CA1063035A - Sound reproduction systems - Google Patents

Abstract

ABSTRACT OF THE DISCLOSURE

In a sound reproduction system, the phasiness of the psychoacoustically most important signals is minimised by subtracting from the velocity signal components of said most important signals a directional bias signal comprising a fraction of the pressure signal phase shifted by 90°.

Description

~063035 This invention relates to sound reproduction systems and more particularly to sound reproduction systems which enable the listener to distinguish sounds ~rom sources extending over 360 of azimuth.
Canadian Specification No. g77687 and co-pending Application No. 222146 are concerned with sound reproduction systems which enable the listener to distinguish sounds from sources e~tending over 360 o~ azimuth and which employ only two independent transmission channels. In one o~ these systems, one channel carries so-called omnidirectional signal components which contain sounds from all horizontal directions with equal gain. The other channel carries so-called azimuth or phasor signal componentæ containing sounds with unity gain ~ro~ all horizontal directions but with a phase æhift relative to the correspondine omnidirectional signal component which is related to, and is pre~erably equal to the azimuth angle Or arrival measured ~rom a suitable re~erence direction. In other systems, the signals o~ the t~o channels comprise linear combinations Or the omnidirectional and phasor signals.
The phasor signal P may be resolved into components X and Y Y/ith a phase dirrerence Or 90.
For a 60und at azimuth ~ ~rom the rorYard direction, the localisation is determined by cos t : sin ~ = Re X : Re Y ;~
~here W is the omnidlrectional signal and ~e ~eans the real part Or~l. Thus the ima~inary parts Or ~w and Y/W do not contribute substantially to the sound localization. Instead they csuse the sound siEnals to ha~e an unple~ssnt quality coc~only called "phasiness"

- 2 - *
' ~ . . , , -.. '' ` ' ' ' ' which manirests itselr ln broad lma~es that are hard to localize and ~ound Yery unnatural It ha~ been ~ound that ~or a particular azimuth, the larger t~e ratio o~ the imaglnary part of Y/W to the real part of ~yw, the worse the phaslness for ~ignal~ ~rom that part-icular azimuth.
An omnidirectional signal is a particular one Or a class o~ ~ignals v~hich represent the acoustic pressure signal available at a listening position.
Similarly a phasor signal is a particular one Or a class of signals which represent the acoustlc velocity 6ignals available at the same listening position.
It should be understood that in the present specirication the signal W may be any signal representing said acoustic pressure signal and the signals X and Y ~ -may be any signals representing orthogonal,components of ~aid acoustic velocity signals.
The present invention i8 concerned with minimising the pha6iness o~ the psychoacoustically most important signals. In general, these are the signals ~rom in ~ront Or the listener. However, lr ~ :-at any time, there is a dominant signal ~rom a particular azi~uth, it may be prererred to minimise the phasiness ror this azimuth and to cbange the para~eters o~ the decoding ratrix as the azimuth of *he ~ost important sound alters. The invention i8 also applicable to decodera ror syatems which are subject to phasiness and heve a hiEher number Or --channels than two and to decoders for three-dimensional systems which additionally distinguished between sounds origlnating at dir~eren~ heights and have a third slgnal Z, representing a third orthogonal component o~ the

- 3 ~

acoustic velocity signals, for this purpose.
According to the invention, there is provided a decoder for a sound reproduction system having at least three loudspeakers surrounding a listening area, the decoder comprising input means for receiving at least two input signals comprising pressure signal components and velocity signal components, means for subtracting f`rom the velocity signal component of a chosen directlon a directional bias signal comprising a signal ell the components of which have a + 90 phase relation with respect to the pressure signal components, and output means for producing a respective output signal for each loudspeaker.
This subtraction procedure is hereinafter called "directional biasing". In general the chosen 'direction will be the direction of the dominant or most significant signal. When the chosen direction is the forward direction, the procedure is called "forward biasing".
In the circumstances when all significant sound sources are, or'a dominant sound source is, located at a particular azimuth at any one instant of time, the invention may provide means for determining such particular azimuth from the input signals and applying a bias signal dependent on such azimuth so as to compensate for phasiness of sources located thereat.
me pressure signal components may be omnidirectional signal co~ponents and the velocity signal components may be phasor signal components.
~0 mus, in acoordance with the invention, the signals W, X and~Y used to produce the output 8ignals for a two-channel input signal, in which ~.

-. , -- . . : ,- : . ...

.

-`- ' 1063035 compensation for phasiness in the forward direction is required are as follows:-W Win X = 1 p ~ ( k W ) = 1 j p _ 1 j k Win where k is a positive constant between 0 and 1, -preferably between ~ and ~. Subtraction of jkWin from Y does not alter sound localizations in any way but merely alters the phasiness by reducing the imaginary part of Y/W.
It shoul~ of course, be understood that :. -decreasing the phasiness at the front has the effect of increasing the phasiness of the back where P is negative. ..
However, phasiness at the rear of the listener is r. ..... - -. . ..
psychoacoustically less important and an overall improvement is obtained.
Embodiments of the invention will now be described by way of example with reference to the accompanying drawings in which:
Figure 1 is a schematic diagram of a sound reproduction system illustrating the disposition of the 25: loudspeakers round a listening position and their :
: :~ connection to a decoder, Figure 2 is a block diagram of a known decoder suitable for use in the system shown in Figure 1, Figure 3 is a block diagram of a decoder in . 30 accordance with another embodlment of the inventlon, Figure 4 is a block diagram of a decoder in accordance with a further embodiment of the invention, and B Figure 5 i8 a block diagram of part of a -. ,: , - ., ' ...............' . :' , ' ' ~.,.; '' ,,, , ''. . ' '' ' ,' ' ,; ' , - . ., :-.,~ .. ... , . :

~06303S

decoder in accordance wlth a third embodiment Or the invention.
It should be understood that, in the following de6criptlon, ~here re~erence is made to a set of pha6e shi~ting circuits applying dirferent phase shirts to difrereDt parallel channels, the phase shirt speci~ied in each case is a relative phase shift and a uni~orm additional phase shift msy be - applied to all channels if de6ired. Simllarly, where it is speciried that particular galns are applied to parallel channels, these gains are relative gains and a common additional overall gain may be applied to all channels ir desired.
Before describing embodiments o~ the invention, it will be con~enient to describe the basic ~orm o~ a type of decoder suitable ror use ~ith rectangular oudspeaker layout~, hereinsrter rererred to as a WXY decoder. The invention may be applied to any decoder Or this type.
Referring to Figure 1, a listening location centred dn the point 10 is surrounded by ~our loudspeakers 11, 12, 13 and 14 which are arranged in a rectangular array. The loudspeakers 11 and 12 each subtond an equal angle ~ at the point 10 relative to a rererence direction indicated by an arro~ 15. A
loudspeaker 13 is dispo~ed opposite the loud6peaker 11 and the loudspeaker 14 disposed opposite the loudspeaker 12.
Thus, assu~ing that the rererence direction l~ the forward dlrection, the loudspeaker 11 is disposed at the le~t rront position~ loudspeaker 12 at the right ~rontpositlon, the loudspeaker 13 at the right back position and the loud~peaker 14 at the lert back position.

All four loudspeakers 11 to 14 are connected to receive respective output signals LF, RF, RB and LB from the decoder 16 which has two input terminals 17 and 18, the received omnidirectional signal W1 being connected to the terminal 17 and the phasor signal P1 to the terminal 18.
Figure 2 shows a known WXY decoder suitable for use as the decoder 16 when the angle ~ = 45. m e decoder takes the form of a WXY circuit 20 and an amplitude matrix 22. The WXY circuit 20 produces an output signal W
representing pressure, an output-signal X representing front-back velocity and an output signal Y representing left-right velocity. m ese signals are then applied to the amplitude matrix 22 which produces the required output -signals LB, LF, RF and RB.
The amplitude matrix 22 fulfils the function of the following group of equations:-LB = ~(-X + W + Y) LF - ~(X + W + Y) RF = ~(X + W - Y) R~ -X + W - Y) Any decoder which produces the.four output signals LB, LF, RF and RB 1s the equivalent of a WXY circuit and an amplitude matrix, and thus constitutes a WXY decoder, provided that ~t-LB + LF - RF + RB) = o ; m e WXY circuit 20 may have more than two inputs. In fact this decoder is the same aq the decoder shown in Figure 5 o$ the above-mentioned Canadian Specification No. 977687 the 90 pha8e hi*t circuits serving as the active part of the WXY circuit 20 and the adder~ and phase inverters 8erving as the amplitude matrix 22.
m e nature of the WXY circuit depends on the form , ~

of the input signals. If, as shown, the input signal~
comprise an omnidirectional signal W1 and a phasor signal P1 of the same magnitude as the omnidirectional signal but with a phase difference equal to minus the azimuth angle, the outputs of the WXY circuit 20 are related to its inputs as follows:-W = W

Y -J~iP1 Figure 3 shows a decoder similar to that of Figure 2 but forward biased in accordance with the invention. The forward hiased decoder comprises a WXY circuit 24 which is similar to the WXY circuit 20 except that it has an additional jW output~ The X
and W outputs are connected directly to the amplitude matrix 22 as before; The jW output i~ connected via a variable gain amplifier 26 to a subtraction circuit 28 where it is subtracted from the Y output of the WXY circuit 24. m e output Y of the subtraction -circuit 28 is connected to th.e amplitude matrix 22.
The gain of the amplifier 26 is set to k, i.e. a positive value between 0 and 1 as stated above.
Con~eniently, in the case when the WXY circuit 20 received two input signals comprising omnidirectional and phasor ~ignal components k may be in the range from ~ to ~. `
A similar modification may be made to any of the WXY decoders described in co-pending Canadian Applicatlon No.222146.. The subtraction of the ~W signal from the Y signal may be carried out at any convenient . ' , - .

. .
..

point between the WXY circuit and the smplltude matrl~.
Conveniently, this subtractlon ls carried out on the output signals from the WXY circuit but other arrangements are pos~ible For example, as shown in Figure 4 Or the present 6pecirication, the output Or the V~Y
circuit 24 may be connected to respectlve shelr filters 30 to 33, the shel~ filter 31 for the W signal being a type I shel~ filter and the shelr ~ilters 30 and 32 ~ill be X and Y signals belng type II ~hel~ fllters as de6cr~bed in the above mentioned co-pending appllcation. The 6helf filter 33 for the ~W signal is a type III shelf filter which has a matched phase response identical to those Or the types I and I~ shelr rilters. This enablesthe constant k to be ~re~uency depend~nt BO that the degree o~ residual phasiness can be controlled according to the sensitivity Or the human ear to phasiness at each ~requency. Howe~er a design 6impli~ication or economy o~ apparatus may be achle~ed by making the type III 6helr rilter the same as the type I shelr filter in whlch case the runction Or the6e'two rilter6 can be per~ormed by a slngle rilter operating on the W eignal, and a 90 phase 6hlrt c~rcult used to produce the jW signal ~roD the output Or thl6 rilter. The signals are then applied to a lay-out control sta~e 34 and a distznce control stage 38 substantially a6 described in the above-mentioned co-pending application.
The subtraction Or the ~U signal may al60 be perrormed aM er the lay-out control staèe 34 and/or the distance control stage 38 although thls ~ill mean that the resulting co~pensatlon for phasiness ~ ary ~ith the~e adJust~entE.

_ 9 _ ,.~, ~ . ~ - "
'., ~

The application o~ the in~ention iB not limited to decoders havln8 omnldirectional and phaaor inputs but can also be applied to more general cla6ses ~r signal~ encoded on two channels. For example i-t may be app~ledtoan encoding method such that one linear combination A Or the two channels may be consldered to be an omnidirectional signal and another llnear combination B may be considered to be (c08 ~ - J q s1n ~) times that Or the linear comblnation A, where ~ 18 chosen suitably for each encoded sound position and q is a resl non-zero constant. ~ may be equ~l to the intended azimuth angle during the encoding process or may be some ~unction Or that angle. In the rollo~ing decoding equations, ~ i8 treated as ~he angle from ~5 ~hich the sound will be heard after decoding.
The decoder ror such signals ~ill have the rOllowing equations;
W = A
X =.~B
~ = ~J q~1 (B kA) where ~ i8 a constant which may be frequency-dependent and k is a positive constant less than 1. The subtraction Or kA rrom the signal Y is the process Or ror-:ard biasing in accordance with the invention so as to ~inimise 90 phase shirted components Or Y ror sounds rOr which ~ is near zero. The value Or ~ ~ill ideally be about~ at rreQuencies substantially belo~
350 Hz and around 1/~ at substanti&lly highei rreQuencies.
- The errect Or the ror.;ard bias ters ln the sbove exprcsslon ror Y 18 not only to reduce the p~siness Or sounds to~ards the rront but also to -increa~e the gain Or sounds from the back and to , --10 ~
j;. .. , -. . . . . .. . . . . .. .
~, . .

'10 63035 reduce that of sounds from the front. Thi~
may help to compen~ate ~or any rel~tive'excessive eains at the rrontin the 6ignals A and B durin~
encoding. There are ~everal ~ystems in which such exces~ive front gain exi6ts.
~or example, the invention may be applied to tv~o channel 6ignalg where the 8ignals in the two channels are linear combinatlons Or C and D ~possibly involving phase shirts) where C has gain ~0 (1 + ~ cos-~'+ ~ ~ sin ~) and D has gain (~ + C08 ~ - ~ sin ~) where ~ is a non-zero constant-Both ~1gnals have t-he same gains ror'all azir,uths and ;-the signal D lags the 6ignal C by a phase angle ~, ~ust'as ror an omnidirectiona~/phasor encoding, but C does not have constant gain with angle, its actual energy gain being (1 ~ p2 + 2~u cos ~) at azimuth g.
~'Jhere y is positive, this gain is higher at the rront than at the back and these signals may be'decoded by treating C as an omnidirectional signal and D as a phasor signal and using the forward biasine to help to re~tore equallty to the gains during reproduction as well giving lower phs~iness ror sounds rrom the front.
The invention may also be applied to three-channel systems Or the type in which the third channel' is Or poorer quality than the other two channels. For éxample, on a three-channel record, the two high quallty channels may be base band channels and the thira channel recorded u8ing a subcarrier.
In one three-channel syste~ the three '~'~
transmltted signals are Wln, P and P~ ~here P~ i8 the 8ignal ~ho~e directlonal gain is the co~plex con~ugate o~ that Or P. Thc respective gains Or the three signals - , . ~ ,- ~ ; .
.
- ' :.

at azimuth ~ are 1, (c08 ~ - ~ sin ~) and (C06 ~ + 3 8in ~).
An "ideal" WXY circult rOr these three channels, without forward blasing, iB given by:
W = Win X -- ,B ( ~P + ~P~') y =,~ P - ~P~) where ~ is a real constant which may be rrequency-dependent.
This decoder does not suf~er from phasiness but gives equal signi~icance to the signals P and P~. In order to ~0 reduce the signi~icance of the supposedly lo~ guality signal P~, the following type Or decoder has been proposed:
W = Win X = ~ ~P + (1 - t)P*~
Y = ~ ~jP ~ t)jP~
where t is a positive number between ~ and 1. Ir t =
the resulting decoder is the rull three-channel decoder described above and where t = 1 the resulting decoder 18 a two-channel decoder. t can vary with rrequency if desired. This system is subJect to phasiness and in order to reduce phasiness ror front images, it may be ~orward biased as rollo~s:
W = Win X = ~ ~tP + (1 - t)P~
Y = ~ Lt~P - (1 - t)~P~ - k(2t - 1)~W
Althou~h the undesirable æide efrects Or increase in the gain at the back relative to that at the rrontalso occurs, the magnitude o~ this errect is less than that ror a t~o-channel decoder.
. .
- In a full three-channel system, there are signals other than ~i that have 90 phase ~hirt relative ~o W ror a~
azimuths. Any real linear combination o~ ~W, ~(P + P~) and (P - P~) has the required 90 phase shift. Con~equently, a threc-channel decoder can be _ 12 -,: ,: -,: .

. .
......... . . - . . . . .
- . ., . .. : . . . ~

rorward biased wlthout arrecting lts ba6ic image localization by adding any real linear comblnation Or the~e three signals to X and Y in the basic decoder equation. Such bias need not necessarily be in the forward direction (in which ca~e it i6 not for~ard bias) and may be used to alter the gain o~ the decoder in 60me direction~ relative to others.
With some encoded signal6, all signiricant sound sources or a dominant sound source may be located at a particular azimuth at any one instant Or time.
In these circumstances, it may be desirable to apply a bias signal to reduce the imaginary components of the velocity signal com~onents signal for this particular azimuth More specifically, a decorder matrix for this purpose may have the following decoding equations :

W 'Wln X = ~(P ~ Ju Win?
Y - ~tiP ~ J v Win~-~here ~is a real constant which may be rrequency dependent and u ar,d v are real numbers, repreæenting gains, which ~ary according to the deduced distribution Or sounds ~ ~ J
- in the encoded signals. '-If it is deduced that all the sounds in the encoded 6ignals sre at azimuth ~ then the ideal vslues Or u and v are uYsin p ~ cos ~
in order to cancel out the 90 phase ahlrted components o~ X and Y. Ir the general tendency o~ sounds 18 to be towsrds azimuth ~, but ~ith a certainty r~ here r may be related to the spread Or sound sources a~ay ~rom ~zimuth ~), then putting:

, u c r æin ~ -rcos ~
gives acceptable results. Inaccuracies in the estimates ror ~ and r do not af~ect the ~ub~ective results very critically becau6e azi~uths near ~ are also decoded with relatively low phasines6.
Several methodæ estimating ~ and r are known and one technique will be described by way Or example.
Figure 5 illustrates a Y~ circuit incorporating variable bias in accordance with the invention for decoding the æignals Win and ~P~
The Win signal is applied to a 0 phase shl~t circuit 50 for producing the signal W and to a 90 phase shi~t circuit 52 ror producing the signal jWin Similarly, the phasor aignal ~P i8 applied to a -90 phase shi~t circuit 54 and a 0 phase shift circuit 56.
The outputs o~ the phase shi~t circuits ~4 and 56 are connected ~ia respecti~e adders 58 and 60 to the X and Y
.
outputs o~ the WXY circuit, the adders ~8 and 60 being used to apply the required biasing aæ will no~ be describe~.
It can be shown that for practical purposes cos ~ and sin ~ can be considered aæ given by -2r cog ~ = En(Win ~ P) - En(Win + p) ~5 En(Win) En (W + jP) - En(W - jP) and 2r sin ~ = ln ln En(Y;in) where En(S) ~ean~ the en~elope o, a ~ave ~orm S.
3 In the circuit shown in ~igure 5, the omnldlrectlonal slgnal Win is applled to an en~elope detector 50~to produce the ~lgnal En~ hich 16 the : ~, - '' `.
.. .

denominator Or both the abo~e expre~si~ns. The ,signal En(W1n + P) produced by an envelope detector 60 respon~ive to an adder 62 and the signal ED(W1n - P) iB produced by an envelope detector 64 which i8 responslve to a.subtraction circuit 66. The outputs Or the envelope detectong60land 64 are applied to a ubtraction circuit 68 to produce the numerator Or the e~pression ror cos ~ and thls i8 divided bg the output of the en~elope detector 58lln a divider 70. The output ' o~ the divider 70 is multiplied by ~Win in a multiplier ,72 to obtain the required blasing signal for,the Y output.
This bias1ng signal 1B then applied via a variable ~ain amplirier 74 to the adder ~
, The blasing slgnal for the X output is obtalned in a similar manner. The signal En(Win + ~P) : is produced by an envelope detector 76 which is.responsl~e ,.
.,, . to an adder.78. The signal En(Win -JP) 18 produced by .-~ -'an envelope detcctor ôO ~hi~ch is responsi~e to a - ~ -subtraction circuit ô2. The outputso~ the envelope detectors 76 and 80 are applied to a subtraction circuit 84, :the output'o~ ~hich is di~ided by tho output Or the - en~eiope detector 58lin a divider 86. The output Or the divider 68 is multiplied by the output of the phase ~:~ : shirt circuit 52 in a multiplier 88 and the resulting ~ s~ ~ ~
biasing signal is applied to the adder1~ia an ~: , ,: : ampli~ier 90. - ' . ~ ' ' ' T~u- the biasing signals applied to the X and Y
;~ outputs o~ the clrcuit sho~n in Figure 5 are dependent on the azimuth Or t~e'dominant sound represented b~
thc coded slgnals Win and ~ and tbe magnltuao o~ tho ,,biaslng sigDals de~ends o~ the amplituae Or tho .~ . .. . .
~ aomlnant slgDal as compared ~ith the amplitude or signals `~ 5 ., " . . . ' ~ .. . ,~ , . .

from other directions. If sounds of equal intensity -come from directions of widely differing azimuth so that there is no dominant signal, the inputs to the subtraction circuits 68 and 84 will be equal so that their outputs are zero.
A simplified variable bias decoder may be obtained by applying a variable bias signal only to the Y output of the WXY circuit and not to the X output, i.e.
by putting u equal to zero. This will "enhance"
directional resolution to the front and/or the back but not at the sides.
Directional biasing may also be applied to non-rectangular loudspeaker layouts. For example, in a regular polygonal array, the signal fed to each loudspeaker may be:- -k1 W + k2 (X' + k3jW) cos~ + k2 (Y' + k4jW) sin~
where X' and Y' are the velocity signal outputs of the WXY circuit and k1 and k2 are both greater than zero and ~here ~ is the azimuth of the loudspeaker to which the signal is fed. The terms k3jW and k4jW are the directional bias terms. k1, k2, k3 and k4 may be frequency dependent and/or may be dependent on the supposed instantaneous direction of the dominant signals but otherwise they are real constants. The circuitry required to implement such polygonal decode~rs differs from that illustrated in Figures 2 to 5 only in that the output amplitude matrix 22 is replaced by an amplitude matrix having n outputs Si (corresponding to --~ -loudspeakers at azimuths e1~ .... ~n spaced apart by 360/n ) given by Si = k1 W + k2 X cos~i + k3 Y sin~i When directional biasing is applied to ~:. ' three-dimensional systems, biasing may be applied to the Z component of the velocity signal as well as or instead of the X and/or Y components.

,; .
.~.. ~ ~ ,., :~

~: ' ' ~ -: , -,

Claims (11)

The embodiments of the invention in which an exclusive property or privilege is claimed are defined as follows:

1. A decoder for a sound reproduction system having at least three loudspeakers surrounding a listening area, the decoder comprising input means for receiving at least two input signals comprising pressure signal components and velocity signal components, means for subtracting from the velocity signal components of a chosen direction a directional bias signal comprising a signal all the components of which have a ?90° phase relation with respect to the pressure signal components, and output means for producing a respective output signal for each loudspeaker.

2. A decoder according to claim 1, wherein the directional bias signal is a fraction of the pressure signal phase shifted by 90°.

3. A decoder according to claim 2, wherein the fraction of the pressure signal is in the range of one third to one half thereof.

4. A decoder according to claim 2, or 3, wherein the pressure signal components are omnidirectional signal components and the velocity signal components are phasor signal components.

5. A decoder according to claim 1, wherein the input means is adapted to receive three input signals and derive therefrom a pressure signal and two velocity signals which, for all sounds are either in phase or 180°
out of phase with the pressure signal, the signal subtracted from the velocity signal being real linear combinations of the pressure signal phase shifted by 90° and the velocity signals phase shifted by 90°.

6. A decoder according to claim 5, wherein the two velocity signals are respectively the sum of a phasor signal and its complex conjugate and the difference between the phasor signal phase shifted by 90° and its complex conjugate.

7. A decoder according to claim 1, including means responsive to the input signals for determining the azimuth angle of the most significant sound source and means for applying a directional bias signal dependent on said azimuth angle.

8. A decoder according to claim 7, including means for producing first and second mutually orthogonal components of the velocity signal of azimuths 0° and 90°
respectively and means for applying a first directional bias signal to said first component and a second directional bias signal to said second component.

9. A decoder according to claim 8, wherein the magnitude of the first directional bias signal is proportional to minus the cosine of said azimuth angle and the magnitude of the second directional bias signal is proportional to the sine of said azimuth angle.

10. A decoder according to claim 9, wherein the bias signal applied to the first of said mutually orthogonal components is proportional to the difference between the envelope of the difference between the pressure and velocity signals and the envelope of the sum of the pressure and velocity signals divided by the envelope of the pressure signal and the bias signal applied to the second of said mutually orthogonal components is proportional to the difference between the envelope of the sum of the pressure signal and a signal comprising the velocity signal phase shifted by 90° and the envelope of the difference between the pressure signal and said signal comprising the velocity signal phase shifted by 90° divided by the envelope of the pressure signal.

11. A sound reproduction system incorporating a decoder according to claim 1, 2 or 5.