Audiophile listening room helps
to define classroom acoustics
Audiophiles and Recording Engineers Generated Advances
in Small Room Intelligibility Classroom intelligibility is a specialized
application in the field of general intelligibility. The subject
of general intelligibility was studied by Art Noxon during the mid
1980's and a number of technical papers were developed and published.
The financing and focus of this work in intelligibility was supported
by "golden ear" audiophiles in their quest to develop
and own high performance "listening rooms". At that time
no one imagined the work accomplished in developing and defining
high end audio playback rooms would be used some 20 years later
to improve the listening experience of students in the classroom.
The search for acoustic perfection in high performance
small rooms lasted one decade, starting in the early 1980's and
coming to an abrupt halt in 1990. This type of investigation was
funded by profits in the flourishing hifi industry. A series of
three economic setbacks essentially finished off the hifi industry
and funding for the research. The first was a national recession
in 1990 and the cutback in consumer discretionary spending . By
1992 home computers were introduced and hifi enthusiasts, being
among the most sophisticated of electronic equipment buyers were
first in line to buy home computers, not more audio equipment. Next
came the DVD and Home Theater and any funds set aside for hifi improvements
became reallocated for home theaters. Home theater, being a highly
visual experience and fairly popular does not yet demand high level
acoustic performance. However, the demand and interest for intelligible
small rooms has resurfaced today, in the classroom.
Intelligibility Became the High Point in Acoustic
Treatment of HiFi Listening Rooms
To become familiar with intelligibility it is very
instructive to revisit those old papers on small room intelligibility
developed some 15 years ago within the hifi industry. The first
paper to study is called "ARTICULATION: Prerequisite to performance".
It was written in 1988 and delivered to the Audio Engineering Society
convention in June of 89 in New York. This paper represents the
overview of the science and applied art of small room intelligibility.
It is an excellent place to start because it defines intelligibility
starting with the most simple signal, the Morse code, and moves
through the range of speech intelligibility and on to the topic
of music intelligibility. It is a pragmatic overview and connects
the dots that are otherwise found scattered among various papers
Further study on intelligibility is available in
another paper. Here the more practical side of intelligibility is
being put to work in developing performance evaluation specification
to be met by audio systems that play in small room acoustics. We
have "Articulation and the Small Room" written by Art
Noxon and presented in November 1988 at the AES Convention in Los
Angeles. Here, the basic terms used in intelligibility are defined:
Signal to noise ratio, modulation index, modulation transfer function,
transmission index speech transmission index and octave masking.
Then they are re expressed in terms of an acoustical testing method
called the MATT, Musical Articulation Test Tones, a gated tone burst
sequence. This test is used in numerous field applications to further
explain what does and what does not work in the quest for better
intelligibility when applied to small room acoustics.
The third paper was actually an appendix to the 89
paper. However, it is sufficiently separate of a topic that it is
now republished under it's own heading. It is a field guide to using
gated tone bursts as a diagnostic tool in the evaluation of room
acoustics. It is intended to provide the field experimenter with
the actual equipment list and settings used to produce modulation
transfer functions. This short paper is called "Diagnostic
MTF Testing Procedure" and is should be read only after both
other papers have been read.
The basic premise for this series of works in intelligibility
is that when working with small room acoustics, room resonance becomes
a very significant problem. Room modes and their Q or damping factor
is a major problem. Acoustic testing of the listening room had been
restricted to balloon bursts, slow sine wave sweeps, pink or white
noise spectrums and octave noise spectrum levels and octave band
RT-60 measurements. But all of these tests missed performance indicators
that directly applied to and were valued by listeners in rooms.
The tonal type of modulation transfer function however directly
addressed and accounted for differences in room acoustic treatments
and their perceived value to the listener. It became apparent that
steady state measurements and broad band transient measurements
were not sufficient to account for intelligibility in small room
Now, with opportunities to improve classroom acoustics
on the horizon, the fruits of years of study and experimentation
in small room acoustics, particularly small room intelligibility
seems to have found a new home. It is fittingly appropriate that
the pioneering work in high performance listening rooms done by
some of the best listeners in the world has now found application
in developing classroom acoustics for those whose hearing capabilities
have been compromised.
There is more to the Signal to Noise ratio than
Our of these studies only certain aspects of listening
room acoustics have been studied. These aspects do not embody "all"
there is to know about quality listening. They merely address one
aspect of evaluating room acoustics, the measuring of the signal
to noise ratio on a tonal basis rather than a 2 octave wide basis
as is ordinarily done. The audio playback in small rooms experience
of the '80's contributed to the recognition of tonal MTF analysis
in determining intelligibility for small rooms. But more, the audiophile
and the recording engineer both work in small rooms and they have
developed a sensitivity and tradition with respect to not only sound
absorption but also to sound reflection.
There are 2 parts to the "signal to noise"
view of acoustics. One is the "noise" part and most of
the effort by all involved in classroom acoustics is focussed on
absorbing or otherwise eliminating the noise. The other part of
the "signal to noise" viewpoint is the "signal"
part. This is a much less addressed factor. The signal of a sound
begins with the direct sound, the one traveling the direct path
between the sound source and the ear of the listener. But it also
includes all early reflections. The contribution of reflections
to listening, how strong they are, where they come from and when
they arrive all make up a second quality of sound. The brightness
of the sound, it's presence, is a consequence of how many reflections
accompany the direct signal. The additive effect of early reflections
to the S/N ration is too often not even considered.
The Recording Industry also Contributes to Intelligibility
The last paper presented here does address the "presence
effect" of sound signals. It is called "Sound Fusion and
the Acoustic Presence Effect" and was presented in 1990 in
Los Angeles at the AES Convention. This work is born out of the
efforts to improve microphone technique for recording engineers.
Here we see that both the direct and the early reflections are important
contributors to the strength of the sound "signal" and
substantial experience with this type of sound is presented. Curiously,
at the end of this paper is found a section that has nothing to
do with the recording industry, it has to do with an experiment
involving learning disabled children at San Diego University. The
value of the Quick Sound Field effect was recognized early on to
also apply to those who are learning disabled.
RT-60 is "Necessary but not Sufficient"
for Good Intelligibility
The ADA has specified that the noise level in our
classrooms should be quiet and the reverb times in the 500, 1k and
2k octave bands be short. The net result is that a substantial quantity
of acoustic material has to be added to any accessible classroom,
typically 200 square feet of material. There are a lot of factors
that restrict what that material might be, and where it can be placed.
We have black boards and clocks, windows and doors, light switches,
air vents, maps and many more things that were there first. Acoustics
is relegated to whatever space is left over. The question remains
as to where in this remaining available space should the acoustics
We have found that it is never satisfactory to simply
own a bundle of acoustical material. Even if you were to place the
bundle of acoustic material in the right room it still isn't good
enough. If you spread out the materials even in the room one effect
will be noticed. Rearrange the materials and another effect is noticed.
How the room sounds depends on where the materials are placed. Reverb
analysis leads to introducing some 200 square feet of acoustic material
into a classroom. The RT-60 requirement may be satisfied with any
number of material distribution configurations. But which distribution
best satisfies the needs of the students? That is the point to which
we direct our attention.
More than Reverb Control, Intelligibility means
Voicing the Room
We have found that in high performance, high intelligibility
spaces, it is critical that the materials introduced into a room
to satisfy the RT-60 requirement also have to be placed is specific
areas, patterns and configurations in order to develop the inner
detail of the listening experience. Absorption manages the RT-60
and knocks down the noise. Reflections enhance the signal and improves
the overall S/N ratio. Therein lies the difference between a room
that meets spec and a room that really sounds good.
Acoustics is sometimes referred to as being half
science and half art. The science of acoustics specifies the material
needed to meet the RT-60 spec and the art of acoustics determines
where that material is to be located. And that's what we call "voicing