Language Identifier List up for comments
Doug Ewell
dewell at adelphia.net
Tue Dec 14 17:39:22 CET 2004
John Burger <john at mitre dot org> wrote:
>> Theoretically, you could create a pronunciation/syntax engine for
>> en-PR as well as en-TX (Texas), en-NYC (New Yawk City), etc, but I'm
>> not sure how well received it would be as a serious tool.
>
> Machine-generated speech is presumably not the only spoken resource
> that needs language codes. I can imagine labeling a library of
> recordings of English speakers from lots of places, and thus needing
> all of en-US-PR, en-US-TX and en-US-NY-NewYorkCity.
One of the "thought experiments" I've been bouncing around concerns the
extension-RFC mechanism in RFC 3066bis, and specifically how one might
construct an RFC to define extension subtags that represent geographical
sub-regions.
If there were such a definition, with (let's say) 's' as the assigned
singleton, then one would be able to say "en-US-s-TX" or "en-US-s-NY-NY"
in the examples above.
"en-PR", of course, can already be used directly.
-Doug Ewell
Fullerton, California
http://users.adelphia.net/~dewell/
More information about the Ietf-languages
mailing list