The Content-Language header was initially presented as a
way to choose the best body part to display to the user,
the best fonts to use, or the best speech synthesis unit
to use for sight-impaired people.
Excellent reminder.
Unfortunatly, it is not simple or (IMHO) desirable to attempt to define a system
to conflate both the audible and visual hints into one header. Far better to
separate them into similar but independant headers.
Content-Audible-Languge
Content-Writing-System
might do. giving us (for documents with uniform content):
Content-Audible-Languge: <Rumanian>
Content-Writing-System: <Roman>
Content-Audible-Languge: <Mandarin>
Content-Writing-System: <GB>
Content-Audible-Languge: <Sindarin>
Content-Writing-System: <Tengwar>
and for marked up body content (by whatever means)
Content-Audible-Languge: Mixed
Content-Writing-System: <Roman>
Content-Audible-Languge: Mixed
Content-Writing-System: Mixed
to give fair warning of poly-content.
I suspect that there is some reference work known to linquists
which would serve our needs better than 639 for values, but I
am ignorant of it, If no one here can come up with anything in
the next coupla days I will take up the matter with librarians
and professori.
Clearly the utility of the audible hints will be limited to availability of
support software, so if any of you is involved
in that area perhaps some references could be turned up that way.