Retyping a page from a book, a letter or a screenshot is slow and error-prone. Optical character recognition (OCR) reads the letters in an image and turns them into text you can copy, edit and search. Modern OCR handles Arabic and English well — if you give it a good image.
What makes OCR succeed
- Sharpness: letters need clean edges. Hold the phone steady and tap to focus on the text.
- Size: small text should be photographed closer; each line should be clearly readable when you zoom in.
- Contrast: dark text on a light background works best. Avoid shadows across the page.
- Straight lines: photograph from directly above. Curved book pages near the spine are the hardest part.
Preparing the image
Crop the image to just the block of text — headers, photos and table edges confuse recognition. Rotate the image so lines are horizontal. For multi-column layouts, crop and recognise each column separately so the lines don’t get mixed up. For a PDF, convert the page to an image first.
Choosing the language
Selecting the correct language matters a lot: the engine uses it to decide which letters are possible. For a purely Arabic page, “Arabic” alone is usually more accurate than “Arabic + English”. Use the combined setting for documents that genuinely mix both, such as forms with English names.
Arabic-specific tips
Printed Naskh-style fonts used in newspapers, books and official letters are recognised best. Calligraphic styles, heavy diacritics and handwriting reduce accuracy considerably. After recognition, check letters that differ only by dots (such as ب ت ث or ج ح خ), as these are the most common errors.
Always proofread
Even a 95% accurate result means a few mistakes per paragraph. Check names, numbers, dates and amounts carefully before you rely on the text, especially for official or financial use.