Re: "El Cheapo" dummy heads?

[Henrik Møller <hm@xxxxxxxxxxxx>]

This is a multi-part message in MIME format.
--------------BB6E4C26C8F9D05B9BA48212
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit

Dear List

Interesting so many responses - public or privately to me - there have been
concerning dummy heads. It seems that many people are relaxed and ready to
"play" in the summertime..... and it is indeed a fascinating technique.

I have had several questions about specific microphones - I think they have all
been mentioned in public postings, so I won't repeat them. We use Sennheiser KE
4-211-2, simply because they were the first choice that we - more than 10 years
ago - came across, which could serve the purpose. And we have not had enough
good reasons to change, but there may be better ones available now. I haven't
tried those mentioned in the postings.

The microphone must be a pressure type microphone (i.e. of the omnidirectional
type, although it will not at all perform as omnidirectional, when it is mounted
in the ear). And, of course, it should have adequate specifications for
frequency range, sensitivity, noise floor, clipping level, stability, matching
between the two, etc.

Furthermore, the microphone should be small enough not to disturb the sound
field it is recording, and this leads me to some questions about microphone
position. The directional information is "encoded" by the outer geometry of the
"head", i.e. the head itself, body and shoulders, and especially the pinna. The
directional information is present when the sound reaches the ear canal. The
further transmission along the ear canal does not add any directional
information, since the same happens to "all the sound", never mind where it
originally came from. (Of course, the travelling along the ear canal changes the
sound, but that's a matter of equalization).

Consequently, you can pick up the sound anywhere from the entrance of the ear
canal to the eardrum. We have found the entrance to a blocked ear canal
especially useful, since it leaves space for the microphone in the earplug, and
it has a number of other advantages (one is that the sound varies less between
people than it does further inside the ear canal and woth the ear canal open).

If you record further out than the entrance to the ear canal, you will get wrong
directional filtering. The technique will not work properly with microphones in
the back of the concha, outside concha, on your glasses or whatever funny
proposals I have seen (- some of them even offered commercially). You may get
the sound external, and also get the feeling of a binaural recording, but the
auditory space is much disturbed, in particular with respect to front/back and
up/down (median-plane and within-cone errors). Actually, some of the microphones
I have seen advertised, may be quite difficult to put in a proper place without
disturbing the sound. Don't fill up the concha with the microphone!

I think it was Jim, who reported of an unsuccessful recording with supraaural
headphones used for recording. I guess that, when used as microphones, these
headphone capsules may work more like pressure gradient mics than as pressure
mics. I have no idea of how the pressure gradient would look like around a
pinna, heavily disturbed by a headphone.....

I have also got questions about the equalization, e.g. don't we get the ear
canal two times?, or don't we lack the ear canal?

The general rule is that you should equalize the headphone to a flat frequency
response at the position where you record. And most easily this is done with the
recording microphone in position - then you can equalize for the microphone,
too, in the same step.

If you do it this way, the ear canal comes in exactly one time as it should.
(Examples (popular and simplified version): A) If you record at the eardrum, the
ear canal transmission is there during recording, and if you use an external
headphone, the canal is also there during headphone measurement, and since the
measurement is used for equalization the two will cancel. Finally, the ear canal
transmission is there during reproduction, which makes one time ear canal
transmission. B) If you record at the eardrum and use an insert earphone for
reproduction, the ear canal transmission is there during, not during measurement
and not during reproduction. Again one ear canal transmission in the result).

Somebody asked, if an insert earphone would be better than supraaural
headphones. No, it does not make any difference, if you equalize properly. (Of
course, you will run into practical problems, if you try to measure an insert
earplug with a blocked entrance microphone......., so that combination requires
more sophisticated actions......)

I made a detailed description of the equalization of binaural recordings some
years ago. For those interested, it appears in:

- Henrik Møller: "Fundamentals of binaural technology", Applied Acoustics, Vol.
36, No. 3/4, 1992, pp. 171-218.

It may be heavy and boring reading, so don't bring it to the beach. Anyway, good
luck with your summer projects....

Henrik

--------------BB6E4C26C8F9D05B9BA48212
Content-Type: text/x-vcard; charset=us-ascii;
 name="hm.vcf"
Content-Transfer-Encoding: 7bit
Content-Description: Card for Henrik Møller
Content-Disposition: attachment;
 filename="hm.vcf"

begin:vcard
n:Møller;Henrik
tel;fax:+45 9815 2144
tel;work:+45 9635 8710
x-mozilla-html:TRUE
url:http://acoustics.dk/
org:Aalborg University;Department of Acoustics
adr:;;Fredrik Bajers Vej 7 B4;Aalborg Ø;;DK-9220;Denmark
version:2.1
email;internet:hm@acoustics.dk
title:Professor, Head of Department
fn:Henrik Møller
end:vcard

--------------BB6E4C26C8F9D05B9BA48212--