PDF OCR (scanned text)
Recognize text from a scanned PDF, in your browser.
PDF OCR (scanned text) runs optical character recognition on every page of a scanned PDF entirely on-device, in English or Chinese, and lets you download the recognized text.
Click to choose a PDF, or drag one in here
One file at a time
How to use
- Upload a scanned PDF and choose its language.
- Click "Run OCR" — each page is recognized in your browser, which can take a while for long documents.
- Review the extracted text, then download it as a .txt file.
Common use scenarios
- Making a scanned document's text searchable and copyable for the first time.
- Extracting text from a scanned form, receipt, or old paper document.
- Digitizing a printed document into editable text without typing it manually.
- Getting a rough text extraction from a scan before cleaning it up by hand.
✓Content & verification reviewed: 2026-09-17
How we verify PDF OCR (scanned text)
How to set expectations for OCR accuracy and processing time.
Accuracy depends heavily on scan quality
Clean, high-contrast, straight scans of printed text recognize well; handwriting, unusual fonts, low resolution, and skewed pages all reduce accuracy — review the output rather than trusting it blindly.
Choose the language before running
Pick English, Traditional Chinese, or Simplified Chinese to match the document's actual language — using the wrong language significantly hurts accuracy.
Processing happens entirely on your device
Since there's no server doing the work, large or many-page documents can take real time and use noticeable CPU — this is expected, not a malfunction.
Before you rely on the result
- You've picked the language that matches the document's actual text.
- You understand the first run downloads a small language model, which only happens once per language.
- You've reviewed the recognized text for errors before relying on it, especially for important documents.
Who this is for
- Anyone who needs to search or copy text from a scanned document.
- People digitizing old paper documents, forms, or receipts.
- Anyone extracting a rough text draft from a scan to clean up by hand.
Real-world examples
Making an old scanned contract searchable
A scanned paper contract has no text layer at all — running OCR extracts its text so specific clauses can finally be searched and copied.
Digitizing a handwritten-free printed form
Someone scans a printed (not handwritten) form and runs OCR to get its field labels and printed values as editable text.
Limitations
- Accuracy depends heavily on scan quality and can struggle with handwriting, unusual fonts, or skewed pages; entirely on-device, so large PDFs take real time and CPU.
Frequently asked questions
Does this work on handwritten text?
It can attempt to, but accuracy on handwriting is much lower than on printed text — expect to correct errors manually.
Why does the first run take a while?
It downloads a language-specific recognition model the first time you use a given language; subsequent runs in the same language are faster.
Is my document uploaded to a server for recognition?
No — recognition runs entirely in your browser, on your device.
🔒 Runs in your browser — nothing is uploaded. Results are for reference only; see the disclaimer. Last updated: 2026-09-17
Subscribe for updates
New tools and tips, straight to your inbox. No spam, unsubscribe anytime.