Why is there not much work being done with voice synthesis? It was all the
rage for a while and now you don’t hear much at all. Even my Commodore 64
could talk without samples. Anyone remember SAM, the Software Automated
Mouth?
They are already making movies with computer-generated actors but we seem
to be stuck with the idea we still need real voice actors. Is it because we
will only pay to hear high paid famous people or maybe the voice synthesis
cannot give us that realistic emotion we need in the dialogue.
Surely someone is working on bringing us emotional sounding computer
synthesised speech?
This should be the next generation of computer interaction. No doubt it
will need some kind of markup language to set which emotion. Or perhaps
CSS? Cascading Style Speech 
Oh please let me be the first to coin the
phrase!!
If the Microsoft Text to Speech application were upgraded with emotional
characteristics, I could get rid of the blue screen and just have it scream
"OH, DAMN!! NOT AGAIN!!". Pebcak errors could be replaced with "You EEDIOT!
Are you trying to kill me, man??". Although a good Windows sound theme
could already do that with a nice WAV sample.
The day when movies are totally created by computers including both
graphics and sound cannot be too far ahead. They’d require a server farm
for voice synthesis to go with the server farm of graphics generation they
already have.
I look forward to the day this dream becomes a reality. Now all we need is
ScriptWriter 4000…
