Accurately extract Arabic text from PDFs and images
Turn your scanned Arabic documents into editable text with OCR and AI.
Extract Arabic text from PDFs and images, keep the content's order and formatting, then export it to Word with ease.
No credit card. No page bundles to buy. Signing up takes about a minute.
- No software to install
- Word (.docx) export
- Page-level retries
- Arabic & English interface
بموجب هذا العقد المُبرم بين الطرفين، يتعهد الطرف الأول بتسليم البيانات المستخرجة كاملةً خلال مدة لا تتجاوز ثلاثين يوماً من تاريخ التوقيع.
Arabic text extraction
Extract Arabic text from PDFs and images quickly and accurately.
Text direction preserved
Detects right-to-left writing and keeps it intact.
Word export
Export the extracted text to a Word file with ease.
Your files stay private
Your files are processed securely and their content is not used for any other purpose.
Designed for Arabic documents
Accurate Arabic text extraction, with support for right-to-left writing.
Arabic is more than text read from right to left. Varied fonts, connected letters, diacritics and the quality of scanned documents all affect how accurately text is extracted.
OCR tools built around foreign languages
- Splits connected Arabic letters into separate glyphs
- Drops or misplaces diacritics and hamza forms
- Reverses reading order, so paragraphs come back backwards
- Fails the whole file when a single page is unreadable
- Exports left-to-right documents you have to reformat by hand
Istakhrej
Istakhrej
- Reads Arabic words as words, ligatures included
- Keeps diacritics, hamza forms and Arabic-Indic numerals
- Returns text in correct right-to-left reading order
- Retries only the pages that failed, without a re-upload
- Exports a .docx that is already right-to-left
Everything you need to extract Arabic text from your documents
Whether it is a scan, an image or a PDF, you get editable, ready-to-use Arabic text in minutes.
Arabic OCR, not an afterthought
Tuned for Arabic script: connected letters, diacritics, hamza forms and Arabic-Indic numerals are read as written, not guessed.
Keeps the direction of Arabic writing
Automatically detects the direction of Arabic text and keeps it in right-to-left order.
Extract text from images and scanned documents
Easily extract Arabic text from images, PDFs and scanned documents.
One bad page will not cost you the file
Failed pages are listed on their own and can be retried in place, without re-uploading the document or starting over.
Word (.docx) export
Download the extracted text as a Word file and use it right away in your work.
Bilingual interface
The whole product — landing page, dashboard and exports — works in Arabic and English, with a proper RTL interface in both.
Three steps, no complicated setup
Upload your file, extract the text, then download it in the format that suits you.
-
1 Takes seconds
Upload your file
Choose a PDF or image that contains Arabic text.
-
2 Usually a couple of minutes
Extract the text
Let the system recognise the Arabic text and turn it into editable text.
-
3 You approve before you export
Download the result
Download the text as a Word file and start using it right away.
You are using the beta version. For now, only a limited number of pages per document are processed.
Try it freeBuilt for Arabic documents
Whether you work in archiving, research, education or business, the tool is built to make scanned Arabic documents easier to work with.
Legal & contracts
Turn scanned agreements, judgments and official gazettes into searchable text you can quote from.
Research & academia
Digitise scanned journals, theses and manuscript transcriptions without retyping a single line.
Archives & records
Make decades of paper correspondence and ministry records findable by keyword instead of by folder.
Business reports
Pull figures and paragraphs out of scanned invoices, statements and annual reports for reuse.
Straight talk about your documents
You should know exactly what happens to a file after you upload it, so here it is in plain language — including the part most tools leave out.
Data Usage Notice
During the beta version, the documents you upload may be used to train and improve our models. Please do not upload sensitive or confidential information.
This applies during the beta only.
Your documents stay yours
Every document is scoped to the account that uploaded it. Other users never see your files or your extracted text.
Working files are cleaned up
The page images and temporary PDF chunks created during extraction are deleted automatically once a document finishes.
Third-party AI is involved
Recognition runs through external AI providers. That is how the accuracy gets there — and why confidential material does not belong here yet.
Frequently asked questions
Nothing while the service is in beta. There is no card to enter and no page bundle to buy. If paid plans arrive later, existing accounts will hear about it well before anything changes.
No. It runs in the browser — no plugin, no desktop app and no command line. Any modern browser on a laptop or a phone will do.
Arabic is the priority, including documents that mix Arabic with English or French. Latin-script text on the same page is extracted alongside the Arabic.
PDF files plus JPG and PNG images, up to 10 MB per file.
During the beta, the first 10 pages of each document are processed and the rest are ignored. The limit is a configuration value and will rise as the service matures.
Clean, high-contrast scans come out very close to the original. Faint photocopies, heavy handwriting and dense tables are harder, which is why every page is shown to you for review before you export.
You read the extracted text page by page before exporting anything, so a mistake is caught in the dashboard rather than in a file you have already sent to someone.
Uploads are processed by third-party AI models and, during the beta, are also used to improve the service. Please do not upload confidential or personal information.
Sign up yourself — it is free and takes a minute. We only ask for your name, email, a password and where you heard about us.
Still unsure? Ask us directly
Turn your scanned Arabic documents into text in minutes
No need to retype Arabic text by hand. Upload your file and extract editable text with ease.
Creating an account means you accept our Terms & Conditions.
Free while in beta
No card required