US 3046337 A
Description (OCR text may contain errors)
July 24, 1962 W. F. HORNYAK STEREOPHONIC SOUND Filed Aug. 5, 1957 i560/30mm fao/w 107 4 Sheets-Sheet l I/IIIIIIII/ IIIIIHIIIIIIIII) Fmi( 4M Fame-12,24
July 24, 1962 w. F. HORNYAK 3,046,337
` STEREOPHONIC SOUND Filed Aug. 5, 1957 4 Sheets-Sheet 3 SELEC 7'/ VE 511A/0 PASS F/l TEF@ IN V EN TOR.
Arrakis/EY July 24, 1962 w. F. HORNYAK sTEREoPHoNIc SOUND 4 Sheets-Sheet 4 Filed Aug. 5, 1957 Ik www OL M O M m 0 w a uw@ w `||`v o NU IO. F NU. fl fbi, L
waL A/A/s a I l I H E0 m V m# w rl. 4 M A 0 U L I. w/Y o B o m w o P y w W m O M W /Oan M l. o i k| ls om m o y L A o United States Patent Jersey Filed Aug. 5, 1957, Ser. No. 676,160 6 Claims. (Cl. 179--1) This is a continuation-impart of application Serial No. 641,910 filed February 25, 1957, now abandoned, by
. William F. Hornyak.
The present invention relates to improved stereophonic `sound systems and particularly to improved apparatus for providing stereophonic and reverberation effects.
Binaural hearing includes the faculty of locating a source of sound by sensing the differences in sound quality reaching each ear. Stereophonic sound reproducing systems attempt to simulate the binaural effect. Such systems include at least two spaced lmicrophones arranged to receive the sound from an orchestra or other performing group in different intensities and phases. The signals from the microphones are applied either directly or after recording or broadcast, to speakers in a listening room in positions corresponding to the microphones. The effect to some extent is similar to that of being present at the live performance. t
One disadvantage of the conventional stereophonic system described above, when used to record stereophonic sound, is that two sound tracks must be recorded, one for each microphone. A similar disadvantage is present in broadcasting the stereophonic signals. Since there are two differently modulated signals to be transmitted, they must be sent either in separate channels or in other ways in which the signals can be maintained isolated from. one another. For example, radio station WQXR, New York, transmits stereophonic signals by applying one of the modulated signals to its amplitude-modulated facility and the other to its frequency-modulated facility. This requires, at the receiving point, an AM and an FM receiver, and separate spaced speakers for each.
An object of this invention is to provide an improve-d stereophonic sound system which overcomes both of the above disadvantages.
Another object is to provide a stereophonic sound system which is entirely compatible with monaural sound systems. The compatibility is such that:
(a) The stereophonically prepared sound track can be used with existing monaural playback systems without equipment modification.
(b) The stereophonic playback system can receive the conventional monaural sound track without equipment modification.
(c) The stereophonic playback system can receive the stereophonically prepared sound track without equipment modification. In case (c), the reproduction of sound is accompanied by stereophonic and reverberation effects.
Another object is to provide improved stereophonic signal producing systems.
A further object is to provide improved stereophonic signal reproducing systems.
The stereophonic sound systems herein comprise ygenerally a stereophonic signal-producing portion or recording system and a signal-reproducing portion or playback system. The output of the signal-producing portion may be used with equal facility on either the stereophonic signal-reproducing systems herein or on conventional monaural signal-reproducing systems. The signal-reproducing systems herein may be used to reproduce the output of either the stereophonic signal-producing signals 'herein or the output of conventional monaural signal-producing systems.
In a simple embodiment of the invention, the stereophonic signal-producing portion or recording system of this invention includes an array of signal receiving means, each for passing audio signals in different portions of an audio frequency spectrumof interest. Each signal receiving means may comprise, for example, a microphone connected to a frequency-selective channel including, for example a band pass filter. The signal receiving means are spatially arranged in an order unrelated to that of the frequencies Vpassed by the channels to which they are connected. Preferably, the frequency band allocation is such as to give a broad frequency sampling per unit of array length. In one form of the invention, for example. the frequency bands are arranged in random order relative to the spatial positions of their associated microphones. Preferably, the entire array of channels passes all, or substantially all, frequencies within the spectrum of interest. The output signals of all channels are combined in a mixer and the composite signal is either recorded, applied directly to the audio signal reproducing portion of the system, or used as the modulating signal for a transmitted carrier wave. In other embodiments of the signal-producing portion, the microphone or other pickup means may feed more than one frequency-selective channel Vup to an array of such channels covering the entire frequency spectrum of interest. Adjustable electrical means are provided to synthesize distance and sound quality for each channel of the array.
The signal-reproducing or playback portion of the system includes a single signal pickup channel, feeding frequency-selective channels which in one embodiment correspond in number and frequency bands to the channels of the signal-receiving means of the stereophonic signal-producing portion of the system. The channels are connected Ito transducers (speakers, for example) spaced in positions in an array which correspond to the actual or synthesized .positions of the signal receiving means (microphones). Thus, spatial coding is achieved through the use of selectively allocated frequency channels. In otherv embodiments, the number of playback channels may be lgreater or less-than the number of recording channels. Y t
The invention will be described in greater detail by reference to the following description taken in connection with the accompanying drawing in which:
FIGURE l is a block circuit diagram of a stereophonic sound recording and playback system according to this invention;
FIGURE 2 is a diagram illustrating the spatial arrangement of the microphones and speakers shown in FIG- URE 1, relative to the frequency bands passed by their associated filters; f
FIGURE 3 is a'diag'ram to explain how the system of FIGURE 11 operates; FIGURE 4 is a block circuit diagram of a portion of the circuit of FIGURE 1 in modified form, to` illustrate a broadcasting embodiment of the invention; v FIGURES 521 and 5b are diagrams to illustrate Waysr in which the number of speakers in the receiver or playback par-t of the stereophonic system may be substantially reduced;
FIGURE 5c illustrates another microphone (or speaker) arrangement according to fthe invention;l
FIGURES 6 and 7 are block circuit diagramsof embodiments of this invention which are useful for converting. the conventional, two channel stereophonic signals to the single channel stereophonic signals of the present system; y v
FIGURES 8a and-8b are plan and elevational views of an embodiment of a playback system of the invention;
FIGURE 9 is a blockcircuit diagram of an embodiment of a recording system of the invention; and
FIGURE is a plan view of a fictitious orchestral arrangement desired to be synthesized.
Similar reference characters are used for similar structures throughout the drawings.
Referring to FIGURE 1, block 10 indicates a recording room at one end of which an orchestra or other performing group may be located. An array of microphones y14-1 to l4-n is arranged in one possible configuration along the opposite wall of the room, where the numbers after the dashes indicate the spatial positions of the microphones, and n is an integer (indicating both the last microphone in the array and the total number of microphones) from about l0 to 50 or more. The number of microphones is not critical, however, the greater the number, up to a practical limit, the better the stereophonic effect. The microphones need not be equally spaced from one another, and need not be at the same height.
Each microphone feeds a different `band pass audio filter in filter bank 16. The filters 16-1 to l6-n are tuned to different bands inthe audio frequency spectrum, without substantial overlap. In other words, the group of filters, taken together, provide an overall pass band covering the audio frequency spectrum of interest say 20-20,000 cycles. The frequency bands are preferably spatially allocated to give as broad a frequency sampling per foot of microphone array as possible. For example, a random spatial allocation of frequency bands gives satisfactory performance, and if n is sufficiently large, the desired broad frequency sampling per foot of array results. A typical `frequency band distribution may in clude the following:
Frequency band covered Position of filter: (in cycles per second) 16n-Z 10,000-12,000 e16n-1 25-30 16u 3300-4000 Another possible distribution of frequency bands is illustrated in FIGURE 2. Each frequency band is represented by the distance between a pair of adjacent lines and, in general, covers a range of frequencies equal to about 5%-20% of the upper frequency of the respective band. Again, these figures are not critical.
For universal application of the system, the spatial allocation of frequency bands once selected should be maintained as a fixed code so that many play back systems may usethe same stereophonically coded input channel. Thus, in the example given microphone 16-4 (the fourth yfrom the left) always works into a band pass filter covering the range 6200-7500 cycles. Further, in the signal-reproducing or playback system, the filter covering the band 6200-7500 cycles is always connected to the rfourth speaker from the left, etc.
Returning to FIGURE l, the output signals of filters 16-1 through 16-n are applied either directly, or through adjustable impedance elements 18-1 through 18-n, respectively, to common mixer stage 20. Adjustment of the impedance elements, shown in the drawing as adjustable resistors, permits one either to accentuate certain portions of the spectrum, the higher or lower frequency components, for example, or to balance all channels (overall fiat response characteristic), or to obtain accentuated stereophonic and reverberation effects. The output signal of the mixer is applied to a recording means, such as a pickup head 22, which records the signal on a recording medium 24 such as a disc, drum, tape, wire, film or the like. Note that only a single recording head is required and that the signal may be recorded on a single track.
The playback portion of the system is the analog of the recorder. For the sake of drawing simplicity, both portions of the system are shown together but, of course, this is ordinarily not the case. The recorded signal is received by pickup head 26 and applied to amplifier 28. The latter amplifies the signal to the required power level and applies it either directly or through adjustable impedance elements 30-1 through 30-n, to band-pass filters 32-1 through .S2-n. The impedance elements perform the same function as the corresponding elements in the recording portion of the system. They, like those in the recorder, are not essential to this invention. The band-pass filters correspond in frequency band to the similarly positioned filters in the recorder. The filters feed an array of speakers 34-1 through 34-n spaced along a wall of listening room 36 in positions corresponding to those of microphones 14-1 through 14-11.
The operation of Athe system can be better understood by referring to FIGURE 3. The rectangle 37 represents a listening room at one end of which is a sound source 38 and at the other end of which is a listener 40. Between the listener and the source is an acoustic partition consisting of a wall to wall array of acoustic band-pass filters 42-1 through 42-n, which correspond to the electrical, audio frequency band-pass filters 16-1 through 16-n. For the purpose of this explanation, consider the partition to be removable. The room arrangement with the partition in place is analogous to the system of FIG- URE 1. Without the partition in place, the room may be taken as the standard of perfection insofar as subjective `binaural hearing effects are concerned.
For the purpose of this explanation, consider the source 3S first to be a sound radiator of white noise (a noncoherent sound signal having frequency components which extend over the entire audio spectrum). Each element of the filter array now receives and passes a signal. The listener at point 40 observes that each foot of the array radiates a sound possessing no characteristic pitch (and thus gives the impression of vbeing a noise generator itself). The portion of the array containing lter x radiates the signal of greatest amplitude, as heard by listener 40, since filter x is located on the straight line between the listener 40 and the sound source 38. Thus, the direction 38-40 is esablished as the one from which the sound is emitted-the same effect as when the filter array is not present.
An additional directional or stereophonic effect is achieved when the white noise sound source 38 is turned on and off. The burst of sound heard lby the listener arrives first from the foot of array containing filter x. The reverberation climate is such as to give the illusion of the source being in the direction 40-38.
If the sound source is a pure tone generator (no harmonics), the listener cannot determine its true direction with the system of this invention. The source will appear to be in the direction of the filter element tuned to the particular tone which in general will not be located on the line 40-38. However, this is not as great a disadvantage as might first appear to be the case. For one thing, it is relatively rare that one is interested in listening to pure tones. For another, tests have shown that the average person has rather poorer sense of directional localization for pure tones than for noise or sound patterns rich in frequency components. Thus, the loss of directional localization for pure tones encountered with the present stereophonic system is less important than it might otherwise be since the average person expects less localization in such cases.
If the sound source 38 is frequency modulated at a sufiiciently rapid rate, the statistical integration of the sound received by the ear provides an effect similar to that observed with the source of a white noise generator. The lowest scanning rate necessary to produce a satisfactory stereophonic effect is reduced considerably when the modulated tone generator is rich in harmonics.
Turning the modulated generator on and otf also results in stereophonic effects similar to those observed with the source of 'a noise generator. Y
Since no two filters radiate the same frequency, it is not possible to have coherent interference effects that might lead to a Huygens Construction reproducing the sound patterns in the listeners portion of the room 37 which would correspond to the c-ase with no partition. However, for certain sound patterns rich in harmonics, the non-linear ear response generates sufcient difference tones to, in elect, convert the ltered array into a coherently radiating array and thereby subjectively produces a directional clue.
In summary, the most pronounced stereophonic effects are obtained when the source 38 produces signals which are rich in harmonics, rapidly Variable in frequency and transient like in behavior. This type vof signal is produced when various orchestral and/ or human voices are engaged in rapid musical dialogue. Thus, in the types of performances in which the binaural effect is most desired, the system of this invention gives optimum peri formance.
An embodiment of the invention for `broadcasting stereophonic signals of the type discussed above is shown in part in FIGURE 4. All of the stages down through mixer 20 are identical to the like numbered elements of the system of FIGURE l, only block 20 being shown in FIGURE 4. However, rather than feeding a recording head, the stereophonic output signal of the mixer is applied to modul-ate the carrier frequency of transmitter 51. This is Ia live broadcast. One may also record the mixer output, then pickup the recorded signal, and apply the picked-up recorded signal to modulate a carrier frequency. This is a recorded broadcast. Any type of modulation desired may be employed las, for example, amplitude, frequency, etc. The modulated carrier is radiated by the transmitter and received by receiver 52. The demodulated signals are applied to the band-pass iilters 'S2-1 through 32-n (not shown in FIGURE 4) and thence to the loudspeakers.
In la preferred playback system of the invention, there are the same number of loudspeakers as there `are microphones. However, it may be desirable in some cases, in the interests of economy, substantially to reduce the number of loudspeakers. The latter, for example, may be part of a home receiver or home playback system where cost is an important consideration. Two ways of doing this, without 'at the same time destroying the stereophonic eect, are illustrated, one in FIGURE a and the other in FIGURE 5b.
In the arrangement of FIGURE 5a, there are n frequency-selective channels. The signals from three of the channels are combined and applied to a single playback channel having a bandpass sufficiently broad to cover the three channels. The frequencies of each group of three recording channels are adjacent to one another in frequency band but not in space. The signals passed by the single playback channel fed by the signals passed by the three recording channels are connected to one loudspeaker. The 'loudspeaker position in the listening room corresponds to the center position of the corresponding three spaced microphones in the recording room. In the example given, the number of playback channels and loudspeakers are reduced by one-third. This method of reducing channels can be equally yas well applied to the combining of 2, 4, 5 or more channels, and, moreover, the combining process need not cover the entire spectral range. i
The method of reducing playback channels and loudspeakers shown in FIGURE 5b is even simpler. Here,
. the spectral range is curtailed. Those channels, for example, covering the range -80 cycles and 10,000- 20,000 cycles are simply removed and the loudspeakers corresponding to these channels are also removed. The
relative spatial positions of the remaining loudspeakers remains undisturbed.
In another practical arrangement, the number of microphones and loudspeakers are both reduced, however, the same number of microphones and loudspeakers are used at each location. The positions of the microphones in the recording or broadcast room correspond to the positions of the loudspeakers in the listening room. -In this arrangement, one channel receives the frequency band from 2li-350 cycles, a second channel the frequency band from 35'0-1-680 cycles, land a third channel the frequency band from 9'520 cycles to 20,000 cycles. The frequency band from 1680-9520 cycles is covered by 10 channels. The vmicrophone for the 20-350 cycle band is spaced centrally, the microphone for the 350-1680 cyle band is immediately to the right of the center, and the microphone for the 9520 to 29,000v cycle band is immediately to the left of center. The remaining microphones are spaced in random ,manner relative to the frequencies passed by the filters to which they are connected, as already discussed in detail. FIGURE 5c illustrates a typical arrangement using a total of 13 microphones. The spectral region from 1680 to 9520 cycles will be observed to be divided into quarter octaves.
While the arrangement of FIGURE 5c does not give quite as good a stereophonic eifect as the one of FIG- URE l, the difference is not as great as one might expect. The reason is that at the lower and higher ends of the audio spectrum, the ear has rather poor directional sensitivity. These bands may therefore be lumped without substantial loss of stereophonic effect.
The arrangement of FIGURE 5c is very advantageous from an economic standpoint. The total number of loudspeakers (and microphones) are substantially reduced and, even more important, the reduction is greatest for the most expensive loudspeakers, that is, those operable at thel low and high ends of the sound spectrum. The speakers for the intermediate portion of the sound spectrum are relatively inexpensive. p
In some situations, it may be desirable to have more loudspeakers in the playback portion than there are channels in the recording portion of the system. This may be done to act as a yblender or to produce -special eifects, particularly with solo voices and instruments. This may also be done to add listening room reverberation effects, or to assist in producing a pleasing elfect when the stereophonic speaker array is used with the ordinary monoaural material.
yFor example, lwith an unaccompanied vocalist, it may be desirable to revert to conventional monaural reproduction or to a system intermediate between stereophonic and monaural reproduction. One method for accomplishing this is to provide additional centrally located loudspeakers which are connected to reproduce the entire audio spectrum of interest, and switching means to allow reproduction through either or both sets of loudspeakers. Another method which may be used is to place additional loudspeakers encompassing a wider than usual frequency band in the loudspeaker array and switching means to allow reproduction through either or both sets of loudspeakers.
.FIGURES 8a and 8b illustrate another arrangement of loudspeakers for the purpose of obtaining special reverberation elfects. A loudspeaker array occupies the corner of the listening room in the13 loudspeaker setup described for FIGURE 5c. bass unit 150 covering the range 20 to 350 cycles is centrally located. A speaker 152 covering vthe range 3.36 to 4.00 kc. occupies the extreme left position; a speaker r154y covering the range `4.0() to.4.'76 kc. occupies the extreme right position .and so forth. addition, an extra speaker .156 is centrally located` as above the bass unit '150 and directed upwardlyl toward the ceilingof the listening room. A composite signal ofv all frequencies in the range 1.68`to 9.52.kc. and at a desired gain level is applied to the speaker. With the proper InV 7 adjustment of the gains, the eifect is to produce reverberation effects appropriate for a room larger than the actual listening room and to provide a more spatially extended sound source.
Existing two channel binaural recordings can be electrically recorded into the one channel stereophonic system of the present invention. FIGURE 6 shows one system for doing this. 'I'he signals recorded on binaural track 60 are picked up by heads 62 and 64 and applied to amplifiers 66 and 68. The amplified signals are applied to spaced speakers 70 and 72 in the transference room 74. The remainder of the system is identical to the recording portion of the system shown in FIGURE 1.
The system of FIGURE 7 is the electrical equivalent of the one of FIGURE 6. Ampliiers 66 and 68 feed two separate, identical arrays of lters 80 and 82. The filters 80k and 82k correspond in band pass to the filter connected to microphone Mk of FIGURE 6. The output signals of filters 80k and 82k are applied through attenuators 84 and 86 respectively, and delay circuits 88 and 93 respectively, to mixer 20.
The attenuations ot introduced are:
D 2 D z Fin-i) f and when) where: D=the distance between the microphone array and speakers in the arrangement of FIGURE 6; and
Rm and Rkg are the distances between the two speakers and the microphone Mk in the arrangement of FIG- URE 6.
'Iihe time delays T introduced are:
where: C=velocity of sound.
The effect is similar to the one achieved with the system of FIGURE 6. Note that the arrangement described is repeated for each frequency selective channel, and each has its own attenuation factor and delay factor, depending on the lter position. For the sake of drawing simplicity, only one mixer input channel is illustrated. Note also that if a less pronounced directional effect is satisfactory, either the time delay or attenuation could be omitted. Y
In some situations, it is desirable to physically separate portions of the original signal source. For example, in phonograph recording, it may be desirable to place the soloists and portions of the orchestra in separate'booths or areas. Physical separation may be desirable to eliminate or to reduce undesired reverberation or positional effects which would otherwise be present in the physical arrangement such as shown in FIGURE 1. Physical separation may also be desirable to permit the accentuation of any particular orchestral unit or units, for example, the tympany or the soloist.
FIGURE 9 illustrates the foregoing principle. A recording room includes n separate booths or reflectors 92a, 92b, 9221 for accommodating separate orchestral units. As shown in FIGURE 9, a booth 92a may contain the soloist; a booth 92b may contain the violins; a booth 92C may contain the tympany; and a booth 92d may contain the'brass section. Each booth is provided with a single microphone 94a, 94b, 9411 in a desired location for picking up the sound of its respective orchestral unit. The signals from each of the microphones 94a, 94h, 94H is applied through its own network 132, 134, 136, etc. respectively, accomplishing electrically that which is accomplished positionally in the arrangement of FIGURE 1. A network for the microphone 94a for the soloist will be described in detail. An identical or a similar network is provided for each microphone or other pickup means.
The output signal from the microphone 94a is applied through an adjustable gain control 96a to an amplilier 98a. The amplifier 98a feeds an array of filters 1112` which Rkz-D C correspond in number and bandpass, for example, to the array of filters, 16-1 to 16-n of FIGURE 1. The output signal of each of the filters of the array 102 is applied through its own attenuator and its own delay circuit to a common mixer 20. For example, the output signal of a filter 102k which may pass for example, the band 4.00 to 4.76 kc. is applied through an adjustable attenuator 112k and an adjustable time delay means 122k to the mixer 20. Each combination of filter, attenuator and time delay means is referred to as a channel. The arrangement described is repeated for each channel, and each channel is adjusted to have its own band pass characteristic, attenuation factor and time delay factor depending upon the effect desired.
Suppose for example, it is desired to synthesize an arrangement of microphones and an orchestral arrangement shown in FIGURE 10. Then the adjustments to the attenuators and time delay means for each channel depends on the fictitious base distance D between orchestra and the array of microphones, and upon the fictitious distance R of the particular orchestral unit and each microphone. The attenuation introduced for channel k of the rst network 132 described for FIGURE 9` is:
and for channel n is:
and for channel nis:
where C is the velocity of sound.
The system of FIGURE 9 is the electrical equivalent of the recording portion of the system of FIGURE 1. The system of FIGURE 9 permits single channel compatible stereophonic sound recording, as in the system of FIG- URE 1, but uses Ian actual number and placement of microphones unrelated to the positional and reverberation effects desired. These effects are synthesized. Further, the system of FIGURE 9 permits emphasizing a particular orchestral unit. Note further that emphasis, distance, direction, and reverberation may be modified for each orchestral unit during the playing of a particular piece. Thus one may also synthesize movement in a par- Tus:
The system of FIGURE 9 contemplates the use of one or a plurality of identical arrays of frequency selective channels, each -array covering the entire or substantially the entire audio frequency spectrum of interest. Each channel is tuned to a different portion of the audio frequency spectrum of interest and includes separate adjustable attenuation means and separate adjustable time delay means. The output of all channels is fed to a common load which includes mixer means for producing a single composite output signal. The composite output signal may be recorded, transmitted, or recorded and transmitted. The system of FIGURE 9 permits syn.
" thesizing a position in space of a particular orchestral unit, the synthetic position being movable by adjustment of each channel.
What is claimed is:
1. A single channel stereophonic sound system cornprising an array of at least 10 microphones spaced from one another; a plurality of band-pass audio filters, one
filter connected to each microphone, each lter tuned to pass a different, discrete frequency band in an audio frequency spectrum to which a normal human being has marked directional sensitivity, said microphones being 2. The system of claim 1 including recording means connected to receive and to record said mixed signal.
3. 'Ihe system of claim l including transmitting means connected to receive and to transmit said mixed signal.
4. In a single channel stereophonic sound system for passing an audio frequency spectrum having upper and lower frequency portions to which a normal human being does not have a marked directional sensitivity, and a center frequency portion to which a normal human being has a marked directional sensitivity; in combination, an array of l to 50 spaced microphones; a plurality of bandpass audio lters, one filter connected to each microphone, one lter tuned to pass at least a major portion of said upper frequency portion, one -filter tuned to pass at least a major portion of said lower frequency portion, and
the remaining filters each tuned to pass a different, substantially smaller band in said center frequency portion, the bandwidth of each band being 5 to 20% of the highest frequency in said band, said microphones being arranged in a predetermined spatial order relative to the frequency bands passed by the filters connected thereto such that physically-adjacent microphones are connected to filters which pass relatively widely-spaced frequency bands and adjacent frequency bands are passed by filters connected to relatively widely-spaced microphones, said remaining filters, taken together, covering substantially the entire center portion of said frequency spectrum; mixer means coupled `to all said filters for combining their output signals to produce a mixed signal; and means connected to receive said mixed signal.
5. In a single channel stereophonic sound system for passing an audio frequency spectrum having upper and lower frequency portions to which a normal human being does not have a marked directional sensitivity, and a center frequency portion to which a normal human being does have marked directional sensitivity; in combination, an array of 13 spaced microphones; a plurality of band-pass audio filters, one filter connected to each microphone, a first filter tuned to pass at least a major portion of said upper frequency portion, a second filter tuned to pass at least a major portion 0f said lower frequency portion,
the microphones connected to said first and second filters being in about the physical center of said array, and the remaining lters each tuned to pass a different, substantially smaller band in said center frequency portion, the bandwidth of each band being 5 to 20% of the highest frequency in said band, the microphones connected to said remaining filters being arranged in a predetermined spatial order relative to the frequency bands passed by the filters connected thereto such that physically-adjacent microphones are connected to lters which pass relatively widely-spaced frequencyr bands and adjacent frequency bands are passed by filters connected to relatively widely-spaced microphones, said remaining filters, taken together, covering substantially the entire center portion of said frequency spectrum, mixer means coupled to all said lters for combining their output signals to produce a Vmixed signal; and means connected to receive said mixed signal.
6. A single channel stereophonic sound system comprising an array of at least 10 microphones spaced from one another; a plurality of band-pass audio filters, one filter connected to each microphone, each lter tuned to pass a different, discrete frequency band in an audio frequency spectrum to which a normal human being has marked directional sensitivity, said microphones being arranged in a predetermined spatial order relative to the frequency bands passed by the filters connected thereto such that physically-adjacent microphones are connected to filters which pass relatively widely-spaced frequency bands and adjacent frequency bands are passed by filters connected to relatively widely-spaced microphones, said frequency bands, taken together, covering substantially all of said audio frequency spectrum; a signal mixer connected to receive the output signals of all filters for producing a mixed signal; and an array of speaker-filter combinations coupled to said mixer, said speaker-filter combinations substantially corresponding in number, relative position, and bandpass characteristic to the combinations of microphone-filters of said system.
References Cited in the file of this patent UNITED STATES PATENTS Re. 24,670 Smith July 21, 1959 2,179,840 Bucky Nov. 14, 1939 2,273,866 Holst Feb. 25, 1942 2,340,365 Bedford Feb. 1, 1944 2,352,696 DeBoer July 4, 1944 2,403,232 Parisier July 2, 1946 2,517,819 Young Aug. 8, 1950 2,520,798 DeBoer Aug. 29, 1950 2,616,970 Broos Nov. 4, 1952 FOREIGN PATENTS 313,022 Great Britain June-4, 1929