US3059053A

US3059053A - Stereophonic sound transmission systems

Info

Publication number: US3059053A
Application number: US835604A
Authority: US
Inventors: Percival William Spencer
Original assignee: EMI Ltd
Current assignee: EMI Ltd; Electrical and Musical Industries Ltd
Priority date: 1958-08-26
Filing date: 1959-08-24
Publication date: 1962-10-16
Anticipated expiration: 1979-10-16

Description

Oct. 16, 1962 W. S. PERCIVAL STEREOPHONIC SOUND TRANSMISSION SYSTEMS Filed Aug. 24, 1959 To'NE AND 2 SheetsSheet 1 PHASE SHIFT NETWORK TO PROCESSING CIRCUIT FOR v 9 RECTIFIER I I r RECTIFIER\ CATHODE ELONGATOR FILTER AND FOLLOWER I AMP. FILTER TRANSIENT EMPHASISING FILTER 2 2050 2Zb\NV- 3500 W 6000 c/s W M4 LOW PASS SUPPRESSED 'FILTER CARRIER MODULATOR E E m FILTER AND AMP. AMP. TRANSIENT CATHODE EMPHASISING FOLLOWER FILTER K 37 5000-620S\FH TER ELONGATOR FILTER AND AMP. RECTIFIER 38 5000-120 c/s N\ 5000 330 (2/5 --4 O W1 5000 -500c/s W 5000 700c/s- M 5000-940c/s W TRANSIENT EMPHASISING AMP FILTER 5O 45 I F u- FORM PROCESSING 51 v' CIRCUIT FOR V 1 DIVIDER L R l menta Oct. 16, 1962 w. s. PERCIVAL 3,059,053

STEREOPHONIC SOUND TRANSMISSION SYSTEMS Filed Aug. 24, 1959 2 Sheets-Sheet 2 CATHODE FOLLOWER FIG. 2.

MUGIUCOI NQSPapazLual c/(D' w 3,059,053 STEREGPHONIC SQUND TRANSMISSIGN SYSTEMS William Spencer iercival, West Ealing, London, England, assignor to Electric & Musical Industries Limited, Hayes, Middlesex, Engiand, a company of Great Britain Filed Aug. 24, 1959, Ser. No. 835,604 Claims priority, appiication Great Britain Aug. 26, 1958 12 Claims. (Cl. 179-1) This invention relates to stereophonic sound transmission systems.

A stereophonic sound transmission system has been proposed comprising means for deriving from sound an audio frequency signal of relatively wide frequency range and representative of the waveshape of said sound and a control signal representative of the direction or position of the source of said sound, and such that said audio signal in conjunction with said control signal can be employed to energise spaced sound reproducers to give stereophonic reproduction of said sound. The two signals may be referred to as the sound signal and the control signal respectively and they may be recorded on a disc or magnetic tape record in different tracks or in the same track but in separate frequency bends, or in the same track by complex cutting techniques. The signals may also be broadcast in a radio transmission. At a reproducer or receiver the sound signals may be applied to two separate loudspeakers or groups of loudspeakers, the gains of said loudspeakers being controlled difierentially by the control signal to convey to a'listener an impression of the relative position or direction of the source of the sound being reproduced.

In such a system it is desirable that the frequency band of the control signal should be small but that the stereophonic eifect should be realistic and be capable of giving the impression of spaced sources or" sound sounding virtually simultaneously.

The object of the present invention is to provide an improved system with a view to achieving these desirable results.

According to the present invention there is provided a stereophonic sound transmission system comprising means for deriving from sound a plurality of audio signals which differ in amplitude one from another in dependence upon the direction or position of the source of the sound, a plurality of groups of processing channels, one group of processing channels for each of said audio signals, filter means for selecting the components of each audio signal into a group of different frequency bands, one frequency band for each channel of a group of channels, and for applying the components in the diiferent frequency bands to their respective channels, each channel including means for processing the frequency components therein, and means responsive to the output signals from the plurality of groups of channels to derive a control signal representative of the direction or position of the source of the sound, and means for transmitting said control signal and an audio signal representing the waveshape of said sound.

It has been found that the sense of directivity of sound waves is associated predominantly with transients, that is rapid changes in amplitude of the sounds. It has also been found moreover that for any particular single sound the sense of directivity is associated principally with the initial transient of that sound although there may be subsequent transients in the same sound corresponding for example to beats, vibrato, etc. When, in accordance with the invention, means are provided for emphasizing transients and for elongating the pulses which arise due to transient emphasis, only the initial transient of any 3,059,053 Patented Oct. 16, 1962 series of transients occurring within an interval which is comparable with the time constant of the elongating means receives full emphasis and following transients of the series are substantially unemphasised or less emphasised. This effect enhances the stereophonic effect achieved on reproduction inasmuch as following transients in a particular sound do not receive emphasis in the control signal. However without precaution to the contrary, the risk arises that a different sound may arise from another source, for example another instrument in an orchestra, having an initial transient within the aforesaid interval and in the absence of precaution to the contrary the elongating means tends to prevent the initial transient of the difierent sound from receiving emphasis. However by virtue of the invention, as the signals from which the control signal is derived are divided into a plurality of components in diiierent frequency Zones and the different components receive individual transient emphasis and elongation, the possibility of giving appropriate emphasis to such a new sound is increased in proportion to the number of components in to which the directional signals are divided since the new sound is likely to have a substantially difierent frequency or frequency spectrum from the first mentioned sound. In this way the invention allows transient emphasis to be confined only to the initial transient of a particular sound but facilitates the emphasis of the initial transients of difierent but overlapping sounds. The invention also enables distortion of the control signal due to reverberation to be re duced or controlled.

In order that the present invention may be clearly underestood and readily carried into eiiect, the invention will now be described with reference to the accompanying drawings in which:

FIGURE 1 illustrates in block form one example of a stereophonic sound transmission system according to the present invention, and

FIGURE 2 is a detailed view of part of FIGURE 1.

Referring to FIGURE 1 a pair of directional microphones 1 and 2;, which may be disposed in a common housing, are arranged with their axes of maximum re-' sponse mutually at right angles in a horizontal plane and each at approximately 45 to the median plane of the auditorium in which sounds are to be picked up from one or more sources. The signals derived from the microphones 1 and 2 will be considered as right and left directional signals respectively and denoted by V and V and they are amplified respectively by pre-amplifiers 3 and 4 and applied thence by tone and volume control circuits 5' and 6. The directional signals V and V from the

circuits

5 and 6 are applied respectively to further amplifiers 7 and 8 of adjustable gains and the outputs of amplifiers 7 and 8 are then summed in an amplifier 9 and dilierenced in an amplifier 10 to form sum and diiierence signals V +V and V -V respectively. The sum and difference signals are applied respectively to phase shift networks '11 and :12 designed to produce a relative delay of between corresponding components of the respective signals over the full frequency range of the signals and the outputs of the phase shift networks 11 and 12 are then added in an amplifier 13 to produce a composite output signal denoted as V The signal V is an audio frequency signal representative of the waveshape of the sounds picked up by the microphones 1 and 2 and is moreover of a form suitable not only for stereophonic reproduction when used in conjunction with the control signal about to be referred to, but also for monaural reproduction for conventional sound reproduction apparatus. An audio frequency signal representative of the sounds to be transmitted, may however be formed in other Ways, for example directly as the sum of the signals V and V or from separate microphone or microphones.

The signals V and V from the tone and volume con- 7 V and thereforethe description will be confinedto the processing circuit for the signal V The processing circuit for the signal V comprises a high pass filter 14 and a low pass filter 15 which respectively transmit frequencies above and below about 1000 c./s.' The signal components passed by the high pass filter '14 are then applied in parallel to five narrow band channels which are all of the same construction. Only the first of these narrow band channels will therefore be described, and this comprises a filter and amplifier 16 which transmits a narrow band of frequencies centred at 1,300 c./s; The output of thefilter and amplifier 16 is applied to arectifier '17 which derives signals representing the envelope of the frequency components of V passed by the filter and amplifier '16. The output of the rectifier is in turn 7 applied to a filter 18 which removes audio components from the envelope signal, and the envelope signal is then applied by a cathode follower 19 to a filter 20 which is arranged to emphasise high frequency components in the envelope signals fed to it by the cathode follower 19, such components corresponding to transients in the original sound. Consequently a pulse is produced in the output of the circuit 20 in response to a transient in the original sound. The construction of the filter 20 will be described in greater-detail subsequently. The output of thetransient emphasising filter is in turn applied to an elongator 21 which is arranged to elongate pulses in the output of the filter 20 the time constant of the elongator being of the order of 100 ms.

As aforesaid the other narrow band channels which receive signals from the high pass filter 14 are of the same construction as that described but the initial filteramplifiers are arranged, as indicated, to pass narrow bands of frequencies centred on 2050, 3500, 6000, 9200 c. /s. respectively. The individual envelope signals produced in this narrow band channel, after transient emphasis as described, arethen fed by a series of summing resistors 22 to 26 to an amplifier 27, e

The output of the low pass filter 15 is applied to an amplifier 28 and thence to a suppressed carrier modulator 29 which receives a carrier oscillation from an oscillator 29a. The carrier oscillation has a frequency of, say, 5000 c'./s. and the output of the modulator 29 is applied to a filter and amplifier circuit 30 which is arranged to suppress the upper side band produced in the modulator 29, leaving only the lower side band, the carrier being of course suppressed in the modulator. The lower side band, is then applied to a series of additional narrow band channels which are similar to those fed by the high pass filter '14. Only one of these-additional narrow band channels is illustrated and it comprises a filter and amplifier 31 which is arranged to amplify only a relatively narrow band of frequencies centred on (5000-62)c./s.

, The output of the filter and amplifier 31 is then applied to a rectifier 32 which is followed by a smoothing filter 33 and the output of the filter 33 is an envelope signal corresponding to the side band components which are at present in the output of the amplifier 31. It therefore represents the envelope of a narrow band of frequency components centred on 62 c./s. modulator 29 transposes the audio frequency components in question to a relatively high frequency range and simplifies the problem of removing audio components from the desired envelope signal by means of the filter 33. It is readily possible to provide a filter which will separate frequencies'of the order of 5000 c./s. from the frequency components which represent the envelope signal, whereas this would not be the case if the audio com- Theponents had been retained in their original frequency range centred on 62 c./s. The envelope signal from the filter 33 is then applied to a cathode follower 34, a transient emphasizing filter 35 and an elongator 36'which are respectively similar to the components d9, 20 and 21 previously referred to. The other narrow'band channels which receive signals from the low pass filter 15 via the

components

20, 29 and 31 are of similar construction to that channel which has just been described. However the initial filters and amplifiers are arranged respectively to amplify components in separate frequency bands centred on (5000-120), (5000-330), (5000-500), (5000- 700) and (5000-940) c./s. respectively. The frequencies 62, 120, 330 9200 are substantially linearly separated on a logarithmic scale. The envelope signals from all the narrow band channels fed by the filter 15 are fed by summing resistors 37 to 42 to the aforesaid amplifier The output of the amplifier 27 is fed to a further transient emphasizing filter 43 which is of the same construction as the filters Z0 and 35 and the corresponding filters in all the other narrow band channels. The output of the filter 43 is passed to an elongator 44 which is similar to the

elongators

21, 36 and so on, but has a shorter time constant, for example'of the order of 35 ms. The output of the elongator 44 is in turn applied by way of a resistor 45 to an amplifier 46.

The directional signal V from the tone and volume control circuit 5 is further applied to an amplifier 47 which is arranged to amplify the full range of frequencies in the signal V The output of the amplifier 47 is applied to a rectifier 48 followed by a filter 49 and the rectifier 48 and filter 49 are arranged to derive the envelope of the signal V;,. This envelope is added in desired proportion to the output of the elongator 44 by way of the resistor 50 at the input of the amplifier 46. a For example the

resistors

45 and 50 may be dimensioned to combine the output of the elongator 44 and of the filter 49 in the proportion of 10: 1. The output of the amplifier 46 is the desired processed signal, denoted by V and it is fed to a dividing circuit 51 which receives the corresponding signal V derived by processing the directional signal V The processing circuit for V has not been illustrated. The processing circuits for V and V may be arranged to restrict the frequencies of V" and V";,, to the band from 0 to c./ s.

The dividing circuit produces the control signal this signal being a function of the ratio of V" and V" As indicated the control signal may represent VVIIL V! IL+ VIIR or may be the logarithm of this ratio. The construction of the dividing circuit is however no part of the present invention and will not be described further. It may for example be of the construction described in the specification of United States application Serial Number 689,431.

The control signal and the aforesaid audio signal V may then be fed to a wireless transmitter and broadcast in any suitable Way. For example the control signal may be caused to modulate a subsidiary carrier wave in amplitude, phase or frequency, andthe subsidiary carrier wave may in turn be caused to modulate the carrier wave for the audio frequency signal. Alternatively the audio frequency signal may be recorded on a suitable record.

Modifications may be made in the construction of the processing circuits. For example a filter having a response which rises linearly with frequency may be provided before the

filters

14 and 15 and the amplifier 47, in order to emphasise tones of high frequency in the directional signals V and V relative to tones of lower'frequencies. Alternatively such filters may be providedafter one or more of the

filters

14 and 15 or in a selection of the narrow band channels fed by these filters. Moreover the number of narrow band channels may be different from that described.

FIGURE 2 shows one of the transient emphasising filters and the associated pulse elongator used in the ar- 'r'angement shown in FIGURE 1. It will be assumed that FIGURE 2 illustrates the filter 20 and elongator 21. The filter 20 comprises a capacitor 60 and a resistor 61 which latter is shunted by a diode 62. The capacitor 60 and resistor 61 are connected in series between the input terminals 63 of the arrangement whilst the output of the filter is taken from across the resistor 61 and applied to a cathode follower 64. The diode 62 may be a crystal diode and is provided so as to render the transientemphasising filter sensitive substantially only to rising transients. When a falling transient occurs the diode 62 conducts and produces substantial attenuation. The elongator 21 which is fed from the cathode follower 64 comprises two series resistors 65 and 66 and a shunt capacitor 67, the resistors 65 and 66 being individually shunted by

diodes

68 and 69 respectively. It will be understood that when a rising transient occurs, a pulse of relatively large amplitude is applied to the elongator from the transient emphasising filter. The capacitor 67 is charged rapidly nearly to the amplitude of the pulse by way of the

diodes

68 and 69 and after the impulse ends, the charge in the capacitor diminishes only relatively slowly by way of resistors 65 and 66 so as to produce elongation of the impulses at the output terminals of the elongator 70. In practice therefore if a series of transients occur within an interval which is comparable with the time constant of the elongator 21, the output signal from the elongator consists of an elongated pulse the leading edge of which corresponds to the initial transient of the series but it does not exhibit individual pulses corresponding to the following transients in the series. In a practical form of the invention, the capacitor 60 has a capacity of 0.2 f. and the resistor 61 has a resistance of 75,000 ohms, making the time constant of the transient emphasising filter (when the rectifier 62 is not conducting) equal to 15 ms. In this example the time constants of all the emphasising filters including the filter 43 are equal. Moreover the time constant of the resistors 65 and 66 and the capacitor 67 is 100 ms. in the case of the

elongators

21 and 36 and the corresponding elongators in the other narrow band channel. However the time constant of the elongator 44 in the common output channel for each directional signal is 35 ms. The time constants may of course be adjusted to give optimum directivity in any particular case and in some cases the time constants of the transient emphasising filters and elongators in the different narrow band channels may be adjusted in dependence upon the frequencies transmitted by the respective channels. For example shorter time constants may be used for the transient emphasising filters, and/ or the elongators used in the high frequency narrow band channels.

The reproducer for giving stereophonic sound reproduction in response to the audio frequency signal and the control signal derived from the apparatus illustrated in FIG- URES l and 2 may, for example, have the construction described in the specification of the aforesaid co-pending patent applications. Such a receiver comprises two spaced loudspeakers which can be referred to as the left and right loudspeakers respectively and in addition to the customary receiver circuits, includes means for separating the audio frequency signal and the control signal and by process involving multiplication forming two audio signals which are respectively proportional to VI! VII (W a) and (w K m) These two audio frequency signals are in turn applied to the left and right loudspeakers so that stereophonic reproduction is achieved. The invention is not however restricted to the form of the reproducer nor is it restricted to a control signal having the specific form indicated above. In a modification of the invention separate partial 6 control signals may be formed for each of the narrow frequency bands into which the direction signal V and V are divided. Each such partial control signal having for example the form VIIL VPIL+VIIR where v is used to indicate the processed signal derived from the particular narrow frequency band under consideration. In such a case the partial control signals may be transmitted in such a way that their individual identity is preserved. A sound reproducer for receiving a transmission including such partial control signals may then include filter means for dividing the audio frequency signal V into narrow frequency bands corresponding to the individual control signals, and two audio frequency signals would then be produced for each narrow frequency band, the resultant narrow band audio frequency signals for one of the loudspeakers being added together and fed to that loudspeaker, and those for the other loudspeaker dealt with in a similar way.

What I claim is:

l. A stereophonic sound transmission system comprising means for deriving from sound a plurality of audio signals which diifer in amplitude one from another in dependence upon the direction or position of the source of the sound, a-plurality of groups of processing channels, one group of processing channels for each of said audio signals, filter means for selecting the components of each audio signal into a group of different frequency bands, one frequency band for each channel of a group of channels, and for applying the components in the different frequency bands to their respective channels, each channel including means for processing the frequency components therein, and means responsive to the output signals from the plurality of groups of channels to derive a control signal representative of the direction or position of the source of the sound, and means for transmitting said control signal and an audio signal representing the waveshape of said sound.

2. A system according to claim 1 wherein said processing means in each channel comprises means for deriving a signal representing the envelope of the frequency components in the respective channel.

3. A system according to claim 1 wherein said processing means in each channel comprises means for individually rectifying frequency components in the respective channel to derive a signal representing the envelope of said frequency components, and means for individually emphasising high frequency components in the respective envelope signal to produce a pulse in response to a transient in said frequency components.

4. A system according to claim 3 wherein said transient emphasising means are more responsive to rising transients than falling transients.

5. A system according to claim 3 wherein said processing means further comprises means for elongating pulses produced in response to transients.

6. A stereophonic sound transmission system comprising means for deriving from sound two audio signals which differ in amplitude one from another in dependence upon the direction or position of the source of the sound, two groups of processing channels, one group of processing channels for each of said audio signals, filter means for selecting the components of each audio signal into a group of different frequency bands, one band for each channel of a group of channels, and for applying the components in the different frequency bands to their respective channels, each channel including means for processing frequency components therein, means for superimposing the output signals derived from one group of channels, means for superimposing the output signals derived from the other group of channels, and further processing means for processing said two groups of superimposed output signals to derive a control signal representative of a the direction or position of the source of said sound and a independent of the loudnessof said sound, and means for transmitting said control signal and 'an audio signal representative of the waveshape of said sound.

7. A system according to claim 6 wherein said processing means in each channel comprises means for individually rectifying'frequency components in the respective channel to derive a signal representing the envelope of said frequency components, and means for individually emphasising high frequency components in the respective envelope signal to produce a pulse in response to'a transient in said frequency components.

8. A system according to claim 7, said processing means in each channel further comprising means for elongating pulses produced in response to transients.

9. A system according to claim 6 wherein said further processing means for processing said two groups of superimposed output signals comprises means for individually emphasising high frequency components in each of said groups of superimposed output signals to provide a pulse in response to a transient in the respective group of superimposed output signals, means for elongating pulses so produced, and means for deriving said'control signal in response to said elongated pulses.

10. A system according to claim 8 wherein saidfurther processing means for processing said two groups of superimposed output signals comprises means for individually emphasising high frequency components in each of said groups of superimposed output signals to provide a pulse in response to a transient in the respective group of superimposed output signals, means for elongating pulses so produced, and means for deriving said control signal in response to said elongated pulses, said means for elongating pulses in each of said channels being arranged to produce substantially greater pulse-elongation than said latter elongating means.

11. A system according to claim 6 wherein said further processing means comprises means for combining each group of superimposed output signals with a portion of the audio signal from which the respective group of superimposed output signals is derived.

12. A system according to claim 6, wherein said processing means includes means for deriving, as said control signal, a signal which is a function'of the ratio of the two processed groups of superimposed output signals.

References Cited in the file of this patent UNITED STATES PATENTS 7 2,249,606 'Friebus July 15, 1941 2,481,576 De Boer Sept. 13, 1949 2,761,897 Jones Sept. 4, 1956 2,852,604 MacCutcheon Sept. 16, 1958 2,874,221 Dauguet Feb. 17, 1959