Turn scanned pages and photos into selectable, searchable text in Hindi, Tamil, Bengali, Telugu, Marathi and ten more languages — without your files ever leaving your device.
Most Indian offices, courts, and small businesses live on scanned documents — a photographed invoice, a scanned affidavit, a faxed purchase order. The text on those pages is locked inside an image: you can see it, but you cannot select, copy, or search it. OCR (Optical Character Recognition) unlocks that text, and MayaPDF does it for 14 Indian languages right inside your web browser.
MayaPDF recognises text in these languages, plus English:
You can OCR a mixed English + Indian-language document — for example a bilingual government form — and MayaPDF will pick up both scripts.
This is the part that matters most for sensitive documents. MayaPDF's OCR runs entirely in your browser using a self-hosted recognition engine (Tesseract, compiled to WebAssembly, served from our own servers — no third-party OCR API is called). When you OCR a scanned contract or an Aadhaar-linked form, that file is never uploaded to any server for processing. Nothing to leak, nothing to subpoena, nothing sitting in someone else's cloud. For law firms, chartered accountants, and government departments, that difference is not cosmetic — it is the whole point.
MayaPDF's OCR produces a searchable PDF: the original page image is kept exactly as it looked, and an invisible, correctly-positioned text layer is placed on top. The page looks identical, but now you (and Ctrl-F, and any PDF reader, and a document management system) can search it, select text, and copy it out. You can also just extract the recognised text if you only need the words.
Because OCR runs on your own device and costs us nothing to serve, we do not cap it. OCR is unlimited in MayaPDF — every language, any number of pages — during the 14-day free trial and on MayaPDF Pro. The trial needs no credit card. (During the trial, exported files carry a small "MayaPDF Trial" footer line; your original upload is always clean. See pricing.)
Good OCR for Indic scripts — with their conjuncts, matras, and complex shaping — is hard, and most tools either send your file to a paid cloud API or ignore Indian languages entirely. MayaPDF bundles the language data and runs the recognition locally, so you get privacy and breadth in one place. It is a genuinely India-first capability, built by a team that has made software in India since 1999.