Orca and non-Latin charactersets

[Jeffery Mewtamer]

I am running a customized Adriane Knoppix 7.2.0 with the version of Orca sourced from Debian Oldstable since the version of Orca installed by default is imcompatible with Iceweasel(the only graphical application I use) Seeing as English is my primary language and I am running the English version of Adriane, Orca's default configuration works for 90% of everything I use it for. However, I have run into two issues involving webpages using non-Latin characters. Issue 1: Greek, the one non-Latin character set Orca seems capable of reading gets read character-by-character instead of string-by-string. Its minor, but it would be nice if I can make Orca read the Greek words I sometimes run into on the web. Issue 2: Japanese characters get read either as "symbol" or "symbol" followed by what sounds like a unicode code point in Hexadecimal. This is a bit more annoying, as I actually know some Japanese and this makes it impossible to work with webpages I could back when I could see well enough to not need a screen reader. Issue 3: This doesn't apply to non-Latin characters, but rather to accented Latin characters, but, I am running Adriane on two different computers, one which has Orca reading accented characters as they should be when used in English text while the other has Orca treated accented characters as separate string(i.e. if cafe was written with the accented e, it would be pronounced caf "e acute"). Its neglible, but I am still curious as to why the discrepency. Any help on resolving these issues would be much appreciated.