Skip to content
Documentation
Cookbook
Multimodal: Vision, Charts, and Transcription
Transcribe documents

Transcribe documents

Document transcription patterns across typed text, handwriting, and forms, including “unstructured → JSON”.

  • Upstream notebook: multimodal/how_to_transcribe_text.ipynb

What to focus on

  • Choosing extraction targets (fields) before transcribing everything
  • JSON as the default contract for form-like documents
  • Dealing with low-quality scans (crop/segment first)

Run locally 

make test-notebooks NOTEBOOK=multimodal/how_to_transcribe_text.ipynb
Charts, graphs, and slide decksCrop tool for image analysis