Skip to main content
PPDF123

Convert a Scanned PDF to Plain Text

Scanned PDF to Text recognizes the image pages of a PDF in the language you choose and returns one .txt file. Pages that already contain text are copied as they are, so a mixed document is only recognized where it needs to be.

Drag and drop files here, or click to choose

Accepts PDF · up to 500.0 MB per file · up to 20 files at once

You can also paste a file with Ctrl/Cmd+V, or from the clipboard menu.

Document languages
Pick up to 4. Fewer languages read faster and more accurately.

When all you need is the words, a plain text file is the lightest result. This page renders the pages that have no text layer, recognizes them with Tesseract in the languages you pick, and joins the pages into a single .txt file. Pages that already carry text are copied from the PDF, so mixed documents are not recognized twice.

Pick up to four languages that appear in the document. Recognition reads better when the language matches, and accented or non-Latin scripts need their pack selected to come out correctly. If a page order matters for your notes, you can add a heading line before each page, and a page range lets you convert just a section.

OCR reads line by line, so columns, sidebars and tables can interleave and the layout is not kept. Use OCR PDF with the Markdown result if you want headings and lists, or the table tool for tabular data. Up to 50 pages that need recognition are handled per request on this site, and files can be up to 500 MB.

Features

  • One plain .txt file for the whole document
  • Existing text is copied, scanned pages are recognized
  • Up to four languages, including non-Latin scripts
  • Optional heading before each page
  • Page range and a stated 50-page OCR limit

When to use this tool

  • Pulling the text out of a scanned letter, form or report
  • Feeding scanned documents into a search index or a text pipeline
  • Copying a few paragraphs without retyping them

How do I convert a scanned PDF to text?

  1. Upload the scanned PDF
  2. Choose the languages that appear in it
  3. Optionally add page headings or a page range
  4. Click Process and download the .txt file

Limits and edge cases

  • Layout, columns and tables are not preserved
  • Up to 50 pages that need OCR per request on this site
  • Accuracy depends on the scan and on selecting the right languages

Test results and conditions

  • Test files: one-page scans made for this project's test suite (200 dpi, light blur and noise, 0.4 degree skew). The text of a scanned sales table, including row labels such as North and Overseas, came out in a .txt file with the default language (English).
  • Layout is not kept: the table arrives as lines of text.

Privacy for this tool

Files are uploaded over HTTPS, processed in memory or a short-lived temporary directory on our servers, and deleted when your result is ready. We do not keep copies for later browsing, training, or advertising profiles. See the Privacy Policy for retention details and AdSense cookie disclosures.

Frequently asked questions

Will the text keep its layout?
No. You get running text, line by line. Use OCR PDF with the Markdown result for headings and lists, or the table tool for rows and columns.
Does it recognize pages that already have text?
No. In Normal mode their text is copied from the PDF. Force OCR recognizes every selected page again.
Which languages are supported?
The installed packs are listed on the page: 20 languages including English, German, French, Spanish, Chinese, Japanese, Korean and Arabic. You can combine up to four.
Why is the output empty?
The page may be blank or too faint for recognition, or the wrong language is selected. Check the scan and try a higher-quality rescan.

Last updated:

Call this from code

Every tool on this site is a plain REST endpoint - no account or API key needed for anonymous use. Built for AI agents and developers as much as for browsers.

curl -X POST "https://pdf123.xyz/api/v1/misc/ocr-text" \
  -F "[email protected]" \
  -F "languages=eng" \
  -F "ocrType=Normal" \
  -F "pageMarkers=false" \
  -F "pageNumbers=all" \
  -o output.pdf

Also available as an MCP tool for agent clients that speak Model Context Protocol (JSON-RPC 2.0 over POST /mcp). Full API reference

Use it from an AI agent

Skill

Claude Code, Codex, Cursor and other AI agents can run OCR PDF to Text for you with this skill: /skills/pdf123-ocr-text.md

All agent skills

Files are used only for this processing job and deleted automatically afterward.