This Streamlit app allows users to generate flashcards from a given document using the power of GPT and Langchain. By analyzing the content of the document, the app identifies key concepts, definitions, and other important information to suggest potential flashcards.
- Document Input: Users can paste their document text into a textbox
- Analysis: After pasting the text, users can click a button to send the input to GPT for analysis.
- Summary & Suggestions: The app will then display a summary of the text with all suggestions for flashcards highlighted.
Flashcards are generated based on the following criteria:
- Definitions: Any sentence that defines a term or concept is a candidate for a flashcard. For example, "Photosynthesis is the process by which green plants and some other organisms use sunlight to synthesize foods."
- Main Concepts: Key concepts or topics that are central to the document's theme or subject.
- Important Dates: Any historical or significant dates mentioned in the document.
- Formulas: In case of scientific or mathematical documents, any formula is a candidate.
- Contrasts or Comparisons: Sentences that contrast or compare two or more things can be turned into a flashcard. For example, "Unlike mammals, birds lay eggs."
- Lists: Any enumerated list or series of related items can be a candidate, especially if they detail steps, types, or categories.
- Paste your document text into the provided textbox.
- Click the "Analyze" button.
- Review the summary and highlighted suggestions for flashcards.
- Use the suggestions to create flashcards in Obsidian or any other flashcard tool.
- Generating .md file for Obsidian flash cards
- pdf, .md, docx and powerpoint file support
- Option to customize the rules for flashcard generation based on user preference.