[LAU] OT - impulse response recording

Kevin Cosgrove kevinc at cosgroves.us
Sun Jan 30 19:59:10 UTC 2011


On 30 January 2011 at 11:43, =?ISO-8859-1?Q?J=F6rn_Nettingsmeier?= <nettings at folkwang-hochschule.de> wrote:

> On 01/30/2011 10:39 AM, Kevin Cosgrove wrote:
> >
> > I think I know how to get aliki to create an impulse response.
> > But, I'm unclear as to how to use that as an input for DRC
> > (digital room correction).
> 
> just export wav files from aliki.

Great.

> > I'm also unclear as to how to use
> > the impulse response to get at things like RT60,  clarity and the
> > like.  I *think* that if I could take the IR out of aliki and
> > bring it into octave, Maxima, or Matlab, that I could figure it
> > out from there.  From the aliki docs it looks like exporting
> > of files outside it's initial format is planned, but not yet
> > implemented.  Do I have this right?
> 
> no. aliki does wav export just fine.

Fabulous.

> > Another thing I might want to do is to have a look at a time
> > plot including incident and reflected pulses.  The time between
> > those should be an indicator of the total path length between
> > the source and the mic.  By inspection of the acoustic space,
> > then I ought to be able to figure out about where reflections
> > are happening (thinking method of images) in that space and
> > apply sound treatments accordingly.  At least, that's my wistful
> > hallucination.
> 
> no problem there.

Whew!

> > The pictures of impulse responses in the various
> > docs I looked at weren't zoomed in enough for me to really tell
> > what might be going on, and they were probably of real spaces,
> > and not simple textbook rectilinear polyhedrons.  I wonder, is
> > aliki a tool that could help in this endeavor, if my idea is
> > worth attempting at all.
> 
> it certainly is. what are you up to? first you mention doing DRC in 
> actual rooms, last paragraph seems to indicate you want to do some 
> theoretical stuff? simulations?

The task I'm starting to plan/attempt is to get my music room treated 
and deal with the remaining artifacts with DRC.  For a simple room, 
e.g. 2 paths from source to listener, I should be able to see the 
incident and 2nd path in the impulse response.  In that case I can 
identify the location of the reflection from the time delay and wall 
location.  In a more complicated room this might not be intuitive.
I was going to leave it there.  But, it should be possible, once a 
reflection is identified, to remove it from the impulse response 
mathematically, leaving the other reflections, which would then show 
up more clearly.  I'm not planning to go down that road, just wanting
to understand.

Cheers....
 
--
Kevin




More information about the Linux-audio-user mailing list