CUNEIFORM(1)

✨ NAME ✨

cuneiform β€” a whole multi-language OCR system, bestie πŸ’– (it lives in section 1 of the manual, iykyk)


πŸ¦‹ SYNOPSIS πŸ¦‹

cuneiform [--dotmatrix] [--fax] [--singlecolumn] [-f format] [-l language] [-o output] input

🌸 DESCRIPTION 🌸

So like, Cuneiform is an OCR system. And not just any OCR system β€” it does text recognition, but it also does layout analysis AND text format recognition on top of that. Main character energy fr fr. πŸ‘‘

And the way it supports several languages?? Obsessed. 🌈


πŸ’… OPTIONS πŸ’…

  • --dotmatrix β€” turns on the recognition mode that's optimized for text printed with a dot matrix printer. yes, the crunchy retro printer vibe πŸ–¨οΈ
  • --fax β€” turns on the recognition mode optimized for text that has been faxed. for all your faxed-document needs, no cap πŸ“ 
  • --singlecolumn β€” disables page layout analysis and just assumes the image is one single column of text. lowkey a shortcut when your page isn't doing anything fancy
  • -f format β€” picks your output format. here's the menu:
    • html β€” HTML format
    • hocr β€” hOCR HTML format
    • native β€” native Cuneiform 2000
    • rtf β€” RTF format
    • smarttext β€” plain text with TeX paragraphs
    • text β€” plain text
and the default? plain text. text. that's the vibe unless you say otherwise πŸ’–
  • -l language β€” by default Cuneiform recognizes English text, bestie. to switch it up, use the command line switch -l followed by a language code (typically an ISO 639-2 three-letter code). here's every language it supports:
    • bul β€” Bulgarian
    • cze β€” Czech
    • dan β€” Danish
    • dut β€” Dutch
    • eng β€” English
    • est β€” Estonian
    • fra β€” French
    • ger β€” German
    • hrv β€” Croatian
    • hun β€” Hungarian
    • ita β€” Italian
    • lav β€” Latvian
    • lit β€” Lithuanian
    • pol β€” Polish
    • por β€” Portuguese
    • rum β€” Romanian
    • rus β€” Russian
    • ruseng β€” mixed Russian/English
    • slv β€” Slovenian
    • spa β€” Spanish
    • srp β€” Serbian
    • swe β€” Swedish
    • tur β€” Turkish
    • ukr β€” Ukrainian

like, twenty-three whole languages plus the mixed one. she's multilingual and she ate πŸ’…βœ¨

  • -o output β€” if you don't define an output file with the -o switch, Cuneiform writes the result to a file called cuneiform-out.format. and yes, the file extension depends on your output format πŸŽ€

πŸŽ€ INPUT FORMAT πŸŽ€

Cuneiform can process any single-page image that GraphicsMagick knows how to open. so like, if gm can open it, we're good 🫢

please consult the gm(1) manual page for the comprehensive list of supported image formats β€” she has the receipts, go ask her.


🌈 HOMEPAGE 🌈

more info about cuneiform can be found at <http://launchpad.net/cuneiform-linux/> ✨


πŸ‘‘ AUTHOR πŸ‘‘

cuneiform was written by Cognitive Technologies and Jussi Pakkanen <jpakkane@gmail.com>.

this manual page was written by Daniel Baumann <daniel@debian.org>, for the Debian project (but may be used by others). so generous of him honestly πŸ’–


ok that's the whole page, thanks for coming to my TED talk 🫢 this was cuneiform version 1.0.0, dated 2010-09-14 β€” yeah she's vintage, and vintage is iconic, periodt πŸ’€βœ¨