Digitise whole books and archives in Urdu and Arabic
For libraries, archives and publishers. LexaOCR turns complete printed books and newspaper runs into editable, searchable text and formatted Word editions, and gives your editors only the words that need a human eye.
Built for whole books, not single pages
- Complete volumes in one upload, up to 1 GB each, read on our servers and saved page by page. Nobody has to keep a browser open.
- Urdu Nastaliq and Arabic Naskh, with several independent readers per line and a dictionary check on every word.
- Only the uncertain words reach your editors, each with the readers' alternatives, instead of proofreading every page.
- Layout kept: columns in reading order, headings, tables, footnotes and verse.
- More from your back catalogue: audiobooks in a calm news-anchor voice, and one-click translation with LexaTranslate.
Who it is for
| Libraries and archives | Make Urdu and Arabic holdings searchable and accessible; export JSON with word positions for your digital library. |
|---|---|
| Publishers | Bring out-of-print titles back as editable files for new editions, e-books, audiobooks and translations. |
| Universities and research centres | Turn primary sources and journals into text for research and teaching. |
| Government and cultural bodies | Digitise gazettes, records and heritage collections, with a data-processing agreement. |
How a project works
- Free pilot. Send us a sample of your collection. We return the finished files and the results measured on your own pages.
- Plan and agreement. Volume pricing, timeline, retention period and a data-processing agreement.
- Digitise. Upload whole volumes in the browser, or we arrange bulk transfer for very large collections.
- Review and deliver. Your editors check only the flagged words; you download Word, review editions, PDF, text and JSON.
Security
Documents are private to your account, encrypted in transit and at rest, and deleted on an agreed schedule. Read our security page and privacy policy.
Questions
How large a collection can you handle?
Each upload can be a whole book of up to 1 GB, and books are read in parallel on our servers. For collections of many thousands of volumes we plan the work with you and can arrange bulk transfer.
How do we know the quality before we commit?
We start with a free pilot on a sample of your collection and report the results measured on your own pages: how many words were verified automatically and how many your editors would check.
What formats do we get back?
A clean Word edition that keeps headings, paragraphs, tables and verse; a review edition with each scanned page beside its text; a searchable PDF; plain text; and JSON with every word and its position, for digital libraries and catalogues.
Who owns the digitised text?
You do. You keep all rights in your documents and in the results, and files are deleted on the retention schedule we agree with you.
Can you sign a data-processing agreement?
Yes. Institutions can have a data-processing agreement, invoices in AED or USD, and a service-level agreement.
Talk to us about your collection
Tell us a little about your material. We reply within two working days, or write to hello@arabianailab.com.