Transcript RAG

A small toolkit that embeds transcript files into ChromaDB and allows simple retrieval augmented generation queries. It exposes a command line interface via transcriptrag.

Installation

pip install transcript-rag

Usage

Embed transcripts

transcriptrag embed /path/to/transcripts --collection osce --model all-MiniLM-L6-v2

Ask a question

transcriptrag ask --collection osce "How does the student greet the patient?" --gen-model gemini-pro

The command retrieves relevant chunks from ChromaDB and sends them to the specified generation model. Ensure the appropriate API key is available in your environment for the generation model (e.g., Google Generative AI).

Publishing to PyPI

1. Install build tools: pip install build twine
2. Run python -m build to generate distribution files in dist/
3. Upload with twine upload dist/*