How it works
OCR: Image / PDF to Text runs on this page, in browser memory. The file you choose is never sent to a server: it is read through the File API, processed locally and handed back as a download.
In three steps
- Choose a file from your device, or drag it onto the drop area.
- Adjust the options on the right, if the tool has settings.
- Press “Process locally” and download the result.
What it takes and what it returns
This tool accepts JPG, PNG, WebP, PDF and produces TXT. The work is done by Tesseract.js.
Real limits
These limits come from what a browser can do safely, not from a paid plan:
- Up to 40 pages per job; split a longer document first.
- The language model downloads once, on first use, then runs locally.
What it does not do
Accuracy depends on the quality of the scan. Skewed, blurry or handwritten text is recognised poorly.
OCR reads images and PDFs. It does not transcribe audio recordings.
Common questions
Do my files reach a server?
No. Processing happens in browser memory. The network is used only to download the application's libraries, and your file's contents are not part of those requests.
Which formats does OCR: Image / PDF to Text accept?
It accepts JPG, PNG, WebP, PDF and produces TXT.
Do I need an account?
No. The tool works without an account and without installing anything during the free pilot.
Does it work on a phone?
Yes, in current browsers. Very large files can exceed the memory a phone has available, in which case the job stops with a clear message.