Optimize PDF
OCR a scanned PDF, make it searchable
This tool reads the text in scanned pages and photographed documents, in your browser, and gives you two files: the same PDF with an invisible text layer you can search and copy from, and a plain .txt file. The OCR engine runs on your device, so the document is never uploaded.
or drop files here, or paste with Ctrl/Cmd+V
PDF · up to 50 files · 100 MB and 50 pages each
No file to hand?
Your files stay on your device. Nothing is uploaded. Check it yourself
How to use OCR PDF
Add your scanned PDFs
Drop one or more PDFs of up to 50 pages each on the box. Up to 50 files.
Choose the language
Pick the language the document is written in. Fifteen languages are available.
Read and download
Press Make searchable, then download the searchable PDF, the .txt file, or both.
How the searchable PDF is made
Each page is drawn at 300 DPI, capped at 3000 pixels on the long side, and passed to the OCR engine. It finds the lines and words on the page and recognises them with a neural network trained for the language you chose.
The engine writes the words as invisible text, each placed over the word it was read from. PlumPDF lays that text over your original page without changing anything you can see: the scan keeps its full quality and the file grows only a little. PDF readers then let you search, select and copy as if the text had been typed.
The .txt file holds the same text in reading order, with a "--- Page N ---" line between pages. Columns and tables are not kept as a layout.
Pages that already contain real text are left alone.
Questions about OCR PDF
How do I make a scanned PDF searchable?
Add the PDF above, choose the language of the document and press Make searchable. PlumPDF reads each page and adds the recognised words as invisible text in the right places. The pages look exactly the same, but searching, selecting and copying now work.
Which languages can it read?
English, Hindi, Bengali, Marathi, Tamil, Telugu, Gujarati, Spanish, French, German, Portuguese, Indonesian, Italian, Arabic and simplified Chinese. If a document in another of these languages also contains English words, tick the box to read both at once. Each language is a separate download of 1 to 3 MB the first time you use it.
Is the OCR done on my device?
Yes. The OCR engine runs inside your browser. The first time you use it, your browser downloads the engine (about 4 MB) and the language data from PlumPDF itself, never from a third party, and keeps them for next time. Your PDF is not sent anywhere.
How accurate is it?
On clean, straight scans of printed text, the OCR engine gets nearly every word right. Accuracy drops with blurry phone photos, skewed or curved pages, unusual fonts and very small print. Check names, amounts and dates in the result before relying on them.
Does it read handwriting?
Not reliably. The OCR engine is trained on printed text. Neat block capitals are sometimes read; joined-up handwriting usually comes out as nonsense or not at all.
What happens to pages that already have text?
They are left as they are, and their existing text goes into the .txt file. Adding a second, invisible copy of text that is already there would make every search find it twice.
Is there a page limit, and can I OCR several PDFs at once?
Up to 50 PDFs at a time, of up to 50 pages each, because recognition is heavy work for a browser: expect a few seconds per page on a laptop and longer on a phone. The PDFs are read one after another with the language you choose, and each gives a searchable PDF and a .txt file. A progress bar shows which page is being read, and Cancel stops straight away. Split longer files first with Split PDF.