US6430293

Extracted document text. Layout, formulas and symbols may not survive extraction; consult the preserved PDF for the original presentation.

Text extraction covers 7 pages, including 7 transcribed by optical character recognition.

Page 1

«2) United States Patent

Finsterle

(10) Patent No.:
(45) Date of Patent:

US 6,430,293 BL
Aug. 6, 2002

(54) RECORDING AND PLAY-BACK
TWO-CHANNEL SYSTEM FOR
PROVIDING A HOLOPHONIC
REPRODUCTION OF SOUNDS

(76) Inventor: Luca Gubert Finsterle, Via Pontida, 6,

Cernusco sul Naviglio Milan (IT)

(*) Notice: — Subject to any disclaimer, the term of this

patent is extended or adjusted under 35

US.C. 154(b) by 0 days.

(21) Appl. No.: 09/242,089
(22) PCT Filed: Aug. 5, 1997
(86) PCT No.: PCT/IT97/00204

§ 371 (c)(1),
2), (4) Date: Feb. 8, 1999

(87) PCT Pub. No.:. WO98/07299
PCT Pub. Date: Feb. 19, 1998

(30) Foreign Application Priority Data
Aug. 13,1996 UT) cesseeccsssssesecsseesssesssesssssenesess MIO6AL752
(51) Int. Cl’. .. HOAR 5/00

381/17; 381/26; 381/19
. 381/1, 17, 18,

(52) US. Cl
(58) Field of Search .

381/19, 303, 309, 310, 26, 74, 56, 61, 91,
122

(56) References Cited
U.S. PATENT DOCUMENTS,

3,892,917 A *
3,940,560 A *

7F/A9TS Sotome ..
2/1976 Condamines

4,060,696 A * 11/1977 Iwahara et al. 381/19
4,072,821 A * 2/1978 Bauer ..
4,119,798 A * 10/1978 Iwahara

*

4,466,117 A 8/1984 Gorike

O

10 44
1273 13

4,868,878 A * 9/1989
5,109,415 A * 4/1992
5,119,420 A * 6/1992
5,173,944 A * 12/1992 Begault ... 381/17
5,208,860 A * 5/1993 Lowe et al. 381/17
5,226,087 A * 7/1993 Onoetal. . 381/26
5,412,732 A * 5/1995 Kanishi et al. 381/18
5,850,455 A * 12/1998 Arnold et al 381/L

FOREIGN PATENT DOCUMENTS
JP 52-2402 * 1/1977

* cited by examiner

Primary Examiner—Xu Mei
(74) Attorney, Agent, or Firm—Hedman & Costigan, P.C.

(57) ABSTRACT

A two-channel sound recording and play-back system for
providing a holophonic reproduction or play-back of sounds
is herein disclosed. The two-channel recording system com-
prises cables and electronic apparatus for providing a ste-
reophonic type of recording, where a plurality of micro-
phones are arranged in an adjoining mutually angled
relationship so as to provide, during the recording operation,
a stereophonic effect due to phase differences at each fre-
quency and an intensity difference at least for middle-high
acoustic frequencies. The two-channel sound play-back sys-
tem comprises, in addition to a source and a two-channel
amplifying system, a plurality of diffusers arranged to cor-
ners of a parallelogram, a listening point being preferably
arranged at the center of said parallelogram, the front or rear
diffusers receiving the recorded signal without
modifications, whereas, respectively, the rear or front dif-
fusers receive a signal the sound pressure of which is
adjusted, by a perspective corrector, from -15 dB to +9 dB,
as modified in the frequency response thereof and with a
delayed emission, so as to broaden the sound range up to
360°, thereby providing a sound holophonic play-back.

5 Claims, 2 Drawing Sheets

Page 2

US 6,430,293 BL

U.S. Patent

10° Si1
127-113

HOO oOw

FIG.1

Page 3

U.S. Patent Aug. 6, 2002 Sheet 2 of 2 US 6,430,293 B1

FIG, 3

Page 4

US 6,430,293 B1

1

RECORDING AND PLAY-BACK
TWO-CHANNEL SYSTEM FOR
PROVIDING A HOLOPHONIC
REPRODUCTION OF SOUNDS

DESCRIPTION

The present invention relates to a recording and play-back
two-=channel system for providing a holophonic reproduc-
tion or playback of sounds.

A holophonic reproduction of sounds involves the capa-
bility of detecting, during a sound reproduction or playback
operation, the space locations of the sound sources,
comprising, for example, musical instruments, in a manner
similar to that which would have been obtained if the
listening person would have been present at the recording
place, for example a theater, and this, tendentially, for any
positions of the sound sources with respect to the micro-
phones.

By a holophonic type of reproduction, accordingly, it
would be possible to perceive in a home reproduction of a
symphonic concert not only the locations of orchestral
instruments in front of a listener, but also echoes and
reverberations coming from the sides and rear of the listener,
as well as related applause.

Thus, the obtained effect substantially corresponds to the
effective original sound or musical event.

Several solutions are already known for providing a
playback more or less fitted to the sound scene, i.e. the
bi/three-dimensional representation of a physical space
built-in only by sound information.

All the two-channel stereo recordings, either of a
multiple-microphone or a two-microphone type, are suitable
to playback, by two diffusers, as in a home stereo system, a
more or less suitable reconstruction of an original musical
event.

In particular, by a stereophonic recording, each stereo
system would be able of recovering a sound scene, which, at
most, evolves through the space included between the two
diffusers in at least two dimensions, i.e. the right-left dimen-
sion between the two diffusers, and the near-far dimension
starting from an imaginary line connecting the two diffusers,
and progressively away from a listening point.

The sole condition to be met by a listener in order to
obtain a satisfactory stereophonic effect is to be located
ideally at the apex of an equilater triangle, the two other apex
points thereof are occupied by the two diffusers.

A bi-three-dimensional representation of a physic space
built-in only based on a sound information obtained from a
stereophonic system, can evolve through a generally trap-
ezoidal region, where the small base apex points are closer
to the listener and are represented by the two diffusers of the
system, whereas the apex points of the major base are
represented by two points defined by a sound information
contents of the recording, such as phase delays and
reverberations, sound reflections caused by a home listening
Toom as well as the detecting properties of the diffusers or
sound emitting sources.

The above disclosed region is defined as a “sound or
acoustic range”. By exploiting the high sensitivity of an
earing system to phase and frequency response variations, it
would be however possible, by a suitable electronic
apparatus, to broaden, during the recording operation, the
sound range, as reproduced by a stereophonic system, or the
area in which it would be possible to reconstruct the virtual
sound scene, ie. that area in which it would be possible to

10

15

20

30

40

50

66

2

form and perceive again virtual sound subjects, since the
latter would be present in said recording.

This technique, which is used, for example, in the
so-called Q discs, will always held the positions of the
diffusers and of the listening point consistant, thereby theo-
retically providing a sound field about a the listener.

In order to further broaden the sound range, other two-
channel recording and playing-back systems have been
made: the so-called quadriphony, ie. four independent
channels, two at the front and two at the rear; the “Dolby-
Surround” and all the improvement variations thereof, with
four playback channels, i.e. three front channels, central
right and left channels, as well as the rear monochannel
splitted on two diffusers, and yet other systems including
more than four recording and playback channels such as the
“Dolby AC3”.

These prior systems, which provide to use, during the
recording operation, multiple-microphone techniques and
electronic signal processing methods, would allow to
broaden up to 360° the sound field with at least three
channels during the recording and for diffusers during the
playback.

Thus, it would be possible to locate sound objects outside
the most marked and clean areas of the sound field using
only channels with phase rotation and two diffusers, like the
above disclosed technique, and as it occurs in the so-called
Q-dises.

While the above mentioned prior recording and playback
systems solve the mentioned technical problem of a holo-
phonic sound reproduction, they have, however, the disad-
vantages of an excessively reduced sound field area, actually
preventing a true and proper holophonic reproduction.

The Q-disc system, on the other hand, provides a rather
unnatural effect due to an insufficient capability of locating
the sound subjects the space arrangement of which must be
considered with respect to the sound field area.

This derives from several reasons, the main of which is.
that said diffusers are anyhow perceived as a sound emitting
source by a listener, thereby providing a perspective distor-
tion effect.

Asound recorded by a conventional stereophonic method,
the original features of which would be such as to always
bring that sound to be perceived outside the sound field area,
would be anyhow tendentially returned to said area, since
the reproduced sound emission occurs at the points occupied
by the diffusers. In order to broaden the sound field area, it
would be necessary in this case to offset the mentioned
prospective distortion by suitable phase rotations and fre-
quency response modifications which, since would not cor-
respond to those naturally perceived for a sound just coming
from that space point and, accordingly, would be unnatural.
Moreover, during a recording operation, it would be neces-
sary to provide an electronic apparatus for modifying the
signal from the microphones and electronic sources, for
example an electric guitar which, while perfect, would
however introduce distortions in the signals, thereby wors-
ening the reproduced sound quality.

The multiple channel recording and playback system,
moreover, involves a coding operation, during the recording
operation, in order to include more than two channels in the
single two-channel arrangement of the stereophonic
software, in addition to a decoding step during the playback
operation. While the compatibility with a stereophonic sys-
tem is satisfactory, the sound field will fall into the sound
field area disclosed for the two channel systems, admitted
that the playback system is of a stereophonic type.

Page 5

US 6,430,293 B1

3

A further disadvantage is the requirement of coding and
decoding recorded signals by an electronic apparatus, which
will further distort an original signal.

Moreover, each system including more than two record-
ing channels can not contain, by definition, phase informa-
tion and frequency response variations like those which
would allow a human ear system to reconstruct a sound
scene identical to that which would be perceived as present
in a original sound event.

Furthermore each effort to broaden the sound field would
involve an use of auxiliary electronic apparatus such as
phase processors and coders, at least in a recording
operation, which apparatus, in addition to worsening the
signal quality, would have a very high cost for the required
software: discs, CD, and so on.

SUMMARY OF THE INVENTION

Accordingly, the aim of the present invention is to over-
come the above mentioned prior art drawbacks.

In order to achieve the above mentioned aim, the present
invention provides a two-channel sound recording and play-
back system, suitable to provide a holophonic reproduction
of sounds which is constructionally simple, of low cost and
provides an end playback result which is more natural and
realistic than that obtained by recording and playback
sounds by three of more channels, thereby providing a sound
reproduction which is substantially identical to the original
sound event, in particular with respect to the original three-
dimensional reconstruction, i.e. an actually holophonic
reproduction of sounds.

According to the invention, a two-channel sound record-
ing and playback system is herein provided, suitable for
holophonically playing-back sounds, wherein said recording,
system comprises cables and electronic apparatus to provide
a stereophonic recording, and in which are moreover pro-
vided a plurality of microphones arranged in a close and
angled relationship, so as to provide, in a recording
operation, a stereophonic effect from phase differences for
each frequency and from intensity differences for at least
sound frequencies greater than 200 Hz, and wherein the
sound playback system comprises a source and a two-
channel amplifying system, a plurality of diffusers being
moreover arranged at corners of a parallelogram, where the
listening point is arranged at the centre of said
parallelogram, the front or rear diffusers receiving the
recorded signals without modifications, whereas,
respectively, the rear or front diffusers receive a signal
adjusted by a perspective corrector for correcting the sound
pressure from -15 dB to +9 dB, as modified in the frequency
response and delayed in emission so as to broaden the sound
field up to 360°, thereby providing a sound holophonic
playback.

The two-channel recording and playback system for pro-
viding a sound holophonic reproduction according to the
invention is characterized in that it comprises the charac-
terizing features of claims 1 and 5.

The two-channel sound recording and playback system,
suitable for providing a sound holophonic reproduction
according to the invention provides the following advan-
tages.

With respect to a conventional stereophonic playback, the
subject system provides, in a playback operation performed
by conventional stereophonic systems, an increase of the
sound field area, independently from the diffusers.

Moreover, the sound information is not further distorted
since no signal processing or codifying system is provided

15

20

30

66

4

for holding further channels in addition to the two-channels
provided in a conventional stereophonic system.

Moreover, the system according to the invention allows to
clearly perceive sounds coming from the rear part of the
shoulders of a listener, as well as laterally, if said sounds are
present in the original sound event: for example the rear
echoes of a concert room, or applause of persons arranged
on the rear portion of the microphonic system.

A further advantage is that of providing a holophonic
sound recording and playback by simply adding, in a record-
ing studio, the elements characterizing the present invention.

This advantage, moreover, would be mainly valid also for
a playback operation for users already having a conventional
stereophonic system.

Finally, independently from the used recording system,
the following advantages will be obtained during the play-
back operation: an increase of the three-dimensional effect
of virtual sounds present in the sound field area; for
example, by listening from the listening point a person
would have the feeling of touching a violin on the front with
a consequent encompassing effect; an increase of the sound
field which is also increased beyond the diffusers while
remaining in front of the listener; an increase of the depth
feeling of the front area of the sound field and, finally, a
general feeling of a greater playback naturalness and real-
ism

BRIEF DESCRIPTION OF THE DRAWINGS

Further characteristics, advantages and details of the
two-channel sound recording and playback system for pro-
viding a holophonic sound reproduction according to the
present invention will become more apparent hereinafter
from the following detailed disclosure, with reference to the
accompanying drawings, which illustrate, by way of an
indicative, but not limitative, example, a preferred embodi-
ment thereof, and in which:

FIG. 1 schematically illustrates a recording step per-
formed by using the system according to the present inven-
tion;

FIG. 2 illustrates the sound field in which are perceived
virtual sound sources, as obtained by the present invention
by using only two diffusers for the playback; and

FIG. 3 illustrates the sound field in which are perceived
virtual sound sources, by using all the means provided
according to the present invention.

DESCRIPTION OF THE PREFERRED
EMBODIMENTS

With reference to the number references of FIG. 1, the
two-channel sound recording system suitable to provide a
holophonic reproduction of sounds according to the
invention, comprises two pickup microphones 10, 11 having
omnidirectional features or cardioid or with all of the
intermediate cases. If a plurality of microphones are
provided, then they could have directive ipercardioid or club
patterns characteristics.

The microphones 10, 11 are arranged with their mem-
branes so that the distance from a microphone to the other
would be from 15 to 25 cm, preferably from 19 to 21 cm,
whereas the angle of said microphones would preferably be
from 120° to 200°.

During the recording operation, the microphones must be
arranged at such physical locations that a listener arranged
herein would obtain a realistic representation of the spatial
reconstruction acoustically obtained from a musical event,

Page 6

US 6,430,293 B1

5

without the need of constructing this representation by
visual images, the representation being only based on a
sound information.

The above mentioned microphone arrangement must be
such as to provide, during the recording operation, a stereo-
phonic effect deriving from a phase difference at each
frequency and, for intensity differences, at least for frequen-
cies greater than 200 Hz.

The frequency response sum of the two channels must
show a total frequency response within +5 dB, for any
locations on the horizontal plane, with respect to the micro-
phone axes, of the sound emitting source.

The recording system should comprise cables 12 and
electronic apparatus 13 adapted, in cooperation with the
microphones 10, 11, to hold the phase response for each
frequency from 20 Hz and 10000 Hz within +20°.

With reference to the number references of FIGS, 2, 3, the
two-channel sound playback system, suitable to provide a
holophonic sound playback according to the invention,
comprises a sound source 7, an amplifying system 8, of a
two-channel type, and two front diffusers 2, 3 so that the
angle formed by the line joining the listening point 1 and
said front diffusers 2, 3 is of 90° with a tolerance of plus or
minus 30°. The above mentioned diffusers 2 and 3 will be
oriented toward the listening point 1.

In such a configuration of the listening system, the sound
field area will also extend to the rear of the listener, thereby
providing an opening of about 180°, i.e. about three times
that value which would be obtained by conventional stereo-
phonic recording systems.

In order to further increase or broaden the sound field
area, the subject system comprises moreover an electronic
apparatus 4, called “perspective corrector” as well as two
rear diffusers 5, 6 which are symmetrically arranged with
respect to the two front diffusers 2, 3, i.e. with respect to the
listening point 1: the rear diffusers 5, 6, accordingly, are
arranged with respect to the first two diffusers 2, 3 so that the
listening point 1 is arranged on the center of an ideal
rectangle formed by the four diffusers, while considering the
disclosed tolerances affecting the locations of said diffusers.

Preferably, the locations of the middle-high frequency
transducers located in the four diffusers 2, 3, 5, 6 will
correspond to the level or height of the ears of a listener
arranged at the point 1: in other words at 1-1.2 meters from
ground in a case of sitting listener.

The electronic apparatus 4 operates a send to the rear
diffusers 5, 6 a signal similar to that sent by the amplifier
system 8 to the front diffusers 2 and 3, so as to obtain from
the rear diffusers 5, 6 sounds which, with respect to the
dynamic frequency response and distortion as well as phase
response, would be substantially identical to those provided
by the front diffusers 2, 3 but having a sound pressure
adjusted in a range from -15 dB to +9 dB, preferably from
-10 dB to +6 dB.

Moreover, in order to further improve, in a playback-step,
the possibility of properly detecting the locations of the
virtual sound events on the rear of the listener, which,
originally, in the recording place, where sent on the rear of
the microphone pick up system, it would be possible to
attenuate or enhance the middle and/or high frequencies, and
in addition to cut off the low frequencies, or slightly delay
the emission of the rear diffusers 5, 6 with respect to the
front diffusers 2 and 3, i.e. in the disclosed example, only for
the rear diffusers 5, 6 and through the perspective corrector
4.

Preferably, the electronic apparatus 4, or perspective
corrector has the following features: a gain varying from -50

15

20

30

40

66

6

dB to +20 dB; a high-pass filter having a cut off frequency
from o Hz to 5,000 Hz, with a slope equal to or greater than
6 dB/oct, preferably from 6 dB/oct to 18 dB/oct; a variable
middle-high frequency adjuster (from -10 dB to +10 dB)
having a center at 4,000 Hz and an operation field from
2,000 Hz to 10,000 Hz; a variable high frequency adjuster
(from -10 dB to +10 dB) for adjusting high frequencies from
10 KHz and 20 KHz.

Said apparatus comprises moreover a circuit for correct-
ing the frequency response and acoustic phase in order to
make the sounds emitted from the rear diffusers 5, 6 similar
to those emitted by the front diffusers 2, 3 or, if desired, to
modify the phase response of the rear diffusers 5, 6.

Finally, the general phase response of the playback system
according to the present invention must be linear within
+#20° for each frequency from 20 Hz to 2,000 Hz at the
listening point 1.

Thus, the recording and playback system according to the
present invention provides to use the two additional rear
diffusers 5, 6 which, together with the control electronic
apparatus called “perspective corrector 4”, will have a
frequency response phase response, dynamic and distortion
characteristics similar or identical to those of the front
diffusers.

This identity can also be achieved electronically if the rear
diffusers 5 and 6 are not identical to the front diffusers 2, 3,
by an equilization and adjustment of the phase response by
a specifically provided circuit to be added to the disclosed
electronic apparatus for sending a corrected signal to the rear
diffusers 5, 6.

More specifically, said diffusers must be arranged in
mirror-like arrangement with respect to the front diffusers 2,
3, ie. with respect to the listening point 1, with the same
tolerances related to an ideally correct position; if they are
closer, then, by the perspective corrector 4, a virtual position
provided with a congruent delay will be simulated, which
will be, from a distance standpoint, similar to that of the
front diffusers 2, 3 from the listening point.

Thus a 360° reconstruction of the virtual sound scene on
a horizontal plane and partially on a vertical plane will be
obtained, since it will be possible to reduce the perspective
distortion occurring in listening from the front diffusers only.

This perspective distortion reduction is obtained, accord-
ing to the present invention, by sending to the rear diffusers
5 and 6 the same identical signals sent to the front diffusers
2, 3, said signals being however adjusted and/or cut and/or
delayed so as to provide a broadening of the sound field up
to 360°, or less, depending on the listener requirements.

The above mentioned adjustments will depend on the size
and acoustic absorption features of the listening room and on
the listener requirements.

In particular, this effect is basically obtained due to the
cooperation of two specific properties of the recording and
playback system according to the present invention.

Amicrophonic pickup which records sounds at two space
points. i.e. those immediately near the ears of a human head,
thereby holding the signal phase characteristics unchanged,
and providing a stereophonic effect derived from a fre-
quency dependant phase and intensity difference.

Moreover, the rear diffusers 5, 6, which would be vari-
ously attenuated and filtered and properly arranged, will
provide, by adding the signals from the front and rear
channels, a virtual sound emission point not coinciding with
the front diffusers 2, 3, thereby reducing the perspective
distortion and broadening the sound field to 360° about a

Page 7

US 6,430,293 B1

7

listener properly arranged with respect to the diffusers 2, 3,
5, 6 at the listening point 1.

This is obtained during the recording operation due to the
particular locations of the microphones 10, 11 allowing the
four diffusers 2, 3, 4, 6 and prospective corrector 4 to send,
through the ear reproduction apparatus, in an optimum
manner, all of the phase information related to sound
sources, by ideally zeroing any perspective distortions from
the front diffusers 2, 3, by adding an identical and opposite
distortion induced by the rear channels and a consequent
generation of an emission point always arranged on the front
or, at most, on one the sides, and different from the localized
diffusers, since the rear diffusers 5 and 6 will be always
suitably adjusted with respect to the front diffusers 2, 3.

Thus, a two virtual diffuser sound emission will be
provided, said virtual diffusers being devoice of any spatial
locating features: thus the sound will be perfectly holophoni-
cally reproduced.

In this connection it should be pointed out that the zeroing,
of the perspective distortion related to two diffusers only can
also be obtained, in a opposite manner to the above, by
holding unchanged the signal related to the rear diffusers 5
and 6 and by operating, by the perspective corrector 4, on the
front diffusers 2, 3, so as to adjust and delay the signal
transmitted to said front diffusers, thereby providing again
an holophonie playback of sounds.

Finally, it should be apparent that the mentioned sound
playback features can also be mainly obtained by a software
recorded by conventional methods.

It should be moreover pointed out that the configurations
and size of the several elements constituting the subject
two-channel sound recording and playback system suitable
to provide a holophonic reproduction of sounds according to
the invention can vary according to requirements, without
departing from the scope of the invention as above dis-
closed.

What is claimed is:

1. A two-channel recording system providing a holo-
phonic reproduction of sounds comprising cables (12) and
electronic apparatus (13) for providing a stereophonic
recording, and a plurality of microphones (10, 11), charac-
terized in that the membranes of said microphones are
arranged in a close and angled mutual relationship so as to
provide, during a recording operation, a stereophonic effect
derived from phase differences for each frequency and from
intensity differences at least for sound frequencies greater
than 200 Hz, the distance between said membranes being,
from 15 cm to 25 cm, and the angle thereof being from 120°
to 200°, that said microphones have omnidirectional direc-
tive characteristics, cardioid with any intermediate patterns,

20

30

40

8

hypercardioids, or club patterns, that the frequency response
sum of said two-channels shows a total frequency response
within +5 dB for each location on a horizontal plane, with
respect to the axes of said microphones (10, 11) of the sound
emitting source, and that said cable (12), electronics appa-
ratus (13) and microphones (10, 11) have a total phase
response for each frequency from 20 Hz to 20,000 Hz,
within +20°.

2. Atwo-channel playback system providing a holophonic
reproduction of sounds, comprising a sound source (7), a
two-channel amplifying system, and a plurality of front and
rear diffusers (2, 3, 4, 6), characterized in that said front and
rear diffusers are arranged at the apex points of a quadrilater,
the listening point (1) thereof is arranged at the center of said
quadrilater, the angle formed by the lines joining the front
diffusers (2, 3) to the listening point (1) and the rear diffusers
(5, 6) to the listening point (1) varying from 60° to 120°,
each said diffusers (2, 3, 4, 6) being oriented toward said
listening point (1), the front diffusers (2, 3) or rear diffusers
(5, 6) receiving an unmodified recorded signal, whereas,
respectively, the rear diffusers (5, 6) or front diffusers (2, 3)
receive a signal adjusted by at least a perspective corrector
(4) controlling the sound pressure of said signal from -15 dB
to +9 dB, as modified in the frequency response and delayed
in emission so as to broaden the sound field up to 360°,
thereby providing a sound holophonic reproduction.

3. A two-channel playback system according to claim 2,
characterized in that said perspective corrector (4) has a gain
varying from -50 dB to +20 dB and comprises a high-pass
filter with a cut frequency varying from 0 Hz to 5,000 Hz
with a slope equal to or greater than 6 dB/oct, an adjuster
variable from -10 dB to +10 dB for adjusting the middle-
acute frequencies with a center at 4,000 Hz and an operating
range from 2,000 Hz to 10,000 Hz, a further variable
adjuster within the range from -10 dB to +10 dB for
adjusting the acute frequencies from 10 KHz to 20 KHz, a
delay circuit providing a delay from 0 ms to 21 ms and a
correcting circuit for correcting the frequency response and
acoustic phase, suitable to modify the signal sent to the
diffusers coupled to said perspective corrector (4).

4. A two-channel playback system according to claim 2,
characterized in that the total phase response of said play-
back system is linear within +20° and for each frequency
from 20 Hz to 20,000 Hz at said listening point (1).

5. A two-channel playback system according to claim 2,
characterized in that said diffusers (2, 3, 5, 6) comprise
moreover middle-acute frequency transducers arranged at
the level of the ears of a listener positioned at said listening,
point (1).