Open Source Ecosystems

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

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

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-4o
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-4o -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