JPS628999B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a stereophonic sound reproducing apparatus that can obtain a listening effect equivalent to that obtained when reproducing a sound from a headphone when a binaural signal such as a dummy head microphone pickup signal is reproduced by a speaker.

Binaural signal sources, such as dummy head microphone sound collection sources, were originally created for the purpose of being listened to through headphone playback, and as is, they are not suitable for listening through speaker playback. In fact, if binaural signals were played directly through speakers and listened to, the sound would not have the same sense of spaciousness as when played through headphones, and the sound image would become blurred and the sound would have a peculiar tone.

These auditory phenomena appear because of so-called spatial crosstalk, in which binaural signals, which should originally be supplied to both ears of a listener independently, are mixed in the space when the sound is reproduced by speakers. This is because spectral changes in the sound due to hearing are added twice.

SUMMARY OF THE INVENTION The present invention has been made in view of the above-mentioned drawbacks, and an object of the present invention is to provide a stereophonic sound reproducing device which eliminates the above-mentioned drawbacks and can provide a listening effect equivalent to that obtained when playing back a binaural signal using a speaker. The present invention will be explained below with reference to Examples.

FIG. 1 is a block diagram of one embodiment of the present invention.

In Figure 1, 1 is a dummy head microphone, and 2
Channel output signals a and b are branched through an equalizer circuit 2 into three outputs c, d and e, and the outputs c, d and e are respectively applied to three speakers 3, 4 and 5 arranged around a listener 6, respectively. and configure it to play.

The distances between the speakers 3, 4, and 5 and the center of the listener's 6 head are all the same, and the speakers 3 and 4 are positioned symmetrically in front of the listener 6.
is placed behind the listener 6.

Now, the acoustic transfer functions from the input terminals of the speakers 3, 4, and 5 to the external ear inlet of the listener 6 are expressed as A(jω), B(jω), and C, respectively.
(jω), and A(jω) is each speaker 3.
Let B(jω) be the acoustic transfer function from the input terminal of the speaker 3 to the left external ear inlet of the listener 6, and from the input terminal power of the speaker 4 to the right external ear inlet, and B(jω) is the acoustic transfer function from the input terminal of the speaker 3 to the left external ear inlet of the listener 6. Let C(jω) be the acoustic transfer function from the input terminal of the speaker 4 to the right external ear inlet of the listener 6, and from the input terminal of the speaker 4 to the left external ear inlet of the listener 6, respectively.
Let the acoustic transfer function be . Further, the acoustic transfer function between the center position of the head of the listener 6 and the input terminals of the respective speakers 3, 4, and 5 (with the head excluded) is P(jω).

Therefore, the equalizer circuit 2 inputs the left channel signal a=L(jω) to the dummy head microphone 1, and calculates the transfer function 1/A'(jω)=P(j
ω)/A(jω) (hereinafter, the electrical circuit network is simply referred to as a circuit network) 7, and the right channel signal b=R of the dummy head microphone 1.
(jω) as an input, a circuit network 8 having a transfer function 1/A′(jω), an adder 9 for adding left and right channel signals a and b, and an output signal of the adder 9 as an input, and transfer Function 1/C′(jω)=P
(jω)/C(jω) and a transfer circuit D(j
Network 1 with ω)=B(jω)/A(jω)
1 and a phase inversion circuit 12 which receives the output signal of the circuit network 11 as input, and the output signals of the circuit networks 7 and 8 and the phase inversion circuit 12 are respectively c, d and e, and the output signals c, d and e are are output to speakers 3, 4, and 5, respectively.

Here, the transfer functions A'(jω) and C'(jω) are head diffraction transfer functions, and the transfer function D(jω) is an interaural difference function, which each depends on the direction of incidence of the sound on the head. It is stipulated that

Examples of the measured values of the transfer functions A'(jω), C'(jω), and D(jω) are shown in Figure 3 a and b, and Figure 3 a is |A'(jω)| ｜
C′(jω)|, the horizontal axis represents frequency, and the vertical axis represents level (dB). It can be seen from FIG. 3a that 1/A'(jω) can be replaced by a peak-dip filter, and 1/C'(jω) can be replaced by a dip filter. Further, Fig. 3b shows the amplitude characteristics and delay time characteristics of |D((jω)|, respectively, where curve E shows the amplitude characteristics and curve F shows the delay time characteristics. From curves E and F, the transfer function D (jω) can be replaced by a cascade of a low-pass filter and a delay circuit.

Therefore, in the equalizer circuit 2 shown in FIG. 2, the circuit network 7 has a peak
In the dip filter 13, the network 8 is a peak-dip filter 14, the adder 9 is a cascade circuit of an adder 15 and an inverter 16, the network 10 is a dip filter 17, and the network 11 is a low-pass filter 18 and a delay circuit 19. It can be implemented by replacing each with a cascade circuit.

Therefore, the dummy head microphone signal R(j
ω) and L(jω) from speakers 3, 4 and 5
When reproduced by , the sound pressure at the left and right external ear inlets of the listener 6 will be as follows.

In other words, the sound pressure R _S at the outer ear inlet of the right ear is R _S = Output of speaker 3 + Output of speaker 4 + Output of speaker 5 = [L(jω)/A'(jω)] ・B(jω) + [R (jω)/A′(jω)]・A(jω)−[R(jω)+L(jω)] [D(jω)/C′(jω)]×C(jω)=L(jω)・D (jω)・P(jω) +R(jω)・P(jω)−P(jω)・D(jω) [R(jω)+L(jω)] =R(jω)・P(jω)[1− D(jω)] …(1). Similarly, the sound pressure L _S at the external ear inlet of the left ear
is L _S =L(jω)・P(jω) [1-D(jω)]
...(2) becomes.

Here, since |P(jω)·[1-D(jω)]|≈constant and does not depend on frequency, it can be seen that the dummy head microphone output signal is transmitted to the left and right ears without being modulated.

As described above, according to this embodiment, the dummy head microphone output signal is passed through the equalizer circuit 2,
When the sound is reproduced by the three speakers 3, 4, and 5, the left and right channel signals are independently reproduced to the left and right ears of the listener 6, similarly to when playing with headphones.

Furthermore, the above effect can be obtained from the left channel signal a of the dummy head microphone 1, as shown in FIG.
is a network 20 with a transfer function 1/A′(jω)
The right channel signal b of the dummy head microphone is transferred from the signal passed through the transfer function 1/A'(j
The subtracter 24 subtracts the signals passed through the circuit network 21 having the transfer function D(jω) and the circuit network 23 having the transfer function D(jω), and reproduces the resultant signal at the speaker 26, thereby transmitting the right channel signal b of the dummy head microphone 1. Zu〓〓〓〓〓
From the signal passed through the network 21 with the function 1/A'(jω), the left channel signal b of the dummy head microphone 1 is transferred to the network 20 with the transfer function 1/A'(jω) and the transfer function D(jω). Even when the signal passed through the circuit network 22 having
The same effect as in the embodiment shown in the figure is obtained. However, the two speakers 26 shown in FIG.
27, there is a drawback that the rear sound image is difficult to stabilize compared to the sound image localized in the front. On the other hand, in the embodiment shown in FIG. 1, one speaker 5 is placed behind the listener 6, so the sound image behind is also stabilized.
This has the effect of enabling balanced 360-degree circumferential localization.

Next, other embodiments of the present invention will be described.

FIG. 6 is a block diagram of another embodiment of the invention.

In this embodiment, the position of the speaker 5 of the embodiment shown in FIG. /E′(j
The signal is replaced with a circuit network having a transfer function of ω)=P(jω)/E(jω), and is reproduced by the speaker 29 through the circuit network 11 and the phase inverter 12.

Here, E(jω) is an acoustic transfer function from the input terminal of the speaker 29 to the left and right external ear inlets of the listener 6.

Normally, when playing binaural signals through speakers as shown in Figure 5, when the listener turns his or her head, the binaural signals are not transmitted independently to the left and right ears, resulting in crosstalk. come.

This is because when the listener turns their head, the transmission characteristics from each speaker to the listener's ears change, which reduces the function of the initially set equalizer circuit, resulting in poor sound localization. It becomes natural.

However, in the case of this embodiment, the crosstalk level when the listener turns his or her head is as follows.

For example, when the viewing angle of speakers 3 and 4 is 120 degrees and listener 6 turns his head ±30 degrees from a predetermined position, the crosstalk level of the 500Hz component (crosstalk of the right channel signal to the left ear) is: The crosstalk level is -34.7dB when turning to the right and -∞dB when turning to the left, and the crosstalk level when turning to the right is -17.8dB and the crosstalk level when turning to the left in the case of the two-speaker system shown in Figure 5. As is clear when compared with -22dB, in the case of this example, the amount of crosstalk based on the rotational displacement of the listener's head can be suppressed to a small level, and the sound image when the listener turns his/her head can be reduced. This has the effect of providing a stable sense of forward localization with little deterioration.

Note that the head diffraction transfer function at the ear near or far from the sound source is exp(−jω・r・
sinθ/C _O ), exp (-jω・r・θ/C _O ), where r is the head radius and C _O is the speed of sound (=340
m/s), and θ is the angle of incidence of sound with respect to the median plane. The crosstalk level above is based on the head radius
Calculations were made assuming a 10cm sphere, treating sound waves as plane waves, and ignoring attenuation due to diffraction.

As explained above, according to the present invention, when a binaural signal is input, the binaural signal is branched into three outputs, and each output is played back by a speaker, the binaural signal is transmitted independently to the listener's left and right ears. Equipped with an equalizer circuit composed of a filter circuit and a delay circuit that simulate the head diffraction transfer function, and by reproducing the outputs of each of the equalizer circuits through three separate speakers, reproduction through the speakers is equivalent to head-on reproduction. The sound image is stable in both the front and rear directions, and if balanced, 360-degree localization can be achieved.

Further, the deterioration of the sound image caused by the rotation of the listener's head is small, and the effective listening position is widened.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of one embodiment of the present invention. Second
This figure is a block diagram of the equalizer circuit of the embodiment shown in FIG. 1. FIGS. 3a and 3b are diagrams showing examples of real-side values of transfer functions. FIG. 4 is a block diagram of a specific example of the equalizer circuit of FIG. 2. FIG. 5 is a block diagram of a binaural signal reproduction device using two speakers. FIG. 6 is a block diagram of another embodiment of the invention. 1...Dummy head microphone, 2 and 2
8... Equalizer circuit, 3, 4, 5, 26, 27
and 29...speaker, 6...listener, 7,
8, 10, 11, 20, 21, 22, 23...Circuit network, 9 and 15...Adder, 12 and 16
〓〓〓〓〓
...Phase inverter, 13 and 14...Peak dip filter, 17...Dip filter, 1
8...Low pass filter, 19...Delay circuit. 〓〓〓〓〓

Claims (1)

[Claims] 1. An equalizer that simulates the head diffraction transfer function so that when a binaural signal is input, the binaural signal is split into three outputs, and each output is played back by a speaker, the binaural signal is independently transmitted to the listener's left and right ears. 1. A three-dimensional sound reproducing device comprising a circuit, and reproducing respective outputs of the equalizer circuit through three speakers.