On Fri, 24 Apr 2009, John Lato wrote:
On Fri, Apr 24, 2009 at 6:00 PM, Eric Cheng <[email protected]> wrote:
Hello folks,
It sounds like there have been a lot of exciting things about Haskell
audio going on lately!
I agree with Balazs that a cross-platform solution would be very nice.
Here at Yale we've been designing the next-generation of the HasSound
real-time sound synthesis DSL. Much of the focus has been on the
theory side, but that has been quite fruitful. We have been able to
derive a new class of Arrows and combine it with fusion techniques,
and preliminary tests show that performance is really, really good.
(By that I mean it is competitive to straight-ahead hand-written
C/C++.)
Is any of this work available yet? I've recently been heading in the
direction of an arrow-based DSP/synthesis DSL as a back-end for one of
my projects, but I haven't put much time in yet. If your group has
already been doing something similar, I'd much prefer to use that
(after a stable release) and spend my time on other stuff, provided
your features overlap with what I need.
Actually, I have also introduced an Arrow into my synthesizer package some
time ago for handling causal processes. I always had the impression, that
Arrows must be something for signal processing, and I tried to use them
for abstraction from the sample rate. But this didn't work. For sample
rate abstraction I use a Reader monad/applicative functor today. However
the Arrow interface proved really useful for the description of causal
processes and can be compiled to fast code without interim signal data
structures.
http://hackage.haskell.org/packages/archive/synthesizer/latest/doc/html/Synthesizer-Causal-Process.html
http://darcs.haskell.org/synthesizer/src/Synthesizer/Causal/Process.hs
_______________________________________________
haskell-art mailing list
[email protected]
http://lists.lurk.org/mailman/listinfo/haskell-art
