcript Quality: YouTube's auto-generated transcripts can contain errors.
• Content Discrepancies: Flashcards may sometimes differ from the exact lesson content due to transcript and AI limitations. Review your final flashcards before importing them into Anki.
• Technical Requirements: Basic command-line knowledge is needed.
• Estimated API Costs per Lesson:
  • gpt-4o: ~$0.01
  • gpt-4o-mini: < $0.0005

Anki: CSV Import

• Open Anki
• Create or select a deck for your target language
• Navigate to File > Import
• Select your generated CSV file for upload
• Select the correct seperator for your CSV file, configure Import options and click on Import

Known Issues

• On Windows PowerShell, URLs with & characters—often from playlist links like

  ltf csv "https://www.youtube.com/watch?v=VIDEO_ID&list=PLAYLIST_ID"
  

  can cause PowerShell to misinterpret parts of the URL as separate commands showing terminal messages like:

  'list' is not recognized as an internal or external command,
  operable program or batch file.
  

  The command still runs successfully. Since the playlist part is usually not needed, you can shorten the URL to just the video or escape & with a backtick in PowerShell

Roadmap

• Publish on PyPI
• Add support for other LLMs (including Ollama)
• Create a web-based interface for non-technical users

Contact

• 🌐 Website: https://charura.com
• 📧 Email: [e-mail]
• 👨‍💻 GitHub: @t-charura

If you find Language Transfer Flashcards helpful, please consider giving it a ⭐️ on GitHub!