Sharing a step that dramatically improved my card-making, in case it helps other heavy-lecture-load folks (med/grad especially).
The problem: making good cards from a 120-slide lecture PDF is miserable when the source is a wall of low-quality text, weird line breaks, and image-only slides. Garbage in, garbage cards.
What fixed it: I convert every lecture to clean Markdown before I open Anki. One .md per lecture, real filename (not “L14_final.pdf”), images OCR’d so the text is actually selectable. Then card-making is just reading clean text and pulling out the testable bits — way faster, and cloze deletions stop breaking on mangled formatting.
I do this with pandoc/MarkItDown for docs and an OCR pass for image slides. (Full disclosure: doing it by hand got old so I built a small Mac app that automates the convert+rename+OCR step — it’s on GitHub as NoteCapt, free beta — but the manual toolchain works fine.)
Curious how others here prep lecture material before cards — do you convert first, or make cards straight from the PDF?