OCR
Recognize text in PDFs and images and write a text layer. It uses Tesseract and stays on the PC. A hot folder starts a finished file immediately and labels an incomplete one. Pages that contain only spaces are not skipped as if they were real text. The language-data failure on Korean, Greek, Serbian, Azerbaijani, Uzbek, and some other languages is fixed in 11.32.0. If OCR still reports Could not open data file, update first, then see the OCR guide.