Quick guide
Thai OCR for Images and Scanned PDFs
Extract Thai or English text from photos, screenshots and scanned PDFs with in-browser OCR. Rotate and enhance pages, choose a layout, then review, copy or download TXT without sending the document to the VPS.
What this tool does
- Supports Thai, English and bilingual documents.
- Reads JPG, PNG, WebP and PDFs up to 20 pages, one page at a time.
- Rotate pages and stop after the current page while keeping completed text.
- Images and recognized text are not sent to AI or the VPS.
How to use Image/PDF OCR to Text
- 1Choose an image or scanned PDF up to 50 MB.
- 2Choose the language, rotate the page and enhance the image for clear text.
- 3Run OCR, then check names, numbers and Thai marks against the original before copying or downloading.
Frequently asked questions
Which type of picture is the most accurate to read?
Use a straight, sharp, evenly lit image and avoid shrinking Thai marks until they break up. Rotate an upside-down page and compare results with contrast or threshold enabled and disabled.
Can OCR preserve handwriting, tables or the original layout?
OCR works best with clear printed text. The output is plain text, so tables, columns, fonts and positions are not preserved, and handwriting may be misread. Always check names, document numbers, amounts and Thai marks.
Which files and page limits are supported?
Supports JPG, PNG, WebP and PDF up to 50 MB, with PDFs limited to 20 pages. Each page is capped at 2,400 pixels or 6 megapixels and processed one at a time to control device RAM.
Can I stop a multi-page OCR job?
Yes. Choose stop after the current page. OCR finishes that page and keeps completed text for review or download.
Are files, images or recognized text sent to the VPS?
No. Documents and text stay in the browser. The VPS only serves the page, OCR engine and language data. Analytics records bounded start, success, error and download states without file names or content.
Why does it take so long to open for the first time?
The browser downloads the OCR core and Thai–English language data from this website on first use. It can then reuse the cached files on the device.