Unfortunately, the inter-next site is temporary closed for renovation.

I'm assuming that at the moment the only solution to the problem is a text
to voice conversion done somewhere between the server and the users.  The
question is where is this done and should it be the responsibility of the
server or the user to arrange for this ability.  Maybe it's because I use to
run a web server, but I think the responsibility for making information
available in a usable form should lie with the centralized server and not
the distributed user base.

I'm real interested in seeing how inter-next does it's think.  Depending on
their solution, this might have interesting implications on the best way for
the visually impaired to access web pages.
I'll get back to you when I've had a chance to take a look at what they
have.