Make a scan searchable
Find a name or an amount in a scanned contract or invoice with Ctrl+F.
OCR PDF
Render scanned pages as images, run OCR, then get both a searchable PDF and the detected text.
- Export
Your files are processed locally in your browser and are never sent to our servers.
Drop your file here
Recommended size: up to 100 MB
Your document stays in the browser. The OCR engine and language data are loaded on demand.
Processing…0 %
Was this tool useful?
How does this tool work?
OCR (optical character recognition) reads the text shown in the images of a scanned or photographed PDF. The tool produces two results: a searchable PDF, in which you can select, copy and search for words, and a plain text file. Pages are analysed in your browser by the open-source Tesseract engine; the document itself is not sent anywhere.
Each page is first drawn as an image at the chosen quality: Fast (scale 1.4, about 100 dpi), Balanced (1.8, about 130 dpi) or High (2.2, about 158 dpi). Tesseract finds the words, their position and their orientation, using the language you indicate: French, English, Spanish, German, Italian, Portuguese, or French and English combined.
The searchable PDF keeps each page as an image and places the recognised words in an invisible text layer exactly on top. On screen the document looks like the original scan; in a reader, search and selection work. The text file gathers the recognised text page by page, preceded by --- Page n --- markers.
On first use, the Tesseract engine and the data for the chosen language are downloaded from a content delivery network (jsDelivr), so an internet connection is needed. Your pages stay on the device. Progress is shown as a percentage during recognition.
How to use this tool
Import the scanned PDF
Choose the document whose text cannot be selected.
Set the language and quality
Indicate the document language and choose Balanced, or High for small print.
Run OCR and download
When it finishes, get the searchable PDF, the plain text, or both, from the buttons provided.
Use cases
Get the text from a document photo
Copy the content of a photographed letter to quote it or rework it.
Prepare a conversion
Add a text layer before converting the PDF to Markdown, HTML or CSV.
Examples
An A4 page at Balanced quality
The page is analysed as a 1,072 × 1,515-pixel image (scale 1.8). In the resulting PDF, searching for the invoice number highlights the right spot on the image, because each recognised word is placed at its original position.
Input
Scanned one-page A4 invoice, language English, Balanced qualityResult
invoice-ocr.pdf (searchable) and invoice-ocr.txt starting with “--- Page 1 ---”Tips and best practices
- A sharp, straight, well-contrasted scan gives far better results; adjust the contrast first if the page is faint.
- Choose the document’s real language: accents and common words are recognised better.
- Check important figures (amounts, IBANs, dates): OCR can confuse 0 and O, 1 and l.
- Remove the blank pages of a long scan first to save time.
Frequently asked questions
Is my PDF sent to a server?
No. Pages are rendered and recognised in your browser. Only the Tesseract engine and the language file are downloaded from jsDelivr on first use.
Does the searchable PDF keep the original text if my document already had some?
No. Each page is replaced by its image and the new text layer comes from OCR. Do not use the tool on a PDF whose text is already selectable.
Why is the resulting file larger?
Because each page becomes a PNG image at the chosen resolution. Fast quality gives a lighter file, at the cost of less reliable recognition of small print.
Which languages are recognised?
French, English, Spanish, German, Italian and Portuguese, plus a French + English combination for mixed documents.
Is recognition 100% accurate?
No. It depends on sharpness, font and layout; tables, handwriting and blurry scans produce more errors.
How long does it take?
A few seconds per page on a recent computer, longer at High quality or on mobile. Progress is shown as a percentage.
Related tools
Keep going with these tools
Tools similar or complementary to this one.
Works well with
Correct contrast, brightness and saturation or convert a PDF to grayscale.
PDF & documentsUse this tool
Automatically detect and remove blank PDF pages directly in your browser.
PDF & documentsUse this tool
Extract selectable text from a PDF and convert it to Markdown locally, without OCR or uploading the document.
PDF & documentsUse this tool
Read next
GuideBest practicesBeginner
How to clean and OCR a scanned PDF
Clean a scanned PDF before OCR to remove unnecessary pages and create a more useful searchable document.
6 September 20262 min
GuideBest practicesBeginner
PDF: understand the format and manage documents properly
Understand how PDF files are structured, the difference between native and scanned PDFs, and the right operations for merging, reorganizing, converting or completing documents.
29 August 20263 min
GuideBest practicesBeginner
Organize a PDF workflow: chain several transformations
Plan PDF operations in the right order and discover the expanded set of reusable PDF Pipeline recipes.
29 August 20262 min
TutorialBest practicesBeginner
How to securely redact a PDF before sharing it
Learn why covering text is not enough, how secure PDF redaction removes sensitive information and why sanitization matters before sharing.
10 September 20262 min