Just realizing that these 2 language can also be encoded with these
charsets (even though ISO-8859-13 would appear to be more common…
maybe?). Anyway now the models are updated and can recognize texts
using these encoding for these languages.
Added some test files as well, which work great.
