Hi, On Tue Jul 4 22:40:16 2023 lbr <lbr7...@gmail.com> wrote: > I'm trying to create a searchable pdf out of a scanned one. I want to > use Textract as an OCR engine instead of Tesseract. Is there a way to > make libtesseract skip the OCR step and just create the invisible text > layer (with the extracted chars from Textract) and apply it to the > input pdf? > > I read that libtesseract is what ocrmypdf uses to create the invisible > text layer for searchable pdfs.
You can use archive-pdf-tools to do this: https://github.com/internetarchive/archive-pdf-tools it has a Python version of the Tesseract text layer generation and can take hOCR as input (you can convert other OCR formats to hOCR). Note that it is not 100% the same as Tesseract currently - I am trying to find the difference/bug in my port. I am the author, so feel free to reach out if you have any questions. Regards, Merlijn -- Sent from my Motorola Droid 4 running Maemo Leste (Beowulf) -- You received this message because you are subscribed to the Google Groups "tesseract-ocr" group. To unsubscribe from this group and stop receiving emails from it, send an email to tesseract-ocr+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/tesseract-ocr/1689181000.27601.14.camel%40localhost.
