On Wed, 2007-08-15 at 08:45 +0100, Nick Collins wrote:
> it's great fun
>
> and who needs LPC or concatenative synthesis when you can make good use of
> Karplus-Strong and a Formant filter... thanks for the description below.
My pleasure, I'm looking forward to trying it out in 'performance'. By
the way once I have a proper working system in a couple of weeks it
would be great to get some user feedback, so if anyone in London or at
the dorkcamp doesn't mind sit before it, playing with it and maybe
filling out a questionnaire that'd be great... And once the thesis is
handed in I'll release it under a GPL.
> Alex, you're aware of the component of the Cylob Music system for speech
> synthesis aren't you?
No I wasn't, thanks a lot! I've dropped him a mail asking if there are
any recordings of it in use...
> he doesn't theorise it in terms of live coding and the possibilties you
> show in the video, but it might be worth referencing? He uses a DIY speech
> synthesis approach like you that gives speech with (uncanny?) character,
> having spent a long obsessive time himself mapping out English phones...
Ace! I'm a real stickler for obsessive mapping although my own mapping
only took about 10 minutes, I just used a mouse to pick values for the
two parameters that were a bit like each consonant. Yep definitely
worth referencing. Another good reference in this area is David Evan
Jones, who wrote a couple of interesting papers about work by himself
and others using speech in music. The ones I've found so far are
"Speech Extrapolated" and "Compositional Control of Phonetic/Nonphonetic
Perception".
alex
Received on Wed Aug 15 2007 - 18:21:57 BST