Recognise the text in a scanned PDF and get back a file you can search and copy from.
PDF OCR
Recognise the text in a scanned PDF and get back a file you can search and copy from.
Recognition runs on your device, so scanned records and archives stay with you. The first run downloads the recogniser and its language data from a public CDN; no page content is ever sent.
A scanned PDF is a picture of a document. You can read it, but nothing else can: search finds nothing, copy gives you nothing, and every tool that works on text has nothing to work with. Text recognition fixes that by reading the shapes of the letters and writing what it finds back into the file as an invisible layer sitting over the printed words. Each page is redrawn as an image at 144 or 216 dpi and the new file is built from those images, so it looks like the original, but a higher-resolution scan comes back at the setting you chose, and links or form fields do not carry over. In exchange, the document is now searchable, selectable and convertible. Recognition runs in your browser, in any of twelve languages, and the extracted text is available on its own as well.
Features
Recognition runs in your browser — pages are never uploaded
Twelve languages, including Hindi, Russian, Chinese, Japanese and Arabic
Invisible text layer sits over each page image, redrawn at 144 or 216 dpi
Two scan qualities, trading speed against accuracy on small print
Plain text output alongside the searchable PDF
Per-page progress, with a stop control for long documents
Language data downloads once and is cached by your browser
How to use the PDF OCR
Drop the scanned PDF onto the page
Choose the language printed on the page and a scan quality
Start recognition and watch it work through the pages
Download the searchable PDF, or just the recognised text
Example
Input
1987-minutes.pdf — 8 scanned pages, typewritten English
Output
1987-minutes-searchable.pdf — pages redrawn at 144 dpi, now with selectable text behind them.
The pages look the same, redrawn at the chosen resolution, and the words are now in the file.
Common errors & troubleshooting
The recognised text is full of mistakes. — Recognition depends on the scan. Try the higher quality setting, and check the language matches the page — English data on a German page produces exactly this.
Recognition is very slow. — It is genuinely heavy work, and it is happening on your own processor rather than a server. Use the faster quality, and split long documents.
A document with two languages recognises badly. — One language is used per run. Split the document by language, recognise each part, and merge the results.
Handwriting was not recognised. — This recognises printed and typewritten text. Handwriting needs a different class of model entirely.
Frequently asked questions
Does the recognised text replace the scan?
No. Each page is kept as a picture, redrawn at 144 or 216 dpi, with the recognised text added invisibly behind it, so the searchable PDF looks like the original. Because the file is rebuilt from those pictures, a scan made at a higher resolution comes back at the one you chose, and links and form fields are not carried over.
Are my scanned pages uploaded for recognition?
No. Recognition runs in your browser and your pages never leave it. On first use the browser downloads the recogniser's program files and the language data from a public CDN; nothing from your document is sent, and the language data is cached for next time.
How many pages can I recognise at a time?
Up to thirty in one run, because recognition is slow enough that longer documents are better handled in batches.
Which quality setting should I choose?
Start with the faster one. Move to the higher setting when the print is small, the scan is old, or the first attempt came back messy.
Can I OCR a PDF that already has text?
You can, but there is no point — check first by trying to select a word. If text is already there, convert it directly instead.
Related tools
PDF to Text — Extract selectable text from a PDF as plain text or Markdown.
Scan to PDF — Use a camera as a document scanner and save the pages as a PDF.
PDF to Word — Rebuild a PDF as an editable .docx document.
PDF to Excel — Pull the tables out of a PDF into an .xlsx workbook or CSV.
PDF to HTML — Convert a PDF into clean, searchable, responsive HTML.
Compress PDF — Shrink a PDF by re-rendering each page to a JPEG at a chosen quality and resolution.
PDF Summarizer — Summarise a long PDF section by section, on your device.