ALTHOUGH MOST AUDI OPHJ LES ARE

familiar with the term binal~ral,

there's still quite a bit of con-
fusion about it. Early in stereo
history the terms binaural and
stereo were used interchangea-
bly, even though the two record-
ing methods are totally different.
Recording pioneer Emory Cook
caused some of that confusion by
cal l i ng hi s earl y 50's t wi n-
grooved stereo LP's bi naural
when they were actually stereo.
Binaural recordings can open
up a whole new realm to your 11s-
tening enjoyment. We'll look at
t he hi st ory behi nd bi naural
sound and how it is made as well
a s present some high-quality
binaural products.
The binaural difference
If you listen to a stereo source
from stereo headphones com-
pared lo the same source listened
to from l oudspeakers, you'll
notice a much different sound
between the two. That's because
most source material isn' t de-
signed for headphone listening.
An unnat ur al l y exaggerat ed
effect i s creat ed wi t h head-
phones, as though half an or-
chestra is on one side of your
head and the other half on the
other side, with a hole in the mid-
dle. Also, the music sounds as if
it's happening inside your head
rather than out in the room. No
serious record producer would
ever monitor a recording session
solely on headphones; a proper
set up of t he hi ghest quality
moni t or l oudspeaker s i s re-
quired to get a feeling for proper
balance in the mix.
Binaural background
True binaural uses only two mi-
crophones, usually small electret
condensers either set into the
outer ears of an artificial human
head. or at least spaced the same
distance apart as an average pair
of ears, and mounted on either
side of a small baffle. The two
INAURAL
Binaural recordings
will add a new
dimension to
your audio world.
JOHN SUNlER
mikes feed two channels which
are kept entirely separated from
the source all the way to the final
listener, whether live, a record-
ing, or a broadcast. The listener
wears stereo headphones and the
original left ear signal must be
routed properly to the left ear and
the right to the right or the effect
is compromised. The final result
is for the listener to be sonically
transported to where the sounds
originated, rather than attempt-
ing to bring the sounds into the
listener's room as with speakers.
The left speaker signal is pre-
vented from feeding into the lis-
tener's right ear, and vice versa,
with binaural playback on stereo
headphones. Figure 1-a-d shows
various types of sound reproduc-
tions, including binaural.
With binaural recording, spa-
tial placement within a 360-de-
gree sphere is so realistic t hat
even vertical placement is per-
ceived. Reproduction of the am-
bience or reflected sounds in a
hall is so correct that acoustical
engineers can listen t o s uch 0
tapes and identify in which hall 8
they were recorded. The only
g
areas of location that are some- n
times problematic are on a line $
directly in front of and to the rear
of the listener. That is dependent
on several factors, including how
our hearing mechanism works,
differences in headphones, and
individual differences as well as
learned responses.
Binaural history
The first use of binaural sound
occurred in 1881 in Paris. Inven-
tor Clement Ader mounted a se-
ries of primitive carbon tele-
phone transmitters along the
front of the stage of the Paris Op-
era House. The transmitters were
grouped in pairs the same dis-
tance apart as human ears, with
several pairs across. The leftmost
of each pair were mixed together
and fed to one telephone line,
which listeners in their homes
directed to their left ears using
the ordinary phone earpiece. The
rightmost of each pair were also
mixed together and fed to a sec-
ond phone line, which each lis-
tener had to have installed in
their home. The result was that
as opera singers moved about the
stage, home listeners could "see"
their movement while hearing
the music with much greater
fidelity than a single phone line
could possibly provide. The origi-
nal patent says "This double lis-
tening to sound. .. produces the
same effects on the ear that the
stereoscope produces on t he
eye." It's fortunate that a wide fre-
quency response is not the most
important parameter for con-
veying the binaural effect; phase
accuracy and correct balance be-
tween the two channels is more
important.
A similar project was carried
out with an improved version of
the Ader experiment in Berlin in
1925. During t hat same year
more than one radio station in
the U.S. did experimental bin-
aural broadcasts using two dif-
ferent frequencies. Listeners
needed two crystal sets, each
feeding a separate earphone. The
mikes in the studio were kept
about seven inches apart, and
therefore listeners with only one
(I)
radio still heard a normal signal.
g
During the last 40 years there
has been sporadic interest in bin-
0
2 aural reproduction around the
2
world, centered primarily in Eu-
-1 rope and Japan. In 1970, Stereo
5' Review issued a binaural demon-
0
o stration LP of music and sound
$ effects using the "Blue Max"
and stereophonic (d).
dummy head, which was hand-
made for the project. Music ex-
cerpts from it are still currently
available on a pre-recorded cas-
sette. The Sennheiser micro-
phonelheadphone people in Ger-
many issued a series of 45-rpm
binaural demo recordings (long
out of print now), which were
well done and designed to pro-
mote their open-air phones and
special binaural mike system. Di-
agrams of the placement of musi-
cians and sounds around the
listener aided in evaluating how
precise the effects were.
One demo in particular was in-
teresting. It featured a woman ar-
riving on a train and being met
by a man at the station. Placed
among all the sound effects of the
train, people, and station en-
vironment were the voices of the
woman, speaking English, and
the man, greeting her in Ger-
man. Eventually they meet in
front of the listener. All the while
you can easily focus on either the
German or the English and un-
derstand perfectly without se-
rious distraction from the other
voice, as would occur with stereo
sound and certainly with mono
reproduction.
This very functional use of bin-
aural is currently being applied
to military aircraft communica-
tion by researchers at the NASA-
Ames Research Center. They use
a powerful computer known as
t he Convolvotron to process
mono speech and signals from
several sources, such as control
con tin ued on page 84
is just as likely to happen when Frankly, experience has shown that of why some records sounded so bad.
monitors with droopy high-frequency there's no reason to trust the hearing But we need look back no. further
response are used. Many years ago, I acuity or good judgement of those than .the hundreds of shrill-sounding
remember reading about a technical responsible for the sound of our mu- CD's released when the format was
survey of the speakers used in sound sical software. True, things are a lot new to realize that incompetent engi-
studios and how appalling their fre- better than in the early 196OVs, when I neering is not a rare phenomena in
quency performance was. first started looking into the problem the record business. R-IE
towers and other aircraft, and as-
sign them places in a 360-degree
sphere so t hat pilots wearing
stereo headphones can "steer"
their binaural hearing t o t he
voice they need to hear and ig-
nore the rest. A demo tape mixed
those artificial binaural voices
into a loud background of heli-
copter noise and contrasted the
very intelligible result with the
same signals in both mono and
stereo.
Seminars Ed a y ............. m t w a l
Meeting Weekend . . . . . . . . . ~ o r k s l i o p
....................
Papers Dai l y Teclmical Paper s
Dai l y .................... Equi pment Exhibits
Exhibits
.................... Tutorials Dai l y Social Event s
Who Will Attend?
* Presidents, CEOs Owners a Cameramen Chief En ' eers * TV Maintenance Supenisors
* Sales & Marketing VPs * Designers * A/" ~ e g c i a n s * Grayluc Artists
* Consultants .Producers Systems Engineers * Tape Editors
Directors of Operatiolis Marketing Managers Teclmical Supenrisors * Video Editors
Students Publishers Production Managers * Digital Effects Designers
. ~o o e o . . o e . . o . o ~. . . . o ~ ~ ~ ~ ~ o e ~ o ~ ~ o o ~ ~ o ~ e o o
Much research i s currently
being done on the human hear-
ing mechanism and the broad
field known as psychoacoustics.
Product approaches being pro-
moted, such as Hughes SRS and
Q Sound, are an outgrowth of
t hat work. Those systems at -
tempt to offer a binaural-like sur-
round localization with loud-
speaker s r at her t h a n s t er eo
headphones, and only a pair of
them at that. With proper source
material, mixing, and proper
continued on page 86
0
For More Information, Mail To:
0
6 :
Society of Motion Picture and
Television Engineers
D I .
595 W. Hartsdale Ave.,
I-
.
White Plains, NY 10607
. (914)761-l l W*hx#914-761-3115
Ltl
PLEASE RUSH THE FOLLOWING INFORMATION:
Exhibitor Full Conference Registration
$ 0 Exhibit Only Tutorial Seminar Workshop
1: E hl 0
lami Title
0
0
Co. e
.
Street
City 0
State z i p Country
~ ~ 1 , Fax 0
0
[email protected]... O ~ b O @ @ ~ O O O ~ ~ ~ O O ~ O ~
84
dard oscilloscope is the vertical cir-
cuitry. Although there are special
considerations, this is really nothing
more than an amplifier designed to
have a response that's as flat as pos-
sible over the rated bandwidth of the
scope. Remember that we're using
the output of the amp to display
changes in the input signal, and we
don't want the amplifier to add its
own two cents to either the shape or
level of the input signal. The point of
using a scope, after all, is to display
unknowns in a signal, not add t o
them.
Just as the horizontal section
should have several accurate sweep
speeds, the vertical section should
have several accurate settings for the
gain. When you look at the display
matrix, going one element to the right
should represent a definite amount of
time and going up one element
should represent a definite amount of
voltage increase. Even if you've never
used a scope, you've probably heard
people talking about the number of
volts per division when they're refer-
ring to scope measurements. In the
scope we're going to build, the equiv-
alent would be to refer to volts per
display element.
These three sections-horizontal,
vertical, and display-together form
the basis of every oscilloscope there
is. Before we can even think about
including some of the features found
on commercial scopes-or even
clearly understand what they do-we
have to get the basics out of the way.
The best way to handle this is to de-
sign a basic scope and then, once we
have that stuff under control, we'll be
,in able to think about things like trig-
gered sweep and other bells and
whi st l es f ound on commer ci al
scopes.
By this time you should know that
the first step in any design job is to
draw up a list of design criteria to
formalize the project you have in
mind. In this case the list isn't too
long because the scope will be pretty
simple-at least in the beginning.
Here's our list:
1. The scope will have a maximum
v, bandwidth of 1 MHz.
2. There will be eight selectable
sweep speeds.
3. There will be eight selectable gain
4 ievels.
4. There will be avariable gain control.
0
5. The display will be in a twenty-by-
$ twenty matrix.
86
You can change any of the criteria
you want but, for the moment, it's a
good idea to leave them all as they
are. Once we get into the specifics of
the design, you'll find it relatively easy
to modify some of the features to
adapt to any particular requirements
you might have.
Before we start the actual design,
we have to talk a bit about the display.
Elementary arithmetic tells you that a
twenty-by-twenty matrix calls for four
hundred LED's and, even though you
can get LED's in bulk quantities from
mail-order houses at extremely low
prices, you still have to do a lot of
wiring to get them set together in the
kind of matrix we need. Let's face it,
it's a real pain in the neck to wire four
hundred LED's.
When we get the scope designed,
we'll investigate some alternatives to
using LED's for the display ele-
ments-LCD screens are a perfect
choice. I've seen pre-made LED ma-
trix-displays that come in various
sizes and I'm currently going through
my parts books and mail-order house
catalogs to see what's available and
who has them for sale at reasonable
prices. If any of you know where
these can be gotten, drop me a line
and I'II put it, along with appropriate
thanks, in the magazine.
I have a working version of the
scope on my bench at the moment
and I used four hundred LED's wired
into a twenty-by-twenty matrix. It took
a bit of time to get it wired but, from
personal experience, I can tell you
that it's not too bad and certainly not
the worst thing I've ever had to do. It
was, however, pretty high on my list of
unpleasant experiences.
As we develop the circuitry for the
scope, I'II base the display on the
same sort of LED matrix I wired up on
my bench but, between all of us, we
should be able to come up with a
more attractive alternative that still
uses LED's. It's a mechanical prob-
lem, not an electronic one.
Once everything is done and we
have the circuit working, we'll take a
look at LCD panels. These have be-
come readily available and you can
find them at reasonable prices. We
won't be doing this right at the begin-
ning because the circuitry needed to
drive them and the memory needed
to hold the display is a separate topic
in itself. First things first.
Next time we'll move into hardware
design. R-E
pl acement of the listener in t he
"sweet spot," those approaches
can be qui t e convincing. When
opt i mum cr i t er i a are n o t met ,
they can also sound worse t han
poor mono, and even at t hei r
best, they don' t equal a good t r ue
bi naur al recordi ng o n good head-
phones.
There i s one hi ghl y successful
speaker approach t hat c an be
used wi t h any bi naur al record-
ing. I t i s t he Bi naural Panorama
ci r cui t i ncl uded wi t h t he nor mal
ambience, reverb, and Dol by Sur-
r o u n d features of t he Lexi con
CP-1 and CP-3 Di gi t al Audi o En-
vi r onment Processors. Cor r ec-
t i on of t he "trans-aural" signals
of t he l eft speaker sounds reach-
ing t he right ear and t he right
speaker sounds reachi ng t he l ef t
ear are at t he heart of t hi s speak-
er system. A n adj ustabl e del ay
l i ne sends t he properly correlated
cancellation signals t o each ear,
somet hi ng l i ke Carver' s Soni c
Hol ography or Polk's SDS, but i s
more sophisticated and freer of
t he "phaseyness" o f t hose ap-
proaches. One does have t o si t in
qui t e a n exact sweet spot. Al -
t hough it works well wi t h onl y a
pai r of speakers, t he addi t i on of
anot her mat chi ng p a i r in t he
rear, bei ng fed a simple L - R sig-
nal improves t he effect even fur-
ther. Wi t h practice (some of t hi s
i s a l earned response too) t he l i s-
t ener c a n even cl ear l y i mage
sounds t o t he rear and far sides,
as wel l as vertically.
The cr ux of the mat t er i s t hat
j ust as some people have troubl e