Re: pitch shifting with grain synthesis

[Hanno <shaftoe nurfuerspam de>]

Am Samstag, 2. Dezember 2006 18:56 schrieb Stefan Westerfeld:

Im looking forward for the 0.7.1 and with this, I hope, I have again the 
ability to write or experiment with certain own modules. How great the 
current modules are, I discover with my organ-instruments - but there are 
still limits I cant overcome without certain small changes to existing 
modules. Like the missing harmonics in the DavOrgan module for example. 

The pitch-shifting would be nice, yeah, but I think you're right that there 
are a lot of more important modules that are easier to implement and of 
greater worth for the project. 

Anyway, I put some ideas into this list, before the ideas are lost. 
granularsynthesis means also a big hunk of work for me, because I have to 
build instruments for that - which is very time consuming. 

Greetings Hanno

> On Tue, Nov 28, 2006 at 10:55:22AM +0100, Hanno wrote:
> > http://de.wikipedia.org/wiki/Granularsynthese
> >
> > Everywhere you hear pitch shifting, its usually done with grain
> > synthesis. But I think, you know that already. I know Stefan is working
> > on this one. Hope we'll get a demo of that in the near future, Stefan?
>
> I don't know if pitch shifting is really a "near future" thing. Its
> true that I have written some pitch shifting code for arts, there is
> mine, and a contributed fft based version, code is here:
>
> http://websvn.kde.org/tags/KDE/3.5.5/kdemultimedia/arts/modules/synth/
>  -> synth_pitch_shift_impl.cc
>  -> synth_pitch_shift_fft_impl.cc
>
> They could be used as base for a beast version. But right now, the
> remaining issues for the 0.7.1 release of beast are my top priority.
> And after 0.7.1, there are a few changes that I have already started
> working on - like the JACK driver - where only some stuff remains to be
> done. I think its better to finish these before filling my mind with the
> necessary theory for pitch shifting.