|Publication number||US2836662 A|
|Publication date||May 27, 1958|
|Filing date||Aug 16, 1955|
|Priority date||Aug 18, 1954|
|Publication number||US 2836662 A, US 2836662A, US-A-2836662, US2836662 A, US2836662A|
|Inventors||Barnet Vanderlyn Philip|
|Original Assignee||Emi Ltd|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (1), Referenced by (8), Classifications (7)|
|External Links: USPTO, USPTO Assignment, Espacenet|
M83127, 1958 P. B. VANDERLYN 2,836,662
ELECTRICAL SOUND TRANSMISSION SYSTEMS Filed Aug. 16, 1955 v 2 Sheets-Sheet 1 y 1958 P. B. VANDERLYN 4 2,836,662
ELECTRICAL SOUND TRANSMISSION SYSTEMS Filed Aug. 16, 1955 2 Sheets-Sheet 2 FIG. 3.
United States Patent Ofice 7 2,836,662 ELECTRICAL SOUND rnANsMtssIoN SYSTEMS Philip Barnet Vanderiyn, Ickenham, England, assignor to Electric & Musical Industries Limited, Hayes, England, a company of Great Britain Application August 16, 1955, Serial No. 528,583 Claims priority, application Great Britain August 18, 1954 Claims. (Cl. 179-100.1)
This invention relates to electrical sound transmission systems wherein electrical signals representative of sounds received at a plurality of microphones and transmitted to a plurality of loudspeakers are so modified that the sounds heard from the loudspeakers will convey to a listener positional impressions of the sound sources, similar to those which would be produced naturally if the listener were present during the emission of the original sounds. Such a system is commonly referred to as a binaural system and it may include means for recording and reproducing the sounds and/ or means for transmitting and receiving the sounds by wire or radio.
In one known type of binaural transmission system as described in U. S. patent specification No. 2,093,540 the apparent spatial distribution of the sound sources is derived on the assumption that a listener localises a source by means of phase differences existing at his ears at low frequencies, and by means of intensity differences at high frequencies. Low frequencies may be defined approximately as those frequencies for which the effective ear spacing is less than one half of the wavelength and high frequencies as those above this limit. The frequency at which the human brain changes over from phase discrimination to intensity discrimination has been stated to occur in the region of 700 cycles/ sec. but depending on the circumstances and from person to person this changeover frequency may vary within wide limits.
As described in the aforementioned patent specification, phase differences at low frequencies can be produced at the listeners cars by the vector addition of the components at each ear due to each of two or more spaced loudspeakers, the phase diiferences being produced at the ears as a result of relative amplitude differences between the loudspeakers. At high frequencies intensity differences can also be produced at the ears as a result of relative amplitude differences between the loudspeakers.
It has been realised that for any given arrangement of two or more loudspeakers the relative amplitude difierences required to reproduce the effect of sound arriving from a given angle will not be the same at both high and low frequencies, and it has been proposed to overcome this difiiculty by employing a multiplicity of loudspeakers,
and distributing the transmitted signals to the loudspeakers on an appropriate frequency selective basis so that the effective spacing of the loudspeakers is different at low and high frequencies.
The object of the present invention is to overcome the aforementioned difficulty without the use of a multiplicity of loudspeakers as heretofore. and the frequency selective networks associated therewith.
A further object of the invention is to provide an improved binaural sound transmission system by means of which more accurate binaural effects can be produced.
In a two-channel binaural system of the type described in the aforementioned patent specification, the impulses from two microphones are modified to produce initial summation and difference signals by transmitting said impulses through so-called summation and difierence channels which are arranged so that the resulting current derived from the summation channel represents half the sum (that is, the mean) of the original microphone currents, while the current derived from the difierence channel represents half the difierence of the original microphone currents. Alternatively, the microphone system may be such that two outputs obtained therefrom are effectiveiy equivalent to initial summation and difference signals. In either case it is known that if the initial summation and difference signals are modified so as to increase or decrease the amplitude of the difference signals as a whole relatively to the amplitude of the summation signals, and the modified signals are again summed and differenced, an angular magnification or reduction will be produced in terms of the original obliquity of the sound source with reference to the microphones. In other words, when sets of signals modified in the manner described are reproduced by spaced loudspeakers the listener will obtain the impression of the sounds arriving from a greater or smaller angle as the case may be, as compared with the original angle of the sounds relatively to the microphone system by which they were picked up..
According to the present invention there is provided a binaural sound transmission system comprising means for deriving two sets of signals with such relative amplitude differences that on applying said sets of signals to spaced sound reproducers a binaural effect is produced, and
. means whereby the relative amplitudes of the high and low frequency components of said sets of signals are modified thereby to produce a frequency sensitive angular magnification or reduction so as to reduce inaccu racies in said binaural effect.
In order that the invention may be clearly understood and readily carried into effect, an embodiment thereof will now be described, by way of example, with reference to the accompanying drawings, in which:
Figures 1 and 2 illustrate networks for insertion into the difference and summation channels, respectively, and
Figures 3 and 4 are explanatory diagrams.
Figure 3 illustrates diagrammatically a pair of spaced loudspeakers 1 and 2 arranged symmetrically with reference to a medial plane AB. Let it be assumed that the loudspeakers are fed with two sets of signals derived from a binaural transmission system according to the aforementioned patent specification, said signals being in phase but ditfering in amplitude in order to convey to a listener at P positional impressions corresponding to an original sound source. Thus the listener at P will estimate the position of a virtual source at an angle 0 to the medial plane AB in accordance with the amplitude differences of the sounds reproduced by the loudspeakers 1 and 2.
If the sets of signals fed to the loudspeakers 1 and 2 are'restricted firstly to the low frequency components below the aforementioned changeover frequency of about 700 cycles/sec. and then to the high frequency components above this frequency, the apparent angle 0 of the virtual source for low and high frequency components respectively can be plotted against the amplitude ratio r of the reproduced sounds 'as shown by the curves of Figure 4, in which values of'r and 0 to suitable scales are represented along the axes XOX' and YOY' respectively. For thelow frequency components the curve obtained approximates to the full-line AOA' passing through the origin 0, but for the high frequency components the dotted line curve is obtained which intersects the low' duce signals derived from a virtual point source having Patented May 27, 1953.
duced or substantially eliminated so that the dotted-line curve of Figure 4 will full-line curve. v 7
According to one embodiment of the invention, as illustrated in Figure, 1, the initial difference signals are fed be in closer conformity with the through a channel including a bridged T-network' which comprises a shunt arm including the series combination of resistor R and capacitor'C and series arms including the resistorsRg and R the resistor R of the shunt arm being connected to the junction of resistors R and R The series arms are bridged by an inductance L This type of equalisersection'attenuates the high frequencies relatively to the low frequencies but in addition it introduces an undesired phase shift in that portion of the frequency spectrum in which the reactances of L and C are comparable with the characteristic impedance of the circuit, and this phase shift produces undesired effects in subsequent summing and. differencing operations. To correct this relative phase shift, a pair of all-pass bridged 'T-networks are connected, one in each channel, to introduce equal and opposite relative phase shifts in a known manner, the network L T ,.C being inserted in the'diiference channel as shown in Figure 1 and the network L 7T C being inserted in the summation channel as shown in Figure 2.
It will be appreciated that the required frequency sensitive angular magnification or reduction could be attained as a result of the insertion of a suitable network in eitherthe summation or the difference channel. However, if the initial summation signals are modified in a frequency sensitive manner an overall frequency discrimination is "applied to the system. Some frequency discrimination will occur as a result of the modification of the initial difference signals but in general to a smaller extent.
As a result of the modification of the relative amplitudes ofthe high and low'frequency components of the initial difference signals or summation signals as the case may be, it is possible to produce, by further summing and differenencing operations, a frequency sensitive angular magnification or reduction. Thus, by suit-ably selecting the extent of the frequency sensitive modification of the difference or summation signals the disparity-referred to in connection with Figures 3 and 4 can be largely overcome, and two sets of signals can be derived for energising spaced loudspeakers so as to convey to a listenerthe effect of both low and high frequency soundsarriving more nearly from the same angle. 7 a
In a two-channel binaural system according to the invention the two outputs from a particular type of microphone system may be effectively equivalent to initial summation and difference signals respectively, and -in such a case the two outputs can be modified in the man? ner describedvbefor'e being subjected to summing and differencing operations to obtain the required frequency sensitive angular magnification or reduction. If, however, the microphone system is not of this type, preliminary summing and differencing'operations are required to produce the initial sum and difference signals and the. signals thus obtained are subjected to frequency sensitive modification according to the invention before being subjected to further slunming and differencing operations so as to obtain the required two sets of signals.
The present invention provides means.
A binaural system according to the invention is parti cularly applicable to arrangements in which only a single pair of loudspeakers is intended .to be employed. The invention is, however, not to be considered as limited to the case in which only two loudspeakers are used since the invention, in general, allows the required number of loudspeakers to be reduced and in particular circumstances it may be desirable to utilize more than two loudspeakers Whilst still obtaining the advantages of the'present invention.
Binaurai sound transmission systems of the type described areparticularly suitable for use in the production of binaural sound records such as twin-track magnetic tapes carrying complementary signals along the respective tracks thereof. To produce such binaural, sound records it is necessary to employ a sound recording system comprising a pair of discrete recording channelsfor recording the two sets of signals derived from the binaural transmission system. It will be appreciated that by employing a binaural transmission system according to the present invention to feed the two sets of signals to the discrete recording channels, the resulting sound records will be improved in that when reproduced bysuitable apparatus they will convey to a listener more faithful positional impressions corresponding to the distribution of the original sound sources.
What I claim is: V
1. In a binaural sound transmission system including means for deriving two sets of in-phase signalswith such' relative amplitude differences that on applying said sets of in-phase signals to spaced sound reproduccrs a binaural effect is produced, the provision of frequency selective means to efiect modification of the relative amplitudes of the high and low frequency components of said sets of in-phase signals thereby to produce a frequency sensitive angular magnification or reduction so as to 'reduce inaccuracies in said binaural effect. v i
2. A binaural sound transmission system according to claim 1 including means for deriving'initial summation and difference signals and means for summing and differencing said initial summation and difference signals thereby to obtain said two sets of in-phase signals, said frequency selective means effecting modification of said initial summation and/or difference signals.
3. A binaural sound transmission system according to claim 2 whereinsaid frequency'selective means reduces the amplitude of the high frequency components of said initial difference signals as compared with the amplitude of the low frequency components thereof.
4. A binaural sound transmission system according to I claim 3 wherein said frequency selective means comprises an attenuator in the form of a T-netwo-rk having a shunt arm includingresistance and capacitance and series arms each including resistance, said series arms being bridged bya path including inductance.
5. A binaural sound transmission system according to claim 4 including all-pass networks through which the modified initial: difference signals and the initial summation signals are fed thereby to produce equal and oppo-- site relative phase shifts to correct for any undesired phase shift introduced by said attenuator.
References Cited in the file of this patent "UNITED STATES PATENTS Blumlein Sept. 21, 1937
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US2093540 *||Dec 13, 1932||Sep 21, 1937||Emi Ltd||Sound-transmission, sound-recording, and sound-reproducing system|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US3088997 *||Dec 29, 1960||May 7, 1963||Columbia Broadcasting Syst Inc||Stereophonic to binaural conversion apparatus|
|US4356349 *||Mar 12, 1980||Oct 26, 1982||Trod Nossel Recording Studios, Inc.||Acoustic image enhancing method and apparatus|
|US5692050 *||Jun 15, 1995||Nov 25, 1997||Binaura Corporation||Method and apparatus for spatially enhancing stereo and monophonic signals|
|US7146010||Jun 19, 2000||Dec 5, 2006||Embracing Sound Experience Ab||Two methods and two devices for processing an input audio stereo signal, and an audio stereo signal reproduction system|
|US7702111||Jul 16, 2004||Apr 20, 2010||Embracing Sound Experience Ab||Audio stereo processing method, device and system|
|US8620010||Apr 19, 2007||Dec 31, 2013||Embracing Sound Experience Ab||Loudspeaker device|
|US20060188101 *||Jul 16, 2004||Aug 24, 2006||Fredrik Gunnarsson||Audio stereo processing method, device and system|
|US20090175472 *||Apr 19, 2007||Jul 9, 2009||Embracing Sound Experience Ab||Loudspeaker Device|
|U.S. Classification||381/1, 381/98, 369/175, 369/86|