Hello, just FYI - I found a commercial Linux software for OCR called Vuescan which should be able to process also non-English texts.
It needs just libusb and it runs OK, although I haven't got right now a scanner and it doesn't allow me to just load a image and to do OCR... :( http://www.hamrick.com/ http://archives.fedoraproject.org/pub/archive/fedora/linux/core/4/i386/os/Fedora/RPMS/libusb-0.1.10a-1.i386.rpm Because (free) Unix OCR for non-English languages is big tragedy, maybe you find this useful. There's also a Linux version of Cuneiform OCR software which used to be top rated Russian OCR that has been open-sourced: https://launchpad.net/cuneiform-linux http://symmetrica.net/cuneiform-linux/yagf-en.html (QT gui) If you have experiences with OCRing non-English texts, share your knowledge ;) jirib