OCR PDF (Scanned PDF to Text)
Read the text off a scanned PDF. Free, and runs entirely in your browser — nothing you enter is uploaded, stored or sent to a server.
About OCR PDF
Reads the text off a scanned PDF. A scan holds pictures of words rather than words, which is why searching, selecting and copying all fail on it; optical character recognition looks at the picture and works out what the letters were.
The engine is Tesseract, compiled to WebAssembly and served by this application, so the recognition runs in this tab and the pages never leave your browser. The one thing fetched from elsewhere is the trained model for the language you pick — a request carrying a language code and nothing else — because there are over a hundred of them at several megabytes each.
Two things come back: the text itself, and optionally a searchable PDF, which is the original scan with the recognised words laid invisibly on top at the co-ordinates they were read from. That file looks exactly like the scan and can be searched, selected and read aloud.
Confidence is reported per page and overall, and it is worth reading. Recognition is a guess made character by character, and a faint, skewed or low-resolution scan produces a confident-looking result that is quietly wrong in places.
Where it helps
- Recognition runs in this tab; the document is never uploaded.
- Fourteen languages, including Hindi, Tamil, Arabic, Chinese, Japanese and Korean.
- Produces a searchable PDF as well as plain text, keeping the scan's appearance.
- Confidence is reported per page, so a bad scan announces itself instead of being trusted.
Limits
- The trained model for the language you choose is fetched from a public distribution, which is the one external request these tools make.
- Recognition takes several seconds a page, so a run is limited to 40 pages.
- Handwriting, unusual fonts, faint print and skewed scans all produce errors — sometimes confident ones.
- The searchable PDF holds the page as a picture, so it is larger than a text PDF of the same document.
Frequently asked questions
Is my document uploaded?
No. The engine and its WebAssembly core are served by this application and run in this tab. Only the language model is fetched, and that request carries a language code — nothing from your file.
Why is the recognition wrong in places?
OCR is a guess from pixels. Faint print, low resolution, skew and unusual typefaces all cost accuracy. Try 300 dpi, and read the confidence figure before trusting the result.
What is a searchable PDF?
The scan with the recognised words drawn invisibly on top, at the positions they were found. It looks exactly like the original, but searching, selecting and screen readers all work.
Can it read handwriting?
Very poorly. Tesseract is trained on printed text; handwriting needs a different kind of model.
Why only 40 pages?
Each page takes seconds of solid computation in your browser. A longer document is better split first and run in parts.
Related tools
- PDF to Word — Turn a PDF into an editable .docx that keeps its layout
- Compress PDF — Make a PDF smaller, by rebuilding it or redrawing it
- PDF to Images — Turn every page into a PNG, JPG, WebP, GIF or JFIF
- PDF Viewer — Read, search and inspect a PDF without leaving the browser
- Icon Font Generator — Build a web font from SVG icons or an SVG font
- Image Converter — Convert between PNG, JPEG, JFIF, WebP, GIF, HEIC and SVG
More Images & Files tools
- Icon Font Generator — Build a web font from SVG icons or an SVG font
- Image Converter — Convert between PNG, JPEG, JFIF, WebP, GIF, HEIC and SVG
- Image Resizer — Resize to a size, a proportion or a longest side
- Crop Image — Cut a rectangle out, or trim a flat border away
- Rotate Image — Turn by a right angle or any angle, and flip
- Image Base64 Converter — Turn an image into a data URI for CSS or HTML, and read one back
- WebP Converter — To WebP for the web, or back out of it
- SVG Converter — Draw an SVG at any size, or wrap a bitmap in one
- Background Remover — Take a flat background out and keep the subject
- Image Compressor — Resize and re-encode images to save bytes
- Image Optimizer — Strip metadata and redundant markup without touching the picture
- Favicon Generator — Build favicon PNGs and a multi-size .ico
- File Checksum — Verify a download, and work out which algorithm made the checksum
- QR & Barcode Reader — Read a code out of an image, and say what it does before you follow it
- MIME Detector — Identify a file from its magic bytes
- Office Password — Protect a Word, Excel or PowerPoint file with a password
- Office Password Remover — Remove the password from a Word, Excel or PowerPoint file
- Image Editor — Crop, straighten, adjust and cut out, with every step visible and reversible
- Passport Photo Maker — Passport and visa photos at each country's size, with head guides and a print sheet
- PDF Editor — Move and retype what is already on the page, reorder, annotate, stamp and sign, with the document in front of you
- PDF Viewer — Read, search and inspect a PDF without leaving the browser
- PDF Password & Permissions — Lock a PDF with AES-256, and say what a reader may print, copy or change
- PDF Password Remover — Strip the password from a PDF you can already open
- Merge PDF — Join several PDFs into one document
- Split PDF — Cut a PDF into separate documents
- Delete PDF Pages — Remove pages from a PDF and keep the rest
- Rearrange PDF Pages — Put a PDF's pages in a different order
- Rotate PDF — Turn a PDF's pages the right way up
- Compress PDF — Make a PDF smaller, by rebuilding it or redrawing it
- Watermark PDF — Stamp text or a logo across a PDF's pages
- Sign PDF — Put a signature and a date onto a PDF
- Edit PDF — Add text to a PDF page, and hide what is underneath
- PDF to Images — Turn every page into a PNG, JPG, WebP, GIF or JFIF
- PDF to Word — Turn a PDF into an editable .docx that keeps its layout
- PDF to Excel — Pull a PDF's tables into a .xlsx workbook that keeps their formatting
- PDF to PowerPoint — Turn a PDF's pages into editable .pptx slides that look the same
- Word to PDF — Turn a .docx document into a PDF that looks the way it does in Word
- Excel to PDF — Print a .xlsx workbook to PDF with its formatting and page setup
- PowerPoint to PDF — Turn a .pptx deck into a PDF that looks the way the slides do
- Images to PDF — Turn photos or scans into one PDF document
- PDF to Text — Pull the text out of a PDF as plain text
- Crop PDF — Trim the margins off every page
- Page Numbers — Number the pages of a PDF
- Header & Footer — Put a running head or foot on every page
- Bates Numbering — Stamp sequential legal numbers across pages
- Reverse Pages — Flip a PDF's page order end to end
- Add Blank Pages — Insert empty pages anywhere in a PDF
- N-Up PDF — Put several pages on one sheet
- PDF Booklet — Reorder pages for folded booklet printing
- PDF Metadata — Read, change or strip a PDF's properties
- Flatten PDF — Make form answers part of the page
- Alternate & Mix Pages — Interleave two PDFs a page at a time
- PDF Overlay — Draw one PDF over or under another
- Combine to Single Page — Stack every page into one continuous strip
- Divide Pages — Cut every page in half or into quarters
- Posterize PDF — Tile a large page across several printable sheets
- Fix Page Size — Put every page onto one uniform sheet
- Page Dimensions — Measure every page, its rotation and its sheet
- Rotate by Any Angle — Turn pages by a fraction of a degree or more
- Remove Annotations — Strip comments, highlights, links and stamps
- PDF Attachments — List, extract, add or remove embedded files
- PDFs to ZIP — Package several PDFs into one archive
- PDF Bookmarks — Read, write or remove a PDF's outline
- Table of Contents — Build a contents page from the bookmarks
- Page Labels — Number front matter in roman and the body in arabic
- Sanitize PDF — Strip scripts, actions, attachments and metadata
- Extract Images — Pull every embedded picture out of a PDF
- Extract Tables — Turn a PDF's tables into CSV, JSON or Markdown
- PDF to Markdown — Convert a PDF into Markdown with its headings, emphasis, lists, tables and pictures
- PDF to JSON — Export a PDF's text as structured JSON
- Compare PDFs — See which pages of two PDFs differ, and by how much
- Remove Blank Pages — Find the empty pages, and take them out
- Invert PDF Colours — A dark-mode version of a document
- PDF to Greyscale — Convert every page to shades of grey
- Adjust PDF Colours — Brightness, contrast and saturation across a document
- Scanner Effect — Make a clean PDF look like a photocopy
- PDF Background Colour — Tint the paper behind the content
- Rasterize PDF — Flatten every page into a picture of itself
- PDF to CBZ — Turn a PDF into a comic book archive
- CBZ to PDF — Turn a comic book archive into a PDF
- Text to PDF — Set a plain text file as a document
- Markdown to PDF — Set Markdown as it renders: headings, emphasis, lists, code, quotes and tables
- JSON to PDF — Print a JSON file as an indented tree
- CSV to PDF — Set a spreadsheet export as a printed table
- XML to PDF — Print an XML document as an indented tree