JPS583640B2 - Sound quality correction circuit - Google Patents

Description

[Detailed Description of the Invention] A pinaural source picked up by a dummy head with individual microphones in both ears can be heard by a listener using headphones with appropriate frequency characteristics. It is possible to hear a pinaural sound that has the three-dimensional sound and sound quality as intended by the producer, but if the listener listens to the above-mentioned pinaural source in a sound field played by two speakers, It is well known that due to the presence of crosstalk between two speakers and the listener, the way the sound image is localized and the sound quality are perceived as being completely different from what the source producer intended. When a pinaural source is played back by two speakers, the pinaural source is A signal conversion circuit that transforms the source is used, and the signal transformed by the signal conversion circuit is supplied to two speakers.

By the way, attempts have been made to broadcast binaural sources as a form of stereophonic broadcasting, but the problem with this is that not all broadcast listeners have headphones. Even if binaural broadcasting is being performed, many listeners will listen to the binaural source within the sound field produced by two speakers.
A listener who has to listen to a pinaural source within a sound field played by a single speaker will only receive completely unsatisfactory stereophonic sound reproduction results in terms of sound image localization and sound quality when listening to a pinaural source. be.

Of course, the above problems could be solved if all listeners of pinaural broadcasts were equipped with headphones or a signal converter like the one mentioned above, but it is difficult to solve these problems. This method would place a huge financial burden on the listeners, so it seems impossible to realize it.

In considering solutions to the above-mentioned problems that are expected to occur when pinaural broadcasting is implemented, the inventor of the present invention discovered that a listener of pinaural broadcasting should listen to a pinaural source played back by two speakers. First of all, we should accept that the way the sound image is localized is unsatisfactory when listening in a sound field, and at least we should be able to improve the quality of the reproduced sound. As a result of various studies focusing on the above, the sound quality correction circuit of the present invention was completed.

That is, the present invention provides a first circuit configured to have a characteristic opposite to the front characteristic of the dummy head used for sound collection, and a crosstalk characteristic that is approximately the same as that which should occur in the reproduction sound field between the two speakers. A sound quality correction circuit in which both channels of a pinaural source are each provided with a circuit connected in cascade with a second circuit configured such that the output signal is fed back in reverse phase through a feedback path having equal characteristics. The sound quality correction circuit according to the present invention is intended to solve the above-mentioned problems in a good manner by providing the above-mentioned sound quality correction circuit.The specific contents of the sound quality correction circuit of the present invention will be explained in detail below with reference to the accompanying drawings.

Figure 1 shows an example of the frequency response characteristic curve of the sound pressure generated in front of the listener's eardrum when a sound source S placed in front of the listener radiates a sound wave with a constant acoustic power over the entire audible frequency band. In the figure, this characteristic curve is, for example, located at a certain distance from a sound source S emitting sound waves with a constant acoustic power over the entire audible frequency band, and directly facing the sound source S. The sound pressure in front of the eardrum of the human ear can be measured using probe-type microphones inserted into both ears of a human being, or the shape and dimensions of the eardrum can be simulated while taking into consideration the acoustic impedance of the eardrum. This characteristic curve is obtained by measuring the sound pressure in front of the eardrum in a dummy head, and this characteristic curve shows the transmission characteristic between the sound source S and the eardrum.

FIG. 2 is a diagram illustrating and explaining how to measure the sound pressure when obtaining the frequency characteristic curve of sound pressure at the front surface of the eardrum as shown in FIG. Or a dummy head, where ec is the auricle, e1 is the auditory canal, ed is the eardrum, P is the probe microphone glove,
Mp is the main body (acoustic-electric converter part) of the probe microphone, and in this Figure 2, the tip of the glove P of the probe microphone Mp is placed just before the eardrum ED placed on the human head or dummy head H. By positioning
It is illustrated and explained that the sound pressure in front of the eardrum can be measured.

Further, in FIG. 3, the main body of the microphone M is embedded in the auditory canal so that the tip of the microphone M is located at the entrance of the auditory canal of the dummy head H, as shown in FIG. 4. A characteristic curve diagram showing an example of the transfer characteristics from the sound source to the microphone obtained when a sound source S placed in front of the dummy head H radiates a sound wave with a constant acoustic power over the entire audible frequency band. Furthermore, FIG. 5 shows a microphone M installed in the dummy head H such that the tip of the microphone M is located at the eardrum position in the dummy head H equipped with the auditory canal EL as shown in FIG. 6. A sound source S placed in front of the dummy head H is used.
, which shows an example of a characteristic curve diagram of the transfer characteristic from the sound source to the microphone obtained when a sound source with constant acoustic power is radiated over the entire audible frequency band.

In addition to the above, for binaural sound collection, a microphone is placed near the entrance of the auditory canal in the actual human head to collect sound without blocking the auditory canal, or a microphone is placed as a dummy head. Various forms of binaural sound collection have been attempted, such as using various simple formats to perform binaural sound collection, but in all of these cases, it is difficult to connect the sound source S to the microphone. The transfer characteristics up to are shown in Figures 1, 3, and
A characteristic curve similar to that shown in FIG. 5 is obtained.

Therefore, in the following explanation, the pinaural source that is collected in the state shown in Figure 2 will be considered as the reference for the pinaural source, and the listener M will listen to the human voice shown by H in Figure 2. Same frontal transfer characteristics as the head or dummy head (
A person is said to have a frontal transmission characteristic as shown in the first diagram (in the following description, "transmission" may be omitted and simply expressed as a frontal characteristic). , Furthermore, the human head or dummy head H in FIG.
The sound quality perceived by the listener M when the listener M's head is positioned is considered to be the standard sound quality.

Now, listener M listens to the two speakers SP as shown in the seventh diagram.
It is assumed that each channel signal of the pinaural north is listened to in a three-dimensional sound reproduction sound field using speakers SPr and SPl, and the sound is collected by the above-mentioned speakers SPr and SPl in the manner shown in the second diagram. In this case, the listener M can only hear sounds with a sound quality different from the above-mentioned standard sound quality, and this point will be explained next.

In FIG. 7, S is a sound source, H is a dummy head, and the sound source S and dummy head H are in a state where the sound source S is in front of the dummy head H, as in the case shown in the second diagram described above. It is said that

Each channel signal of the pinaural source picked up by the dummy head H is given to input terminals 1 and 2 of the playback device, and after being amplified individually by amplifiers 3 and 4, it is sent to the right speaker SPr and the left speaker SPl. , speaker signal SPr
s, SPls.

By the way, in a three-dimensional sound reproduction sound field using two speakers SPl and SPr, the listener M listens to two speakers as shown in the seventh figure.
The transmission characteristics of the reproduced sound field when the two speakers SPl and SPr are equidistant from each other and facing toward the center of the two speakers SPr are the transmission characteristics from the right speaker SPr to the front of the eardrum of the right ear. A, left speaker SP
Transfer characteristics from l to the front of the eardrum of the left ear are A. Transfer characteristics from the right speaker SPr to the front of the eardrum of the left ear are B1 Transfer characteristics from the left speaker SPl to the front of the eardrum of the right ear are B. The front of the eardrum of the right ear. The sound pressure is Er, and the sound pressure in front of the eardrum of the left ear is E.
l, right speaker signal SPrs, left speaker power signal SP
Assuming ls, it is expressed by the following equation (1).

In the above equation (1), left and right speaker signals SP
ls and SPrs are signals caused by changes in the sound pressure in front of the eardrum in the dummy head H. Now, as shown in FIGS. 2 and 7, the dummy head is affected by the sound from the sound source S placed in front of the dummy head H. Let C be the signal due to the change in sound pressure generated on the front surface of the eardrums of the left and right ears of H, then the sound pressure El generated on the front surface of the eardrums of the left and right ears of listener M who is in the reproduction sound field shown in Figure 7. , Er (which may also be signals El and Er due to sound pressure changes) are expressed as in the following equation (2) where both speaker signals SPls and SPrs in equation (1) above are set as C.

That is, if listener M places his head at the position of dummy head H in FIGS. 2 and 7 and listens directly to the sound from sound source S, listener M will hear Er=C, El=C. Even though the listener M can hear the sound (standard quality sound), the listener M in the reproduction sound field using two speakers hears the sound quality (A + B) C, which is different from the original sound quality C. You will be listening to it.

Transfer characteristic A between two speakers and both ears of listener M,
B has a time difference from each other as is clear from FIG. 7, and the characteristic (A+B) shown as the sum of the two transfer characteristics A and B described above exhibits characteristics like a comb-shaped filter. Therefore, in the sound field reproduced by the two speakers as shown in Figure 7, the listener M is listening to the sound C through the comb-shaped filter having the characteristic (A+B) described above. Therefore, the listener M will listen to sound with poor quality compared to the standard sound quality.

The transfer characteristics A and B are shown in FIG. 8, and (A+B) is shown in FIG.

Based on the above, the inventor of the present invention believes that in order to enable listener M, who listens in the sound field produced by two speakers, to hear the reproduced sound of good quality, it is necessary to It is only necessary to make it possible to apply the signal to the peaker.If it were constructed directly, it would become an extremely complicated circuit, and its realization would be extremely difficult.

The present inventor has completed the present invention with a relatively simple circuit configuration without prior various researches, and FIG. 10 shows a block diagram of one embodiment of the sound quality correction circuit EQ of the present invention.

In Fig. 10, 5 and 6 are input terminals for each channel signal Ir and Il of the binaural source, EQ is a sound quality correction circuit consisting of a cascade connection of a first circuit EA and a second circuit EB, and 7 and 8 are outputs. terminals, input terminal 5 and output terminal 7
The circuit provided between the input terminal 6 and the output terminal 8 and the circuit provided between the input terminal 6 and the output terminal 8 are configured to have the same characteristics.

The first circuit EA includes circuits 9 and 1 that transmit signals between the speakers and both ears of the listener in the sound field reproduced by the two speakers.
0, one channel signal of the binaural source is applied to the second circuit EB through the circuit 9, and an Ikegata channel signal of the binaural source is applied to the second circuit EB through the circuit 10.

In the second circuit EB, a sign adder 13 and a cross path to be generated in the reproduction sound field consisting of two speakers are configured between the circuit 9 in the first circuit EA and the output terminal 7, Further, a circuit to be generated in the reproduction sound field consisting of the sign addition adder 14 and two speakers is connected to the circuit 1 in the first circuit EA.
0 and the output terminal 8.

The feedback path 11 described above is used as a feedback path for feeding back the output signal Or appearing at the output terminal 7 to the input side of the sign addition adder 130 in reverse phase. is also given a signal from circuit 9.

Further, the feedback path 12 described above is used as a feedback path for returning the output signal C1 appearing at the output terminal 8 to the input side of the sign addition adder 14 in reverse phase, and A signal from the circuit 10 is also provided to the circuit 14.

The characteristics between the input terminal and the output terminal of the correction circuit shown in FIG.

By transforming equation (3), equation (5) is obtained from equation (4) as shown above, but when the input signals Ir and Il in equation (5) are placed in front of dummy head H, When the sound emitted from the sound source S is collected, that is, when Ir=Il=C, the above (5)
The equation is as shown in equation (5a) below.

Therefore, the binaural source that collects the sound from the sound source S placed in front of the dummy head H is passed through the sound quality correction circuit EQ shown in FIG.
SPl, that is, when the listener M who is in the sound field reproduced by the two speakers as shown in Figure 11 listens, the dummy head H used by the listener M to pick up the sound of the binaural source. This means that the user can hear the same sound quality as when listening to the sound from the sound source S while placing the head in the position.

This point can be explained using a mathematical formula as follows.

In the case shown in Figure 11, the speaker signals SPrs and SPls given to the two speakers SPr and SPl are the output signals Or and Ol of the sound quality correction circuit EQ, respectively, and SPrs=Or and SPls=Ol, so By substituting Or and Ol for SPrs and SPls in the above-mentioned equation (1), the following equation (6) is obtained.

In addition, in the case shown in Figure 11, since the output signals Or and Ol of the sound quality correction circuit EQ are respectively shown by equations (5a), equation (5a) is substituted for Or and Ol in equation (6). As a result, the following equation (7) is obtained.

Therefore, when the sound quality correction circuit EQ is used as shown in FIG. It becomes equal to the signal C of the sound pressure change in front of the eardrum.

In this way, if the binaural source is transformed through the sound quality correction circuit EQ configured as shown in Figure 10 and then broadcast, the sound from the sound source S in front of the dummy head H can be collected by the dummy head H. In other words, if the signals of both channels of the binaural source are in phase and at the same level, the listener will be able to pick up the sound even if the received binaural broadcast is played back by a normal 2-channel 3D sound reproduction device. It is possible to listen to the same sound quality as when listening to the sound from the sound source S at the position of the dummy head H used in the above.

The binaural source to be supplied to the sound quality correction circuit EQ shown in FIG. , a listener M listening in a sound field reproduced by two speakers can hear sound with no problem in terms of sound quality when the sound is reproduced from the front of the listener.

Return circuit 1 to the above-mentioned sound quality correction circuit EQ shown in Figure 10
1, 12, etc. are used, but these circuits can be easily constructed by combining a filter with required passband characteristics and a delay circuit.

By the way, as mentioned above, the circuits described above in FIG. 10 have the transmission characteristics A,
However, as can be seen immediately by comparing curve A in FIG. 8 with the characteristic curve shown in FIG. 1, the transfer characteristic A shown by curve A in FIG. Since the transfer characteristic A in , and the front characteristic in the dummy head H used for sound collection as shown in the second figure are approximately equal, the sound quality correction circuit EQ
The characteristics of the circuits 9 and 10 constituting the first circuit EA may be the opposite of the front characteristic of the transfer characteristic in the reproduced sound field.

In addition, as mentioned above, the front characteristics of the dummy head shown in FIG. Transmission characteristics from the sound source to both ears or the microphone when the sound source is in front of the head, and from the sound source located around 30 degrees in front of the left or 30 degrees in front of the right of a human or dummy head to the ear or microphone on the side closer to the sound source. Since the transmission characteristics are very similar to that of the dummy head, the circuits 9 and 10 in the first circuit EA of the sound quality correction circuit EQ shown in FIG. Alternatively, it may be the opposite of the characteristic around 30 degrees left and right in front of the dummy head.Furthermore, the front characteristic of the dummy head (the range from one side to the front side to the other side is defined as the front side) A circuit with opposite characteristics is used as a sound quality correction circuit EQ.
It may be used as circuits 9 and 10 in the first circuit EA.

It should be noted that the second circuit EB of the sound quality correction circuit EQ. The feedback circuits 11 and 12 are configured to cross the reproduced sound field.

In the explanation so far, the sound quality correction circuit EQ shown in FIG.
was supposed to be configured to perform sound quality correction for a pinaural source picked up by a sound pickup device configured as shown in the second figure and having the frontal characteristics as shown in the first figure. For example, if the pinaural source given to the circuit EQ is generated by a sound pickup device configured as shown in FIGS. 4 and 6, It is necessary to further correct the difference between the front characteristics shown in FIG. 1 and the front characteristics of the sound pickup device configured as shown in FIG. In this case, a troublesome configuration such as cascading a correction circuit for correcting the difference in front characteristics between the sound pickup devices described above to the sound quality correction circuit EQ shown in FIG. 10 is not required. Instead, as circuits 9 and 10 in the first circuit EA of the sound quality correction circuit EQ shown in FIG. It is sufficient to use a circuit with characteristics opposite to the front characteristics of the device, and even in this case, the sound quality correction circuit EQ
A circuit 9,1 in which a circuit having a characteristic opposite to the forward characteristic in the sound pickup device used to pick up the sound of the pinaural source supplied to the sound quality correction circuit EQ is provided as the first circuit EA of the sound quality correction circuit EQ.
It may be used as 0.

So far, we have focused on the sound quality of the sound image in front of the listener, but in actual program sources, the sound image does not exist only in front of the listener.

However, when the sound source is located in a position other than the front of the dummy head, analysis using mathematical formulas is extremely complicated and difficult to perform. We decided to conduct experiments to determine the configuration of a sound quality correction circuit that would function effectively.

As a result of various experiments, the feedback path 1 in the second circuit EB of the sound quality adjustment circuit EQ shown in FIG.
It has been found that by appropriately changing the magnitude of the signals fed back through the listeners 1 and 12, relatively good sound quality can be achieved even for sound images other than those directly in front of the listener M.

FIG. 12 shows a sound quality correction circuit EQ having a configuration in which variable attenuators 15 and 16 are inserted in the transmission path of the feedback signal in the second circuit EB of the sound quality correction circuit EQ shown in FIG.
FIG. 12 is a block diagram of one embodiment of FIG. 12, in which the second circuit is designated as EBα.

It should be noted that the first
The circuit FA is the sound quality correction circuit EQ shown in FIG.
It has the same configuration as the first circuit in .

According to the results of experiments, in the sound quality correction circuit EQ shown in FIG. It has been found that the sound quality of the sound image can also be made relatively good.

The curve EBα in FIG. 13 represents the second circuit E in the sound quality correction circuit EQ having the configuration shown in FIG.
This is an example of the frequency response characteristic curve of Bα, and the first
The curve EB in FIG. 3 is an example of the frequency response characteristic curve of the second circuit EB in the sound quality correction circuit EQ having the configuration shown in FIG. The frequency response characteristic of the second circuit EBα in the illustrated sound quality correction circuit EQ is such that the feedback amount of the feedback signal is several dB higher than that of the circuit arrangement shown in FIG.
Correspondingly, the sound quality correction circuit EQ having a configuration as shown in FIG. 10 indicated by a curve EB
The frequency response characteristic of the second circuit EB is increased by several dB compared to the low frequency part.

Incidentally, the binaural source whose sound quality has been corrected by the sound quality correction circuit EQ configured according to the present invention can be listened to with headphones without any loss of binaural effect and with good sound quality.

This means that the sound quality correction circuit of the present invention does not include any component that changes the localization of the sound image;
and that ordinary headphones have characteristics such that when a signal for stereophonic sound reproduction is supplied from two speakers, good sound quality is given to the listener;
This can be easily understood from the following.

As is clear from the detailed explanation above, if the sound quality correction circuit of the present invention is used, when broadcasting or recording binaural sources and supplying them to listeners,
Even listeners who do not own headphones and who play binaural sound using a three-dimensional sound reproduction device using two speakers can hear good quality sound, and the present invention allows the listeners to hear good quality sound. The problem is successfully resolved.

[Brief explanation of the drawing]

1, 3 and 5 are examples of transfer characteristic curves between a sound source and a microphone placed in front of a dummy head with different configurations, respectively; FIGS. The figures are an explanatory diagram of a sound collection device using dummy heads with different configurations, respectively. 8, 9 and 13 are explanatory characteristic curve examples, FIGS. 10 and 12 are block diagrams of different embodiments of the sound quality correction circuit of the present invention, and FIG. 11 is a diagram of the characteristic curves of the present invention. FIG. 2 is a block diagram showing an example of application of a sound quality correction circuit. EQ...Sound quality correction circuit, EA...First
circuit, EB, EBα... second circuit, Mp
, M...Microphone, H...Dummy head, M...Listener, SPr, SPl...
...Speaker, P...Probe, 9,10...
... Circuit with characteristic 1/A, 11, 12 ... Feedback path, 13, 14 ... Sign adder, 15, 16
...Variable attenuator.

Claims (1)

[Claims] 1. A first circuit configured to have characteristics opposite to the front characteristics of the dummy head used for sound collection, and crosstalk characteristics that should occur in the reproduction sound field between the two speakers. Both channels of the pinaural source are independently provided with circuits connected in cascade with a second circuit configured such that the output signal is fed back in opposite phase through feedback paths having substantially equal characteristics. Sound quality correction circuit. 2. The sound quality correction circuit according to claim 1, wherein the entire output signal is fed back through the feedback path in the second circuit. 3. The sound quality correction circuit according to claim 1, wherein the magnitude of the signal fed back via the feedback path in the second circuit can be variably adjusted.