Skip to content

Home

Academic Doc Generator Logo

Workflow Infographic

Welcome to the Academic Document Generator documentation!

Transform annotated PDFs into professional LaTeX documents using AI. Generate thesis colloquium protocols, project grading letters, peer review comments, and translate LaTeX exams automatically.


πŸ“Š Four Main Use Cases

  • πŸŽ“ Colloquium Protocols


    • Notes β†’ clear questions
    • Auto-metadata extraction
    • Thesis summary
    • Form pre-filling
    • Email generation

    Documentation

  • πŸ“Š Project Work Grading


    • Metadata extraction
    • Salutation determination (Mr/Ms)
    • Grading letter template
    • Feedback summary
    • Student email

    Documentation

  • ✍️ Peer Review Comments


    • Notes β†’ constructive feedback
    • Auto line number detection
    • Markdown export
    • Always in English
    • Scientific tone

    Documentation

  • πŸ”€ LaTeX Exam Translator


    • German β†’ English
    • Exam class optimized
    • Preserves math formulas
    • Protected comments
    • Structure-aware

    Documentation


✨ Key Features

  • πŸš€ Unified CLI - Single academic-doc-generator command for all tasks
  • πŸ” PDF Annotation Extraction - Extract text and annotation positions with Docling + PyPDF
  • πŸ€– Multiple LLM Support - Works with OpenAI, Groq, Google Gemini, or Ollama
  • 🎯 Context-Aware Rewriting - Maps annotations to exact highlighted text and paragraphs
  • ✍️ Intelligent Comment Refinement - Rewrites terse notes into full questions
  • πŸ“ LaTeX Generation - Creates professional letters with TH KΓΆln formatting
  • βœ’οΈ Automatic Signature Detection - Automatically includes signature from data/signature.png
  • πŸ“‹ PDF Form Pre-filling - Auto-fills official grading forms (auto-mapped to course of study)
  • πŸ“§ Email & Outlook Integration - Creates registration emails and Outlook drafts
  • 🌐 Unicode Support - Handles German special characters correctly

πŸ› οΈ Requirements

  • Python: 3.9 or higher
  • LaTeX: LuaLaTeX recommended (for Unicode support)
  • LLM API: At least one of OpenAI, Groq, Gemini, or Ollama

🀝 Contributing

Contributions are welcome! Please see CONTRIBUTING.md for guidelines.

πŸ“„ License

This project is released under the MIT License.


Note: This tool aids in document template creation β€” it does not mark or make evaluative decisions automatically. All academic assessments remain the responsibility of the examiner.