Skip to content

Documents & scans

Make scanned PDF text searchable

Get extracted search text: page text, headings, paragraphs, tables as text, labels, captions, and visible notes from scanned pages.

  • Scans detected automatically · AI Deep Read only when needed
  • Free: 5 conversions a day · 5 MB each
  • Deleted after 24 h · never used for training

Before you start

What to expect

  • Files are retained only for the product retention window and deleted after 24 hours unless your account policy says otherwise.
  • Uploads are not used to train models; AI Deep Read reads the page only to produce your requested output.
  • AI Deep Read costs 1 credit per rendered page and still needs human review for names, dates, legal text, and totals.
  • Searchable-PDF output with the original look is not offered yet; this workflow creates extracted Markdown/text instead.

What the output contains

  • Markdown or text extracted from scanned PDFs, TIFFs, PNGs, and JPGs using AI Deep Read where OCR is needed.
  • Readable headings, paragraphs, labels, and simple tables suitable for search, copy/paste, or conversion to Word.
  • The original look of the PDF is not preserved in the output because this workflow does not create a hidden text layer.
  • For a true searchable PDF that keeps the original page image, use ocrmypdf locally today.

A real run

Scanned policy binder

Input
32-page scanned PDF binder with section headings, stamps, two simple tables, and no embedded text layer.
Output
Markdown with page-ordered headings, paragraphs, and two GFM tables; it was searchable as text, but not a PDF preserving the original scanned appearance.

Known failure modes and how they're handled

User expects searchable PDF output
Be explicit: FileType Converters does not yet output a PDF with an invisible text layer. It outputs Markdown/text/Word from the scan.
Huge scanned books or archives
Each rendered page consumes credits and time. Split very large scans and review output section by section.
Language, tables, and reading order
Deep Read extracts visible text and table-like Markdown, but complex multi-column reading order must be checked.
Password-protected PDFs
Unlock the PDF first. The workflow cannot OCR pages it cannot render.

Command line

Do it yourself

ocrmypdf

bash
ocrmypdf --deskew --rotate-pages scan.pdf searchable.pdf

This is the right DIY route when you need a PDF that looks the same but has an invisible searchable text layer.

pdftotext -layout

bash
pdftotext -layout searchable.pdf searchable.txt

Extracts the OCR text layer after ocrmypdf creates it.

Questions about this workflow

Will I receive a searchable PDF?

No, not yet. You receive Markdown/text/Word-style extracted content. Use ocrmypdf locally if you need the original PDF plus invisible OCR text.

Can I convert TIFF scans?

Yes. TIFF, PNG, JPG, and PDF scans can be read into Markdown text through this workflow.

Does a digital PDF consume Deep Read credits?

With auto detection, a PDF that already has usable text can avoid Deep Read. Scanned pages consume 1 credit per rendered page.

Can I turn the Markdown into Word?

Yes. After extraction, convert Markdown to DOCX if an editable document is the goal.

← All workflows