Fixing the old font problem
Lots of Tamil documents were made with legacy fonts such as TAM, TAB or Bamini rather than Unicode. Copy text out of those files and you get gibberish. Running the page through OCR sidesteps that entirely: the tool looks at the image and writes out proper Unicode Tamil, whatever font produced it.
What people use Tamil OCR for
- Copying paragraphs from Tamil textbooks and guides
- Turning newspaper clippings into searchable text
- Extracting text from screenshots of Tamil posts and videos
- Typing up printed forms, notices and certificates
- Getting text from old printed books for study or archiving
Grantha letters and numerals
Grantha letters used for borrowed words (ஜ, ஷ, ஸ, ஹ, க்ஷ) are recognised along with the core script. Tamil numerals such as ௧ ௨ ௩ are kept if the page uses them; otherwise standard digits are kept.
Scanned PDFs
This page accepts images only, which keeps each page sharp and fast to process. For a Tamil PDF, screenshot or export each page to PNG and upload up to five at a time.
Mostly handwritten pages? Try Tamil handwriting to text.
Frequently asked questions
Is this Tamil OCR free to use online?
Your first image is free without an account. A free account gives 3 credits and 1 free credit per day.
Can I convert a Tamil PDF to Word?
Screenshot the PDF pages, run them through this OCR, then paste the text into Word. Direct PDF upload isn't supported on this tool.
Does it work on old printed Tamil books?
Yes, as long as the print is reasonably clear. Torn or blotchy pages may produce a few [?] marks.
Will it read handwritten Tamil?
Yes. For mainly handwritten pages, see the Tamil handwriting to text page for photo tips.
What output format do I get?
Plain Unicode Tamil text that you can edit, copy, or download as a .txt file.