OCR
Turns scanned pages and images into editable text.
- Text types
- Text in Arabic script
- Text in Latin script
- Input
- PDF, PNG, JPG, WEBP, TIFF
- Output
- Word (.docx), Markdown (.md)
- Models
Features
Nüsha OCR is built for text recognition, transliteration into Latin script and translation of Arabic and Ottoman Turkish sources. Every operation starts from the same work page and is corrected in the same editor.
Turns scanned pages and images into editable text.
Transliterates Ottoman Turkish pages into Latin script, with two spelling options.
Translates recognized works or Word and Markdown files into 22 languages.
Basic, Standard and Pro are the names we give to the AI models we use. Each has its own price per page, and the amount shown on the confirmation screen applies.
| Model | Actions | Note |
|---|---|---|
| Basic | OCR | The lowest price per page. |
| Standard | OCR · Transliteration · Translation | The middle tier. |
| Pro | OCR · Transliteration · Translation | The top tier. |
To see which model suits your work, use your free trial to process the same page with every model and compare the results. Current prices per page
The type of file you upload determines which operations are available.
| Source | Actions | Note |
|---|---|---|
| PDF (including multi-volume works) | OCR · Transliteration | Choose the pages to process; crop, split and rotate them. |
| Images: PNG, JPG, WEBP, TIFF | OCR · Transliteration | Images are collected into a single PDF volume; every frame of a multi-page TIFF is processed. |
| Word (.docx), Markdown (.md) | Translation | The text is already in the file, so no recognition is needed; it is translated directly. |
The tools waiting for you once you create an account.
Your works, their status and outputs in one list, with search and pagination.
Side-by-side editing with the page image, find and replace, and a visible save status.
Low-confidence pages are gathered in one place for you to review.
Every save is kept; you can return to an earlier version when needed.
Helper tools that are free for members and work without AI.
Converts Shamela's HTML book exports and other HTML files to Markdown, keeping headings, footnotes and page numbers. Open the tool →
Conversion runs on our server without AI; your files are not sent to service providers. Results are deleted automatically after 24 hours.
Source files of completed jobs are cleaned up after 270 days for accounts that have paid and after 14 days for other accounts.
You can delete your works and your account at any time. Privacy Policy
Which service providers receive documents, and for what purpose, is published separately. Data Recipients and Transfers
Sign up, process a sample page with every model and compare the results side by side. No payment is needed for the trial.