Read and improve a scanned PDF locally
A scanned PDF contains pictures of pages. OCR reads those pictures into text. A readable preview, the correct language and careful review matter more than simply seeing a completed status.
How to do it
Choose the document language
Pick the languages actually present in the scan before converting. Interface language and document language are different settings. The first use downloads the chosen OCR resources.
Review a difficult page
Open the converted result and inspect reading order, tables, numbers and names. Compare against the source when accuracy matters. OCR confidence is not a percentage of correctness.
Try one page improvement
Open Improve this page. Rotate a sideways page, preview straightening or speckle cleanup, then explicitly reread the page. Compare the result and restore the earlier text when the change is worse.
Export only after checking
Review the final text before copying or downloading. Keep the original PDF for signatures, form fields and visual details that are not included in Markdown.
A worked example
Illustrative correction of a sideways synthetic invoice. This example is not an accuracy benchmark.
What goes in
A sideways photograph of a page Invoice 104 Amount: 250
What comes out
## Invoice 104 Amount: 250 Verify the numbers against the source image.
Try the scanned PDF sample first. Page improvements require review and may not improve every scan.
What it will not do
- Handwriting, faint images, complex layouts and unusual fonts can be misread.
- A cleaner-looking image can still produce worse text. Compare and restore when necessary.
- Large scans take time and memory. A phone may have lower practical limits than a laptop.
Questions
- Does image cleanup certify the text?
- No. Review names, amounts and other critical text against the source. OCR is not a certified transcript.
- Can I use it without a connection?
- Load the site and required language resources first. Cached resources and browser storage determine what remains available offline; test before relying on it.
Where your file goes
Nowhere. Reading, converting, cleaning and counting all happen in this browser tab, on your own device. The file, its name and its text are never uploaded to a server, and there is no account to create.
One thing does use the network, and it is worth being precise about: reading a scanned page needs an OCR model, which downloads from this site the first time you convert an image or a scan. That is a program arriving on your device, not your document leaving it. After it has been fetched once, that conversion works offline.