OCR to searchable PDF on your computer

Your scans become searchable and selectable. OCR runs entirely in your browser: the document never gets uploaded to a server. Free includes one PDF per day; the 7-Day Pass and Pro are unlimited.

100% private: everything runs on your computer, your files never leave your device.

  1. Upload your scanned PDF - it opens locally on your computer
  2. Text is recognized locally (Spanish and English); pages with text are kept as-is
  3. Download an identical PDF with a searchable text layer

What Searchable OCR does

Searchable OCR takes a scanned PDF, recognises the text on each scanned page and writes that text back into the PDF as an invisible layer placed exactly over the printed words. The page looks the same as before, but now you can search it, select and copy its text, and any indexer or document system can read it. Recognition runs in your browser: the scan is never uploaded.

The tool recognises Spanish and English, both loaded together, which covers invoices, contracts, official letters, receipts and old reports in either language or mixed. It is the step that turns a folder of scans into documents you can actually find things in, and it is the natural preparation before summarising, extracting text or anonymising a scanned file.

How to make a scanned PDF searchable

  1. Add your PDF. The setup screen shows the file name and page count and reminds you what will happen: Spanish and English are recognised, pages that already have text are left untouched, and everything runs on your computer.
  2. Press Recognize text. The first time, the browser downloads the recognition engine and the two language packs, about 8 MB in total, from OléPDF's own domain; afterwards they are cached.
  3. Wait a few seconds per scanned page. Each page is rendered at 350 dots per inch on a computer, or 200 on a phone, its contrast is normalised, sideways pages are detected and straightened, and low-confidence words are discarded.
  4. Download the result, named with the suffix _ocr. Open it in any reader and try the search box.
  5. With the 7-Day Pass or Pro you can add several PDFs and press Run OCR on N PDFs to receive a ZIP with one searchable file per scan.

Limits and behaviour worth knowing

Only pages that have no usable text are processed. A page is treated as scanned when it contains fewer than ten characters of real text; pages with an existing text layer are copied unchanged, so running the tool on a mixed document is safe.

Privacy: recognition happens on your device

The recognition engine is a WebAssembly build shipped by OléPDF, and the language packs are served from the same domain. Page images are rendered in your browser, handed to the engine in memory and discarded when the job finishes. The document is never uploaded, no text is sent anywhere and no third-party script is involved. Even the daily allowance of the free plan is counted locally in your browser.

That makes the tool appropriate for scans that would be a problem to upload: identity documents, medical records, payroll, court filings.

Typical uses

Frequently asked questions

Is OCR free on OléPDF?

The free plan includes one OCR document per day, counted per tool. The 7-Day Pass, a one-time payment, and the Pro plan give unlimited OCR and allow batches of several PDFs.

Is my scan uploaded for recognition?

No. The engine and the language data are downloaded once from OléPDF's own domain and run in your browser. The PDF and its text never leave your computer.

Which languages are recognised?

Spanish and English, always both at once. There is no language selector and no support for other scripts.

Does the PDF look different afterwards?

No. The scanned image is kept as it is. The only change is an invisible text layer aligned with the printed words, which is what makes search and selection work.

What if some pages already have text?

They are left exactly as they were. Only pages with fewer than ten characters of real text are treated as scans and recognised.

How long does it take?

A few seconds per scanned page on a modern computer, more on a phone, where recognition runs at a lower resolution to save memory. The first run adds the one-off download of the engine.

PDF Converter

All tools

Photos rather than PDFs? Try OléPIC, our image tools.