Leo Tools

Extract text from a PDF or image

Upload a scan or a photo and Leo reads the text out of it. No account required to start.

Drop your file here, or choose one

PDF, PNG, JPEG or WebP up to 20 MB

No account needed to start. Your file is processed in this browser tab.

Privacy and file handling

Processed on your device

Your file is processed in this browser tab. Nothing is uploaded until you choose to download or save it.

Removed automatically

A finished file waiting to be downloaded is held for up to 24 hours, then deleted. Files saved to an account stay until you delete them.

Not used for training

Leo does not use your documents or images to train models.

What this tool does

  • Read text from PDFs, PNG, JPEG and WebP
  • Choose the recognition language
  • Live recognition progress
  • Review the extracted text before you keep it
  • Per-line confidence, with low-confidence lines flagged
  • Copy the text
  • Download the text as a .txt file

Supported files: PDF, PNG, JPEG or WebP up to 20 MB.

What it does not do

  • Recognition quality depends on the scan. Clean, straight, high-contrast pages read well; skewed phone photos and faint fax copies do not.
  • Leo does not produce a searchable PDF with an invisible text layer. It returns the extracted text.
  • Handwriting is not supported.

Frequently asked questions

What is OCR?
Optical character recognition: turning a picture of text, such as a scan or a photo, into text you can select, search and edit.
Which languages are supported?
English, French, German, Spanish, Portuguese, Italian, Dutch, Swahili, Arabic and simplified Chinese. Pick the language before you run recognition, since it materially changes accuracy.
How accurate is it?
It depends on the source. Leo shows a confidence figure for the page and flags individual lines it is unsure about, so you can check those rather than re-reading everything.
Do you upload my document?
Recognition runs in your browser. Leo receives the extracted text only when you choose to download or save it.
Do I get a searchable PDF back?
No. This tool returns the text. A searchable PDF with an embedded invisible text layer is not supported yet.

This tool is one small part of Leo

Leo is an intelligent working environment. It understands how an organisation works, keeps track of projects, tasks, meetings and matters, and helps turn what you know into coordinated action. You stay in control of every consequential decision.

Know

Leo holds the context: people, systems, commitments and live work.

Decide

It reads situations, sets out options and prepares plans for you to approve.

Act

Approved plans become coordinated work across people and agents.

OCR a PDF or image online, free. Extract the text.