OCR PDF – Make Scans Searchable
Recognize text in scanned PDFs so you can search, select and copy it. Runs entirely in your browser.
- ✓Files never uploaded
- ✓No file limits
- ✓No sign-up
- ✓No watermark
How to OCR PDF
- 1Drop your scanned PDF into the box or click to choose it.
- 2Choose the document language.
- 3Click Run OCR and wait while each page is recognized.
- 4Download your searchable PDF.
Why pdfdots?
Private by design
All processing happens locally in your browser. Your documents are never uploaded to a server, so nobody else can see them.
No limits
No daily task caps, no file-count limits and no premium tier. Process as many PDFs as your device can handle.
Fast
There's no upload or download queue. Work starts the moment you drop a file and finishes at the speed of your own computer.
Works everywhere
Windows, Mac, Linux, iPhone, Android or Chromebook — if it has a modern browser, pdfdots runs on it. Nothing to install.
Turn scanned pages into searchable text
A scanned PDF is really just a stack of pictures: you can read it, but you can't search it, select a sentence or copy a number out of it. OCR (optical character recognition) finds the text in those images and adds it to the PDF as an invisible layer placed under the words you see. The pages look the same, but Ctrl+F, copy and paste, and screen readers now work.
pdfdots runs OCR with Tesseract, an established open-source engine, directly in your browser. Your document is never uploaded. The only download is the language model for the language you pick (English, Turkish, Spanish, Portuguese, German or French), roughly 10–20 MB, and your browser keeps it so later runs start faster. Choosing the right language matters, because it helps the engine recognize accented letters and common words.
Results depend on the scan. Clean, straight, well-lit pages at a decent resolution give very good results; blurry photos, skewed pages, tiny print or unusual fonts lead to more mistakes, and handwriting is not reliably recognized. Each page is re-rendered at about 200 DPI before recognition, so the finished file can be larger than the original — run it through Compress PDF afterwards if needed.
Recognition takes a few seconds per page on a typical computer, so long documents need some patience, and phones are slower. If you only need part of a long scan, extract those pages first. Once your PDF has a text layer, you can also convert it to Excel or Word much more reliably.
Our take
We run OCR when a scanned document has to become searchable: an archive of old letters, a scanned contract where we need to find a clause, or a paper statement whose figures we want to copy.
We always set the document language correctly, because it noticeably improves recognition of accented letters. The language model downloads once and the file stays on the device. For long scans we extract only the pages we need first, since recognition takes a few seconds per page.
The pitfall: OCR on a crooked or blurry phone photo produces odd characters, and handwriting is mostly lost. Rescanning straight is often faster than correcting. When you need tables in a spreadsheet, follow OCR with PDF to Excel; for editable text, PDF to Word.
Related tools
Frequently asked questions
Is my PDF uploaded?+
No. OCR runs entirely in your browser. Only the language model is downloaded once; your file stays on your device.
Which languages are supported?+
English, Turkish, Spanish, Portuguese, German and French. Pick the language of your document before starting.
Will the pages look different?+
They look the same. The recognized text is added as an invisible layer under the page image.
Why is the output file bigger?+
Pages are re-rendered at about 200 DPI for recognition. Use Compress PDF if you need a smaller file.
Can it read handwriting?+
Not reliably. OCR works best on printed text from clear, straight scans.
Why is it slow?+
Recognition takes a few seconds per page on your own device, so long documents take a while.