text-to-speech API? SAPI?

[Nicolas Cueto]

Hello,

Minutes ago just had a bit of a minor programming epiphany.

A teacher on an EFL (English as a Foreing Language) mailing list pointed me to:

http://neospeech.com/

It's a text-to-speech service, and I was amazed how well intelligible
and ear pleasing it was. The state of the art has certainly improved
from what I remember way back when!

So, seeing profound new possiblities for my language classroom,  I
immediately popped open Rev to see if it had text-to-speech functions
and, lo and behold!, there was the revSpeak command.

Gave it a whirl with...

revSpeak "hello world'"

... and two reactions.

First, wow that was simple. Rev is brilliant!

Second was, what the hey!! For my target users -- young EFL learners
-- not only is "Microsoft Sam" unintelligible but he'd either give
them the giggles or scare them into tears.

Looking at the Rev documentation, I guess the problem is the API. But
this is all very new to me, so I don't really know.

If the holy grail of voice-quality I'm after here is the API (or is it
SAPI?), would anyone have a recommendation or experience with great
Windows text-to-speech APIs/SAPIs?

I'd of course also welcome general advise about text-to-speech
APIs/SAPiS, especially how to integrate one with my standalones.

Thank you.

--
Nicolas Cueto