Free tool

Marathi image to text, read on your device.

सातबारा, शासन निर्णय, marksheets, leave-and-licence agreements and book pages in मराठी, read in your browser and given back as text you can paste, search or translate. English on the same page is read too.

2. The languages on the page, up to 3
Indian languages are read along with English, for the numbers and English words.

The first time, your browser fetches the reader: about 9 MB, and 1 to 6 MB for each other language. After that, a page takes a second or two, even a tilted phone photo.

Runs in your browser. Your files are never uploaded.

The text appears here, ready to copy or fix.

Several files, or a whole form? Tell Sochen “Read this scanned rent agreement and tell me the notice period and the deposit” and it does every step, checks each against the notification, and still keeps your files on your device.

Try Sochen free →

How Marathi reads

  • Marathi is written in Devanagari, like Hindi, but with its own letters and spellings: ळ appears constantly (in words like महाराष्ट्र, कळवणे), and the eyelash-ra in words like दर्‍या or तऱ्हा. The Marathi model knows these; the Hindi one guesses.
  • Maharashtra government documents, land records and older books often print numbers as Devanagari digits ०–९. The reader gives them back as Devanagari digits. If a form needs 0–9, retype them or use find-and-replace.
  • Anusvara (ं) is everywhere in Marathi (केंद्र, संस्था) and is a single dot above the headline. On a faint scan it disappears, which turns a word into a near-miss.
  • Conjuncts in Marathi include some Hindi rarely uses (ज्ञ read as dnya, श्र, त्त्व). They are single shapes on the page and are the words to check first.

Marathi documents people bring

  • 7/12 extracts (सातबारा उतारा), 8-A extracts and property cards from Mahabhulekh, to read the survey numbers and owners' names.
  • Maharashtra government resolutions (शासन निर्णय) and circulars, to search for a clause or translate.
  • SSC and HSC marksheets and school leaving certificates from the Maharashtra board.
  • Leave-and-licence agreements and society letters typed in Marathi.
  • Pages of Marathi books, newspapers and exam notes.

Getting a better read

  • Pick Marathi rather than Hindi for a Marathi page, even though both are Devanagari. The Marathi model's dictionary chooses Marathi spellings and reads ळ correctly.
  • 7/12 extracts are printed in small type with ruled boxes. Scan at 300 DPI or zoom the phone camera so a row of the table fills the width; the reader straightens the lines, but tiny text stays tiny.
  • Check every number in a land record or marksheet against the original. Devanagari digits look alike at low resolution (१, २, ३ and ७, ८ especially).
  • An old Marathi PDF typed in a Shree-Lipi or DV-TTYogesh font copies out as gibberish. Choose only Marathi and the reader reads the picture instead.

Worth switching on alongside Marathi: Hindi, if the page mixes both. The reader takes up to three languages, and English is read with every Indian language automatically.

Scanned PDFs are read page by page, up to 30 pages at a time; a PDF with its own text layer is copied exactly, without reading the picture.

Questions

Should I choose Marathi or Hindi for a Marathi document?+
Marathi. The script is the same, but the Marathi model expects Marathi words and letters like ळ, so it makes fewer spelling mistakes. Add Hindi only if the page has Hindi passages too.
Why are the numbers in my 7/12 extract coming out as ०, १, २?+
Because that's what's printed: Mahabhulekh prints Devanagari digits. The reader keeps them as Devanagari so nothing is silently changed. Retype them as 0–9 if a form needs that.
Can it read a photo of a Marathi handwritten application?+
Only roughly. The reader is trained on printed type. Use it for typed and printed pages.
Does the file get uploaded?+
No. The Marathi model is fetched to your browser once and the reading happens on your device. Nothing is sent to Sochen.