That's OK with me, if the consensus is to include my code in the same package, although I'm not using any of the functions in the Audio pkg (at first, I thought I might, but I haven't had any reason to do so). However, all my code is really aimed at musical applications, and I don't think it would be of any use to people interested in voice recognition, speech generation or other audio apps. I'd prefer to make it a separate package (called 'music', for example).
I do have a couple of questions about the existing code in the Audio pkg. First of all, the 'aurecord()' function seems to be missing from the Windows binary (I'm using Octave 3.0.1).
Secondly, on line 41 of 'au.m' (version 1.1.1 of Audio), there's the following code, which seems to be meant to swap the values of two parameters, but it doesn't seem like it would work to me:
  if hi<lo, t=hi; hi=lo; lo=hi; endif
Shouldn't it be changed to this?
  if hi<lo, t=hi; hi=lo; lo=t; endif
Is there any one on this list that's interested in using Octave to make music? In particular, if any has used Csound, Nyquist, or Haskore (an add-on package for Haskell), I'd be interested in getting feedback from them, in regard to the idea of using Octave for musical apps. Any suggestions would be appreciated.
Also, I'm doing my development on Windows XP, using Octave 3.0.1. If anyone is using Octave on other platforms and would like to test it there, I'd appreciate it.

I'll try to post some code by the end of the week. At the moment, I've got all my code in a single file. Is it a standard or convention to have one source file per function? If so, where should global variables be defined? Or is the use of global variables frowned upon?
Thanks for your help.
On Mon, Jan 5, 2009 at 10:18 AM, Søren Hauberg <> wrote:
man, 05 01 2009 kl. 09:54 -0500, skrev David:
> They have a lot in common, of course, but I'm wondering if I should
> try to extend the Audio pkg, or start from scratch, building on the
> built-in audio functions documented in the Octave reference manual.

I think you should just extend the audio package. From what I remember
it is currently a fairly simple package, so you should easily be able to
wrap your head around the code.

> I'm not sure how the process works as far as contributions are
> concerned, either. Who evaluates the contributions, tests them, etc.?
> Or is there any review process at all?

The procedure is very open to new comers. Basically, you should send
your first contribution to this list. Then somebody will have a look,
and see if things are okay. If so, you will get access to commit your
code to SVN. This check is mostly there to keep the number of inactive
accounts low. So, that's the review process.

If you've written some code that you would like to commit, but you'd
like some comments on it first, then you can always send it to this
list, and hope somebody will comment. However, we don't have a clear
policy here, except we try to help each other out as much as
possible :-)