How it works

1. 01 Add images or a PDF

   Drop, paste or browse any photo format or a PDF. Each PDF page becomes its own page in the workspace.

2. 02 Choose a language and recognize

   Pick Vietnamese, English, or both, then recognize all pages. Processing runs locally using Tesseract.js.

3. 03 Review side by side

   Compare the recognized text against the source image for each page, and fix any mistakes directly in the text box.

4. 04 Export your result

   Download the combined text as TXT, a Word document, or a searchable PDF with an invisible, copyable text layer over the original image.

FAQ

Are my images or PDFs uploaded anywhere?

No. Recognition runs entirely in your browser using Tesseract.js. Only the OCR engine and language data (not your files) are downloaded from a CDN the first time you use it, then cached. A recovery copy of your current session is stored on this device and expires after 48 hours.

Which languages are supported?

Vietnamese, English, or both combined in a single pass, for documents that mix the two scripts.

What does the searchable PDF export do?

It draws the recognized words as invisible, selectable text directly over the original page image, so the PDF looks unchanged but its text can be selected, copied and searched.

What are the limits?

Images up to 30 MB, PDFs up to 50 MB, and up to 30 pages processed per PDF. Very large or low-quality scans take longer and may recognize less accurately.

Can I fix mistakes in the recognized text?

Yes. Each page's text is editable — correct it directly before exporting.