DE2659757C2 - Device for recording and reproducing spatial sound events using two-channel spatial sound signals - Google Patents

Description

The main patent relates to a device for recording and reproducing spatial sound events according to the preamble of claim 1, see DE-PS 26 16 665 IM.

According to the teaching of the main patent [1], as well as according to "Advances in Acoustics", DAGA 76, Heidelberg, p. 483 [2] and NTG-Tagung, »Hörrundfunk«, Düsseldorf 1976 [3], spatial sound events are caused by a facility transmitted from a directional microphone with a rapidly rotating directional characteristic on the Admission side consists. On the playback side, the voltage of the "rotating microphone" becomes a loudspeaker arrangement fed, in which a switch is arranged in front of each loudspeaker, each synchronous with the rotating microphone characteristic is switched (acoustic stroboscope).

The voltage of a rotary microphone contains the direction-independent low-frequency component (base signal), which is supplied by the »spherical microphone« and the signal component Ä 'of the rotating figure eight microphone with the

Information about the direction of arrival of the ship. If the rotary microphone has a directional characteristic of the first order and rotates at an orbital frequency of 38 kHz, its total signal (spatial tone signal B + find a frequency range from 30 Hz to 53 kHz.

The spatial tone signal basically consists of three low-frequency microphone voltages, which are optimally are coded [I].

ω The decomposition of the Räümtönsignäls in the Känälmultiplex leads z. B. on three low-frequency voltages (B, /? + ^ R-φ) according to Fig. La, of which the two, which are derived from the room signal R ' , differ from each other in their phase position by twice the sound angle of incidence. The three audio frequency voltages can be shown in the phasor diagram (FIG. 1 a). This representation makes it clear that the sound angle of incidence φ determines the phase position of the three voltages to one another, that the direction of sound incidence is transmitted undistorted when the linear phase and amplitude distortions of these three sound voltages are the same at the same frequency. This is important if, for example, a pre-emphasis or dynamic compression of the spatial sound signal is to be carried out or if phase distortions (e.g. in the case of magnetic recording) have to be feared in the spatial signal range R '. In such cases we recommend

It is not possible to transmit or store the surround sound signal in channel multiplex.

The spatial tone signal can be stored in both frequency and channel multiplexing. at Tape recordings will be preferred to channel multiplex because multi-track devices with the same phase and frequency responses in all channels are common in studio technology today. In the case of record storage you only have two low frequency ranges up to 20 kHz available. The record transfer of high Frequencies (up to 53 kHz) and thus the transmission of the entire spatial tone signal is possible today, but unsatisfactory because this method is too expensive and prone to failure playback devices (pickups, Pickup needle) and because the scanning of higher frequencies (> 20 kHz) due to increasing acceleration forces causes higher wear of the record.

The object of the invention is therefore to use these three signals present in the low frequency position in a rotary microphone to process in such a way that the three signals result in two low-frequency signals for their recording or transmission, only two low-frequency channels are necessary.

This object is achieved according to the invention by what is specified in the characterizing part of claim 1 Characteristics.

Advantageous refinements of the invention are characterized in the subclaims The invention is explained with reference to the drawing figures:

Fig. La For the formation of three low-frequency signals (channel multiplex) from the signal of a microphone with a rotating characteristic, which is formed from the superposition of a fixed omnidirectional characteristic (B) and a rotating figure-of-eight characteristic (R ')

F i g. 1 b To form the channel voltages Kf , Kj , respectively. Ka, Kb

F i g. 2a Generation of the low frequency signal of the nth loudspeaker of the loudspeaker configuration from the channel voltages Kf, Kf with the help of modulators (coherent demodulation)

F i g. 2b Generation of a signal in spatial pitch from the channel voltages Kf, (Kj and further processing with the help of counter-rotating switch demodulators

F i g. 2c device according to FIG. 2b with additional reproduction of a vertical microphone plane according to FIG the main patent [1]

F i g. 3a, b Some loudspeaker allocation functions depending on the parameters K, β

The tone voltages of FIG. 1 b are according to the type of F i g. 1 c summarized, so that the voltages Kf, Kj arise: ~~

Kf = Bcos a + R cos (a - φ) + P

(1) Kj = B cos a + R cos (a + φ) - P

mü

P 19 kHz pilot tone

a = wt + α instantaneous phase of the audio frequency

The spatial signal components R- ₉ and R ₊₉ may have been obtained from the spatial tone signal, as sketched in FIG. La. P is the 19 kHz pilot tone, which synchronizes recording and playback modulation ([1J [3]) and z. B. is transmitted with about -20 dB under full modulation. Kf and Kj are treated like the AB signals of conventional two-channel stereophony (tape, record). After Summenbildupg (Kf + Kj) , the two pilot tone components cancel each other out; after the difference is formed, they add up. Splitting the pilot tone has the advantage that if the phase differences between the two transmission channels are small, an equalization takes place

The two Tonspannungen Kf and Af ⁺ ₂ can be transmitted on two channels and stored. Before their two-channel integration stereophonic reproduction, their linear combination must be carried out again; When scanning records, this is known to happen through the 45 ° position of the scanning systems. By linear combination it follows:

Af (i +2, = K * + Af ₂ ⁺ = 2 B cos α + 2 R cos α · cos o>

Kt + Kt

= B ■ cos α ■ (1 + K cos φ) _5Y

(2) £ <i-2) = Kt -! ^ ^{= R} {cos (α-?)) - cos (α + φ)} + 2P_

A ⁺ + K ⁺

= R sin α ■ sin φ + P)

with K = R.'B
If the pilot tone is filtered out, the level difference between these two channel voltages is:

ΛΙ -VM 1 + AfCOSfl?

^ '+ W.-2) = 20lg (3)

If the pilot tone P is only added to the voltage Kf , it is extinguished during playback for a listening position exactly between the loudspeakers.

Both channel voltages JjC ₁ *, Kf always show a phase shift of 90 ° from one another, corresponding to the shift between the sine and cosine functions in equation 2. Their voltage ratio is only dependent on the direction of sound incidence for a given K (recording parameter). Except for the 90 ° phase shift of both channels, this technology is compatible with intensity stereophony. For K = 1, for example, a characteristic curve that is symmetrical around φ = 90 ° results, which at φ = 10 ° or 170 ° achieves level differences of ± 20 dB. In the angular range of ± 60 °, the characteristic curve is remarkably straight. The rear listening scene is of course mirrored in the front. For larger values of K (e.g. 1.5 or 2) the listening scene becomes narrower ίο and its center moves to smaller angles, e.g. B. 70 °. Instead of one, there are three reflections at the same time.

The constant phase shift of 90 ° probably causes a certain "spatiality", especially of the center impression, which - if this is not desired - can be avoided by using phase shifters with a constant phase for reverse rotation. The single-channel reproduction of the tension (Kf - / C ₂ *) is compatible with the two-channel stereo reproduction, because it is greatest in the middle of the scene (φ - 90 °), which is why the two tone tension from Ka and Kb should be formed (Fig. 1 d):

V. -If * SS. Λ- IS- - IS- - K * »ink-analia * WioHoroahp

Linear combination: (4) M Kb Kf K _a -Kb = Kt + K}

with

Ka = Bcos a + R [a — φ) Kb = -Bcos a — R (a + φ)

The reproduction of the direction of sound incidence with the aid of shell modulators in front of each loudspeaker analogous to the method described in [1] is achieved by the circuits according to FIG. 2 enables each of the voltages Kf, K; first of all, after the pilot tone P has been filtered out, the rotary cha is at the rotational frequency Ω characteristic multiplied, whereby according to [1] the pilot tone is used to synchronize the modulators. This can e.g. B. done in a known manner with a PLL circuit. At the output of the modulators, the signals are in Room signal position available, d. H. as required by the "eidophonic rendering". Unlike the Signal from the rotary microphone [1], parts of the base signal are now also included in the room signal range. Both Voltages pass through filters to directional modulators that are assigned to each loudspeaker channel [I]:

Each loudspeaker channel is supplied with the voltages Kt and Kt transposed into the room signal position via two channel demodulators modulated in opposite directions. The channel modulators in FIG. 2a have been assumed to be a four-quadrant multiplier for the sake of simplicity. In addition, the sum of the voltages Kf, Kf is added to each channel. For the loudspeaker occupancy function (it describes the attenuation of the sound voltage as a function of the sound incidence angle φ) it finally follows after low-pass filtering (f _g = 15 kHz):

40

{{Lit x Kt) x (12 / + φ) - [ΤΡ] - B cos (a - ψ) + R cos (fl + (φ - φ)) (iit χ Κ;) χ (Ut + ψ) - [TVj - B cos (α - ψ) + R cos (α + (ψ - φ))

⁴⁵ C.

2 B cos α ■ cos φ + 2 R cos α cos (ψ - φ)

1 (£ Γ + Kj = 2Β cos α + 2R cos α cos φ

1/2 total sum = B cos a \\ + K cos φ + cos φ + K cos (φ - ψ) \

U = Bcosa [l + K cos φ + cos φ + K cos (φ - ψ) (5a)

a = t + λ instantaneous phase of the low frequency

φ direction of sound incidence -φ »channel angle«, ie the angle at which the loudspeaker in question is set up

U = Bcosa \\ + K cos φ + cos (ψ - β) + K cos {φ - {φ - β))} (5b)

60

with /? phase shift of the spatial tone modulator.

F i g. 2b shows a circuit variant which can be used for room demodulation with switch modulators according to [1] and which leads to equation 5b if, in addition, the 38 kHz oscillations are transmitted to the two four-quadrant multipliers via a "forward" or "reverse" phase shifter l ± ß \ according to Eq. 5b are supplied.

The pressure in brackets in equations 5a / b each describe the occupancy function of the loudspeakers. The loudspeaker occupancy function indicates the weighting under which the loudspeaker in question emits its signal. An example is shown in Fig. 3a for β = 0. In contrast to the method described in [1]

(Eidophonie) the loudspeaker occupancy function is now also dependent on the channel angle, so that the reproduction of the directions of sound incidence is dependent on the directions in the reproduction room; for example, nothing at all is reproduced in the shawl incline direction φ = 180 ° at K = 1. In addition, the listening scene is slightly contracted towards the middle.

For full compatibility of two-channel playback using switch demodulators, it is desirable that the middle of the listening scene is the same in both modes of reproduction.

In the case of monophonic and two-channel playback, the center is identified by φ = 90 ". With eidophonic playback according to Eq. 5a, there would then no longer be any amplitude symmetry at the center of the listening scene (φ = 90 °) (see FIG. 3), ie right and left Sound sources positioned in the middle would be reproduced with different loudness. This deficiency could easily be countered by appropriate placement of the sound sources, but this is not possible for the room echoes Phase β of the 38 kHz oscillations that are fed to the room tone modulators changes, if, for example, a phase shift of 90 ° is provided, the symmetry of the listening scene by 90 ° is fully preserved (see Fig. 3b).

The two-channel transmission proposed here can also be used if, according to [1], two planes are scanned (horizontal and vertical) and reproduced in plane time division multiplex, since no special restrictions were made for the voltages Kf, IQ when encrypting the room signal '. A short-term memory is to be connected to each of the outputs of the two room tone modulators, the outputs of which are alternately connected to the outputs of the room tone modulators via the level switch with the direction modulators of the two levels, see FIG. 2c. The demodulator effort increases naturally.

For this purpose 4 sheets of drawings

Claims (6)

Patent claims: 1. Device for recording and reproducing spatial sound events with a directional microphone arrangement with moving characteristics, consisting of two gradient microphones of the first order, which are arranged at an angle of 90 ° to each other and their Output signals in amplitude modulators with sinusoidal ones that are phase-shifted by 90 ° Vibrations of the same frequency are modulated, the frequency being so high that the resulting Sidebands are outside the listening area, an adder that combines the output signals of the amplitude modulators into a spatial signal, and an omnidirectional microphone, which is arranged closely adjacent to the gradient microphones, and a base signal supplies, and with an arrangement of loudspeakers with moving radiation characteristics, which is synchronized with the recording characteristics via a pilot tone,
according to patent 26 16 665, characterized in that that the sidebands of the spatial signal are transported through filters and coherent demodulators into a first (R-φ) and a second low-frequency signal (R ₊₉ ) , that a first channel signal (Kt) is generated by adding the first low-frequency signal (R-φ) to the base signal (B) and a second channel signal (Ki) is generated by adding the second low-frequency signal (R ₊₉ ) to the base signal (B) , that On the playback side, the two channel signals (Kt, (Kj) are transported back in the spatial signal position, and that the loudspeaker signals are obtained from the back-transported signals. 2. Device according to claim 1, characterized in that the synchronization of the recording and Modulators on the playback side necessary pilot tone (e.g. 19kHz) about 2OdB under full modulation either only one of the channel signals is added, or both in antiphase. 3. Device according to claim 1 or 2, characterized in that zj; r reproduction of the two low-frequency channel signals according to FIG. 2 these are initially transposed into the room signal position by modulation with the rotation frequency, so that these components in the room signal position are combined with components of the channel signals in the low frequency position, whereupon the two resulting overall signals are demodulated and fed to the associated loudspeakers. 4. Device according to claim 1 or 2, characterized in that the channel signals according to Eq. 1 at the Playback duro «Addiüjn or subtraction (linear combination) can be reshaped, whereby their two-channel, Intensity stereo reproduction becomes possible. 5. Device according to Ansp cattle 1,2 or 3, characterized in that the room signal from the scanning comes from two recording levels and, according to the level multiplex method, the two decoders each Playback level is fed. 6. Device according to claim 1, 2 or 3, characterized in that one of the channel signals is applied so that sound events from the middle of the listening scene (φ = 90 °) are reproduced the loudest, so that the listening scene is reproduced with one-channel, two-channel or playback with Schaltdemodulatorcn matches (compatibility).