[LAU] OT - impulse response recording

[Ken Restivo]

On Fri, Jan 28, 2011 at 04:44:47PM +0100, fons at kokkinizita.net wrote:
> On Fri, Jan 28, 2011 at 03:37:21PM +0100, Renato wrote:
> 
> > fons at kokkinizita.net wrote:
> >
> > > Starter guns, electric sparks, exploding balloons or preservatives,
> > > etc. are not the best way to do this. You get a much more accurate
> > > result and a much better S/N ratio using a sine sweep and
> > > deconvolution instead. Aliki (available on my website) can do this.
> > > See <http://www.kokkinizita.net/papers/aliki.pdf> for an introduction.
> > > The actual application does not have all the features explained in
> > > the paper but it will do the basic IR measurement.
> > 
> > question: why the amp/speakers don't have to be high quality
> > flat-response monitors (as I think Julien is saying), but the mic does?
> 
> They have to be of reasonable quality and FR, but any remaining
> small errors can easily be calibrated out. You just take the
> direct sound (for a good speaker + mic this will be the first few
> ms of the IR) as a reference. 
> 
> A few percent distortion (most speakers will have this) does not
> matter if the measurement is done in the correct way: after the
> deconvolution the distortion products will appear *before* the
> real IR and they can be removed easily.

Really? So that little bit of mung before the "snap" in an IR is distortion, not "predelay" or "early reflections" or any such thing? Wow, that's good to know. I have to get busy and clean up some IR's then.

> 
> > Also, do mic vendors offer files with the frequency response of the mic
> > so one can "multiply the inverse of it" with the spectrum of the
> > signal recorded with said mic, taking mic charateristics out of the
> > equation (at least theoretically)?
> 
> The high quality ones will have tight specs and usually come with
> a measured FR or other calibration data.
>  
> There are a few other points to consider. 
> 
> Most speakers will be directional at all but the lowest frequencies.
> It depends on your application if this matters or not. If it does
> (e.g. to measure room acoustics according to international standards)
> you need an omnidirectional speaker. These are usually made as a
> dodecahedron of small speakers. See <http://www.gizmag.com/go/3894>
> for an example.
> 
> Also omni mics will have some directionality at high frequencies, and
> that is why they come in two forms: calibrated for flat FR on-axis
> (plane wave), or for flat FR in a diffuse field (averaged over all
> directions). The difference matters only at very high frequencies.
> Some measurement come with a small conical diffuser which can be
> attached in front to improve diffuse-field response.
> 
> Ciao,
> 
> -- 
> FA
> 
> There are three of them, and Alleline.
> 
> _______________________________________________
> Linux-audio-user mailing list
> Linux-audio-user at lists.linuxaudio.org
> http://lists.linuxaudio.org/listinfo/linux-audio-user