OCR PDF Converter
Turn a scanned PDF into editable text with OCR that runs entirely in your browser.
Pages are drawn and read inside this page. Your PDF is never uploaded.
What this tool does
A scanned PDF is a set of pictures, so ordinary PDF text extraction returns nothing. Choose the file, pick the language and pages, and the tool draws each page and reads it with OCR in your browser. Pages that already contain real text are used as they are. Copy the result or save a .txt file.
Draws each chosen page at the resolution you pick and runs optical character recognition on it, all inside your browser.
Skips OCR on pages that already contain real text and shows progress and the editable result as it goes.
How to use it
- Choose the scanned PDF.
- Pick the language of the text and, if you only need some pages, a page range such as 1-3,7.
- Choose the reading resolution: 150, 200 or 300 dpi.
- Press Read the PDF and wait; the status line shows the page being read.
- Edit the result if needed, then press Copy or Download .txt.
Why use this converter
Reads text out of scanned, image-only PDFs with an OCR engine that runs in your browser, so the document is never uploaded. Scanned contracts, receipts, forms and old documents are often PDFs that contain only pictures of pages, so you cannot search, copy or reuse their text. OCR turns the pictures back into text, and doing it in your browser means a private document does not have to be uploaded to anyone.
Why we built it
We built the OCR PDF converter to run text recognition in your browser and to use real text on pages that already have it. Read single pictures with the image to text converter and extract embedded text with PDF to text.
Who this tool is for
- People holding scanned contracts, receipts and forms
- Researchers digitising archives
- Anyone who cannot copy text from a PDF
When to use it
- You need to copy text from a scanned contract, letter or form.
- A PDF cannot be searched and you want a text version you can search.
- A document is sensitive and you do not want to send it to an online service.
- You want text from only some pages of a long scan.
How it works
Each page is drawn to a canvas at the resolution you choose. If the page already has a text layer of at least 25 characters and the "use the file's own text" box is on, that text is used directly. Otherwise the canvas is passed to the Tesseract OCR engine, which finds lines and words and returns the text with a confidence score.
Higher resolution gives the engine more detail to work with and is slower; 200 dpi is a balanced default and 300 dpi helps small print. The language setting selects the trained data the engine uses, so reading Spanish with the English setting gives poorer results.
OCR is never perfect. Always proof-read figures, names and dates against the original page before relying on them.
What you give it, and what you get back
| Item | Details |
|---|---|
| PDF file | A scanned or image-only PDF. Text PDFs also work, using their own text. |
| Language | English, Spanish, French, German, Portuguese, Italian or Hindi. |
| Pages | All pages, or a range such as 1-3,7. |
| Resolution | 150, 200 or 300 dpi for the page images. |
| Output | Editable text, one block per page, to copy or download as .txt. |
Issues you may run into
| What you see | Why it happens | What to do |
|---|---|---|
| Output is empty or gibberish | The scan is faint, skewed or very low resolution, or the language is wrong. | Pick the right language, try 300 dpi, or rescan at higher quality. |
| Reading is slow | OCR is heavy work done on your own device. | Read fewer pages at a time, or use 150 dpi for a quick look. |
| Browser tab runs out of memory | Very large pages at 300 dpi need a lot of memory. | Lower the resolution or split the PDF first. |
| Tables and columns are mixed together | OCR returns reading-order text, not layout. | Copy the text and rebuild the table by hand. |
| Handwriting is not read | The engine is built for printed text. | Type handwritten notes manually. |
Assumptions and limitations
- Seven languages are included; others are not available.
- Layout, images and formatting are not reproduced - only the text.
- Accuracy depends on scan quality; always proof-read.
- Large or high-resolution documents are limited by your device's memory.
Frequently asked questions
How do I extract text from a scanned PDF?
A scanned PDF contains pictures, so use OCR. Choose the file here, pick the language and read it; the text appears in the box for you to copy or save.
Is my PDF uploaded anywhere?
No. The pages are drawn and read inside your browser, and the file never leaves your device.
Why does normal PDF to text return nothing?
Scanned PDFs have no text layer, only images of pages, so plain extraction finds nothing. OCR reads the images instead.
What resolution should I choose?
200 dpi suits most documents. Use 300 dpi for small print and 150 dpi for a fast first look.
Which languages can it read?
English, Spanish, French, German, Portuguese, Italian and Hindi.
Can it keep the layout or make a searchable PDF?
No. It returns the text only; it does not rebuild the layout or write a new PDF.