LANGUAGES · 100+ SUPPORTED
Image to text in 100+ languages
Pick the language of your image or let the tool guess. Both work. Language support covers the scripts people actually write in — Latin, Cyrillic, Arabic, Hebrew, CJK, Indic, and Southeast Asian.
Most used 18
Latin script 40
Cyrillic 10
Arabic & right-to-left 10
Chinese, Japanese & Korean 4
Indic scripts 14
Southeast Asian 6
Other scripts 7
A few honest notes on language support
Auto-detect is a starting point
If you leave the language on auto-detect, the converter looks at the letter shapes and picks the most likely script. It's usually right for whole pages of text, and it can be wrong for very short snippets. For a document in a single language, choosing it manually is always the safer option.
Right-to-left text
Arabic and Hebrew are read right to left, and the extracted text keeps that reading order — useful when you're pasting into a document or editor that expects it.
Accuracy varies by script and image
This is the part nobody likes to hear: OCR accuracy depends more on the quality of the image than on the language. A sharp, flat scan of Cyrillic or Hangul converts very well. A blurry phone photo of a Japanese sign at an angle? Less so. The accuracy article explains the fixes that actually work.
One language per image
The engine works best when an image contains a single language. A document that mixes two scripts wildly — headings in one, body in another — will return text with more errors in the secondary script. It happens; both languages usually still come through well enough to read.
What language is in your image?
Find it in the list, or just let the converter figure it out.
Try it now