Here's how Hughes' Sound Retrieval System (SRS) works
Bruce McDiffett - no good deed goes unpunished
System, a new system for expanding the soundfield in stereo playback. I
was talking to the people at Hughes last week, and they gave me the
following reprint about SRS.
Transcribed without permission from Radio-Electronics Magazine,
September 1989 issue. Copyright Gernsback Publications, Inc., 1989.
Beyond Stereo : The Sound Retrieval System adds a new dimension to audio
reproduction.
By Len Feldman
Stereo sound on broadcast television has been available for about five
years, and nearly half of the TV sets sold last year incorporated MTS
(Multichannel Television Sound) decoders. They enable viewers to hear
stereo sound tracks from programs that are transmitted locally in
stereo, as well as a SAP, or Secondary Audio Program channel, which can
be a second-language translation of a movie sound track.
However, for all the success that stereo TV has enjoyed, there are still
two major problems with the way stereo TV sound is reproduced in the
home. If you depend on your TV's two built-in speakers to provide the
stereo effect, you are likely to be disappointed. That's because the
two speakers are usually spaced too close to each other to provide a
pleasing stereo "sound stage."
If your TV monitor/receiver has jacks for external speakers, you can use
a pair of separate speakers, or even your home-stereo speakers. The
stereo effect will be best if they are separated by eight feet or more,
but then a new problem arises: When actors on the TV screen speak, if
you are even slightly off center between the two speakers, you will get
the impression that sounds are coming from the nearest loudspeaker
instead of from the actor who is speaking.
A Psychoacoustic Solution
More than two years ago, Hughes Aircraft, a division of the General
Motors Corporation, developed what originally started out as a car-
stereo enhancement system. It is called the Sound Retrieval System, or
SRS for short. Hughes demonstrated the system by placing two small
speakers so close together that their side panels practically touched.
As you might expect, music played over the speakers sounded "cramped."
because it lacked proper stereo imaging and spread. By merely pressing
a button, the two little speakers could be made to *sound* as though
they had been moved apart to the corners of the room.
Two more knobs allowed the music to be adjusted so that you could walk
all around the room without altering the tremendous spread of sound and
sense of ambience. The demonstration was so dramatic and so effective,
that people couldn't help but look for additional hidden speakers. Of
course, there were none.
A simialr demonstration, this time in a car, revealed that same spread
of sound when the "magic" button was pressed. Instead of the sound
being confined to the narrow width of the automotive interior, it seemed
to extend well outside the car. And while Hughes' primary interest was
in enhancing the stereo effect in car-stereo systems, it was clear that
stereo TV, which was then just gaining a foothold, could also benefit
from this remarkable enhancement. More than a year and a half later,
Sony Corporation, having licensed the SRS technology from Hughes, has
incorporated it into nearly a dozen new stereo TV models, and the sound
is truly incredible. (Editor's note: For those of you who don't
believe that this new system sounds as good as we say it does, we
encourage you to prove it to yourself by going to your local Sony dealer
and listening to one of their new SRS-equipped televisions.)
Theory and operation of SRS
SRS is based upon several psychoacoustic principles that have been
written about in obscure scholarly papers over the years, but have
never been put to practical use until now. SRS technology is so
valuable that Hughes has been granted one comprehensive patent
containing no fewer than 159 separate claims. Two additional patents
involving further improvements and additional claims may well have been
granted by the time you read this article. Here is how SRS works.
Much of what happens when SRS is working involves psychoacoutics. As
Arnold Klayman, the inventor of the system explains it, humans perceive
the direction from which sounds come by at least three different means.
We detect the relative phase of sounds in the case of low frequencies
(between about 20 Hz and 200 Hz.) For mid-range sounds (300 Hz to 4000
Hz), we detect the relative intensity. That means sounds coming from
one side sound louder to the nearest ear, and softer to the other ear.
For higher frequency sounds - those having a fast rise time - we judge
direction by relative time of arrival. Those sounds reach the closest
ear sooner than they reach the farther one.
There is, however, a fourth factor that governs the way in which we
judge where sounds originate, which, up to now, has been ignored in
stereo-reproducing systems. That factor has to do with the way our
hearing system's frequency response varies.
The outer ear, known as the pinna, has an effect on the spectrum of
sound reaching the eardrum, while the concha (the section that leads to
the ear canal) has an effect on the frequency at which the ear canal is
resonant. Together, those two parts of the ear control the spectral
shape (frequency response) of the sounds reaching the eardrum. In other
words, the system functions as sort of a multiple filter, emphasizing
some frequencies, attenuating others, and letting some through without
any change at all. The ear's frequency response changes with both
azimuth and elevation, and together with our binaural (two ear)
capabilities, they help us determine whether a sound is coming from
above, below, the left, the right, ahead, or behind.
As Klayman went on to explain, "Microphones used in making recordings
don't behave like human ears. Omnidirectional microphones have a flat
frequency response for sounds coming from all directions. Cardioid, or
directional microphones have a flat response for sounds coming from the
sides and from the front, but are "dead" to rear sounds. So, during
playback, if sounds that originally came from the side are reproduced by
speakers located "up front," those sounds are heard with an incorrect
spectral response. The result is a spatial distortion of the sound
field, and we are prevented from hearing the proper spatial cues of what
was originally performed.
The SRS technique helps to correct those problems by processing the
audio signals so that the spatial cues are restored. SRS first combines
or adds together the left and right channels to create a sum signal
(L+R). It then subtracts each one from the other to create to
difference signals; (L-R) and (R-L).
The signals are then subjected to various forms of processing and
equalization. Ambience and spatial characteristics are derived from the
processed difference signals. Dialogue, vocalist, and soloist sounds
are derived from the processed sum signal. Once the complex and dynamic
processing sequence has taken place, the signals are revised and
reconstructed into new (L+R) and (L-R) signals, which are matrixed back
together in the same fashion used in stereo FM and stereo TV. That is,
using simple algebra, the new (L+R) is added to the new (L-R), yielding
a new L signal (actually 2L, but the 2 is simply an amplitude
coefficient which can be ignored). The new (R-L) signal is added to the
new (L+R) signal to form a new R signal.
In a stereophonic signal, sounds coming from "in front" of you produce
equal-amplitude sounds in the left and right channels, and are therefore
present in the sum signal (L+R). Ambient, reflected, and side signals
produce a complex sound field, and are present primarily in the
difference signals (L-R and R-L).
The Hughes SRS circuitry processes the difference signals to bring back
the missing spatial cues and directional information. The difference
signals are then dynamically increased in amplitude in order to increase
apparent image width. However, since the ear has increased sensitivity
to mid-range frequencies, selective de-emphasis of the difference
signals is necessary to produce a realistically wider stereo image
without introducing annoying image shifts.
The selective emphasis of certain frequencies in the difference signal
accomplishes several things. For the quieter components, it further
enhances the stereo image by restoring the ambience of the live
performance, which is normally masked out by the louder, direct sounds.
It also provides a much wider listening area, as you can walk about the
room and still retain a sense of direction of all the musical
instruments - you no longer have to sit midway between the two
loudspeakers!
There's a lot more going on inside the SRS circuit than what we've
briefly described. There are many control circuits that detect the
content of the music, and dynamically adjust both the level and the
spectral content of the sum and difference signals.
The selective boosting of the difference signal is automatically
adjusted while audio signals are applied so that the perceived stereo
effect is relatively consistent. Without such automatic adjustment, the
amount of enhancement provided would have to me manually adjusted for
different program material. For example, in order to avoid excessive
boosting of artificial reverberation, which is sometimes added to stereo
recordings, the enhancement technique of SRS deemphasizes the frequency
range in ehich excessive reverberation is most likely to occur. That
area is then reinforced by appropriate injection or addition of the sum
signal. The perceived effect is that the amount of artificial
reverberation does not change appreciably when the SRS is turned on and
off.
In the SRS-equipped Sony TV sets, there is only an SRS on/off switch.
That's because the speakers are a known distance apart, and because
stereo TV brodcasts are handled in a predictable manner.
The license acquired by Sony for SRS is non-exclusive, so don't be
surprised if you see other manufacturers offering SRS technology in
their TV sets and audio components in the not too distant future.
[end of article. I didn't attempt to transcribe the figures.]
The people at Hughes mentioned that the manufacturing cost for the
original (analog, 90 dB SNR) versions of the circuit was around $40.
The costs of the latest digital version (which only has a 60dB SNR) has
dropped to about $1 - yep, about a buck. How much more are retailers
charging for SRS sets? :-> :->
Hope this was informative.
--
-Bruce
Anyone notice you haven't seen much of Dan Quayle since
Bush threatened to send the reserves to the Middle East?