Creating a new language

[GoogleCodeExporter]

m following the instruction of the Train Tesseract section and I would like
to do it with my own language

I understand that I need to create the 8 (empty) files, however if I do
that I cannot run the command:
tesseract fontfile.tif fontfile batch.nochop makebox

since the unicharset file is empty. If I modify the English one according
to my requirements it wont execute either cause in the inttemp file there
are defined 112 characters and not 30 like in my language. I dont know how
to modify the inttemp file so Im stucked.

Any help in here?
Thank you and regards

Original issue reported on code.google.com by jjar...@gmail.com on 8 Apr 2009 at 4:05

[GoogleCodeExporter]

It works for everyone else, but it si complex. Read the documentation at the 
wiki
http://code.google.com/p/tesseract-ocr/wiki/TrainingTesseract
You need existing data files in place when you are running makebox.

Original comment by theraysm...@gmail.com on 9 Apr 2009 at 5:40

• Changed state: WorksForMe