The Voices in My Head: My Latest Super-nifty Idear

Posted by:: Artemia Salina
Date: Sat, 12 Mar 2005 22:00:21 -0500

--------
First, have a look at http://www.festvox.org.

"Another speech synthesis program. Big whoop," you say.
Well it's HOW it synthesizes speech that makes it interesting.
Basically you take a bunch of digitized speech samples from
a given orator and generate a database to be used by the text
to speech program. In other words, you can get the text to speech
program to SOUND LIKE ANYONE that you have a database for!!!!1

So my idear is to take samples of Rev. Stang's speech from Hours
of Slack episodes and generate a database with his voice, and do
the same with other speakers, like Huey, and Joe Aufricht. Once that
is done I can write a filter script for my newsreader that will
parse the "From:" headers of articles and pipe the text into the
text to speech program using the speech databases that I assign
to different posters. Huey's posts will get processed with Huey's
voice, Rev. Stang's posts with his own voice. Posters like J'lan,
purple, and Quirk can get Aufricht's voice. Posters for whom I do
not have special databases for can use generic voices.

When done I should be able to just click on an article and have
it read to me automatically using that poster's voice!

Hmmm, where did I put those Tex Avery audio clips?

--
0:-) 0:-) 0:-) 0:-) (-:0 (-:0 (-:0 (-:0
0:-) Artemia Salina (-:0
0:-) Surrounded by Angels (-:0
0:-) 0:-) 0:-) 0:-) (-:0 (-:0 (-:0 (-:0

Posted by:: "frater S.O.D.D.I."
Date: 13 Mar 2005 13:40:57 -0800

--------
Unix only.

Poo.