JPH06303691A - Stereo phonic microphone - Google Patents

Abstract

PURPOSE:To collect remote sound stereophonically with high S/N. CONSTITUTION:The microphone is provided with a first and second ultra directivity microphones 1, 2 whose major axis is directed in front and installed at a position at an equal distance from a front side sound source in parallel, a 1st signal subtraction means 31 taking a difference between a signal from the 1st ultra directivity microphone 1 and a signal obtained by delaying a signal from the 2nd ultra directivity microphone 2 by a time tau1, a 2nd signal subtraction means 32 taking a difference between a signal from the 2nd ultra directivity microphone 2 and a signal obtained by delaying a signal from the 1st ultra directivity microphone 1 by a time tau2. Thus, an output from the 1st signal subtraction means 31 has a directivity collecting sound in a range at a narrow angle toward right side from the front and an output from the 2nd signal subtraction means 32 has a directivity collecting sound in a range at a narrow angle toward left side from the front and then the stereophonic microphone having an ultra directivity is formed.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention is designed to produce high S / N for distant sounds.
It relates to a stereo microphone for collecting stereo sound.

[0002]

2. Description of the Related Art A conventional stereo microphone will be described below.

FIG. 7 shows the structure of a conventional stereo microphone. Reference numeral 60 is a first directional microphone unit, 61 is a second directional microphone unit, and the first directional microphone unit has a main axis oriented in a direction of an angle θ from the front direction to the left, and a second directional microphone unit. The main axis of the sex microphone unit is oriented in the direction of an angle θ from the front direction to the right. A first output terminal 62 receives the output signal from the first directional microphone unit. 63 is a second output terminal, which receives the output signal from the second directional microphone unit.

The operation of the conventional stereo microphone configured as described above will be described below.

The first directional microphone unit 60 in which the directional main axis is oriented in the direction of the angle θ from the front direction to the left side emphasizes and collects the sound on the left side, as shown in FIG. 7B. Get directional characteristics. Further, the second directional microphone unit 61 in which the directional main axis is directed from the front direction to the right in the direction of the angle θ collects the sound on the right by emphasizing the sound.
A directional characteristic as shown in (c) is obtained. As a result, stereo sound collection is realized. At this time, the range of θ is usually 45.
The angle is set to ≦ θ ≦ 90 °.

[0006]

However, in the above-mentioned configuration, since the sound pickup angle is wide and a wide range of sound is picked up for the purpose of obtaining the stereo feeling of the nearby sound field, the superdirective microphone is used. There was a problem that it was not possible to pick up a distant sound that could be picked up, and even if the surroundings were quiet and could pick up a distant sound, it would not be possible to obtain a stereo sense of the distant sound field. .

The present invention solves the above-mentioned problems and provides a stereo microphone that distributes superdirective directional characteristics to the left and right and collects distant sounds in stereo.

[0008]

To achieve the above object, in a stereo microphone of the present invention, a first super-directional microphone and a second super-directional microphone are provided in parallel with each other with their main axes facing forward. A first signal delay means provided after the first super-directional microphone, a second signal delay means provided after the second super-directional microphone, and a first super-directional microphone Output signal from the second signal delay means and first signal subtraction means receiving the output signal from the second signal delay means, an output signal from the second superdirective microphone and an output signal from the first signal delay means. A microphone having superdirectivity and capable of collecting stereo sound can be obtained by providing a second signal subtracting means for inputting, and obtaining output signals from each of the first signal subtracting means and the second signal subtracting means.

Furthermore, by varying the signal delay amounts of the first signal delay means and the second signal delay means, it is possible to adjust the stereo feeling according to the sound field. The range of the signal delay amount τ 1 at this time is the range of the first super-directional microphone and the second
The distance between the sound holes of the super directional microphone of
Then, 0 <τ1 ≦ d / c.

Furthermore, the first super-directional microphone and the second super-directional microphone are waveform type super-directional microphones having sound holes on the side surface of the acoustic tube, and By providing the sound holes and the sound holes of the second superdirective microphone in opposite directions so as to maximize the distance, the distance between the sound holes can be increased in a limited size, It is possible to improve the characteristics in the low frequency range.

Further, the first super-directional microphone and the second super-directional microphone are provided in parallel with each other with their main axes facing the front direction, and the first super-directional microphone receives the output signal from the first super-directional microphone as an input. Signal delaying means, second signal delaying means for inputting the output signal from the second super-directional microphone, and third signal delaying means for inputting the output signal from the first super-directional microphone. A fourth signal delay means for receiving an output signal from the second superdirective microphone, a first amplitude / phase correcting means for receiving an output signal from the first signal delay means, and a second Second amplitude / phase correction means that receives the output signal from the signal delay means, and the first amplitude-phase correction means that receives the output signal from the third signal delay means and the output signal from the second amplitude / phase correction means. A signal subtracting means and a fourth Compensating for variations in the characteristics of the first and second superdirective microphones, the second signal subtracting means having the output signal from the signal delay means and the output signal from the first amplitude / phase correcting means as inputs. Thus, the signals from the specific direction are surely subtracted in the first and second signal subtraction means, and the directivity can be improved.

Further, in addition to the above configuration, a cross-correlation calculating means for inputting an output signal from the first super-directional microphone and an output signal from the second super-directional microphone is provided, and the first amplitude phase The correcting means is replaced with the first adaptive amplitude / phase correcting means, and the amplitude / phase characteristic thereof is updated based on the output signal from the second signal subtracting means and the output signal from the cross-correlation calculating means to obtain the second amplitude. The phase correcting means is replaced with the first adaptive amplitude / phase correcting means, and the amplitude / phase characteristic is updated based on the output signal from the first signal subtracting means and the output signal from the cross-correlation calculating means. As a result, the first adaptive amplitude phase correcting means and the second adaptive amplitude phase correcting means can realize appropriate characteristics by self-learning.

[0013]

The stereo microphone configured as described above is the sound picked up by the first super-directional microphone installed on the right side and the second super-directional microphone installed on the left side with respect to the front direction. The signal is calculated and separated into left and right signals. It is assumed that the characteristics of the first and second superdirective microphones are substantially equal. In the first signal subtraction means, the output signal from the first super-directional microphone and the second signal
The output signal from the super-directional microphone of is delayed by time τ1, and the result of the subtraction is a signal in the direction of angle θ = sin ^-1 (τ1 · c / d) from the front direction to the left. Is canceled (where d is the distance between the sound holes of the first and second superdirective microphones, and c is the speed of sound). Therefore, the output from the first signal subtraction means becomes a directivity characteristic in which a blind spot is formed in the direction of the angle θ to the left of the directivity characteristic of the first superdirective microphone, and the superdirective sound taking the sound from the front to the right direction. It becomes a characteristic.

Similarly, in the second signal subtraction means, the second signal subtraction means
Of the output signal from the super-directional microphone of and the output signal from the first super-directional microphone delayed by time τ 1 are performed, and as a result, the output of the second signal subtraction means is This is a directional characteristic in which a blind spot is formed in the direction of the angle θ to the right of the directional characteristic of the super directional microphone, that is, a super directional characteristic in which sound is emitted from the front to the left. At this time, the value of τ1 is 0 <θ ≦ 90
As 0 °, 0 <τ1 ≦ d / c. The angles θ and τ 1 from the front in the blind spot direction are expressed by (Equation 1).

[0015]

[Equation 1]

Further, by making τ1 variable, it becomes possible to change the angle in the blind spot direction, so that the directivity can be set according to the sound field and the condition of the sound skipping in the front direction can be adjusted. Become.

Further, the third and fourth signal delay means and the first
And the second amplitude / phase correcting means are provided in the above-mentioned configuration,
When the transfer characteristic of the amplitude / phase correcting means is H1 (ω) and the transfer characteristic of the second amplitude / phase correcting means is H2 (ω), the respective characteristics are determined by (Equation 2) and (Equation 3).

[0018]

[Equation 2]

[0019]

[Equation 3]

However, M1 _R (ω) and M2 _R (ω) are frequency characteristics of the first and second superdirective microphones in the right θ ° direction, M
1 _L (ω) and M2 _L (ω) are frequency characteristics of the first and second superdirective microphones in the left θ ° direction. If the characteristics of the first and second super-directional microphones are exactly the same, M1 _R (ω) = M
2 _R (ω), M1 _L (ω) = M2 _L (ω), but in reality there are variations in the characteristics of individual superdirective microphones. Therefore, first and second amplitude / phase correction means are provided.

Further, the third and fourth signal delay means are the first
And is provided for the purpose of compensating for the signal delay in the second amplitude / phase correction means. In this way, the directivity is improved by more accurately canceling the signal in the blind spot direction.

Further, the cross-correlation calculating means is provided as in the above-mentioned structure, and the transfer characteristic H1 (ω) of the first adaptive amplitude / phase correcting means is set to the second value only when the correlation of the sound wave from the right θ ° is high.
The signal is updated using the output signal from the second signal subtracting means so that the output signal from the signal subtracting means of (1) becomes small. Similarly, the transfer characteristic H2 (ω) of the second adaptive amplitude / phase correction means is set to the output signal from the first signal subtraction means only when the correlation of the sound wave from the left θ ° is high in the calculation result of the cross-correlation calculation means. The value is updated using the output signal from the first signal subtraction unit so as to be smaller. With such an operation, the first and second super-directional microphones have the first characteristic in the direction of the directional blind spot (the direction of θ ° in each of the left and right directions).
And the second adaptive amplitude / phase correcting means self-corrects, so M1 _R (ω), M2 _R (ω), M
It is no longer necessary to obtain 1 _L (ω) and M2 _L (ω), and even if τ1 is changed and the blind spot direction is changed, the appropriate H1
(ω) and H2 (ω) are obtained.

As described above, the output of the first signal subtracting means has the super-directional characteristic of collecting the right side from the front, and the output of the second signal subtracting means is the super-directive characteristic of collecting the left side from the front. It becomes a characteristic.

[0024]

DESCRIPTION OF THE PREFERRED EMBODIMENTS A stereo microphone according to a first embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 shows the configuration of a stereo microphone according to the first embodiment of the present invention. In FIG. 1, 1 is a first super-directional microphone, 2 is a second super-directional microphone, and a first super-directional microphone is provided on the left side of the first super-directional microphone with respect to the front. Install in parallel and equidistant from the front sound source. 1
Reference numeral 1 is a first signal delay means, which receives an output signal from the first superdirective microphone 1. Reference numeral 12 denotes a second signal delay means, which is the second superdirective microphone 2
The output signal from is used as the input. Reference numeral 31 is a first signal subtracting means, which receives the output signal from the first superdirective microphone 1 and the output signal from the second signal delay means 12. A second signal subtraction means 32 receives the output signal from the second superdirective microphone 2 and the output signal from the first signal delay means 11. Reference numeral 51 denotes a first output terminal, which is provided at a stage subsequent to the first signal subtracting means 31. Reference numeral 52 denotes a second output terminal, which is provided at a stage subsequent to the second signal subtracting means 32.

The operation of the stereo microphone configured as described above will be described below with reference to FIGS. 1, 2 and 3.

In FIG. 1, the directional characteristics of the first super directional microphone 1 and the second super directional microphone 2 are shown in FIG.
It is assumed that the characteristics are almost equal as shown in (a). The first signal subtracting means 31 subtracts a signal obtained by delaying the output signal of the first super-directional microphone 1 and the output signal of the second super-directional microphone 2 by time τ1. As a result, the directivity characteristic of the output signal from the first signal subtracting means 31 is shown in FIG.
The dead angle is formed in the left θ direction of the super directional characteristic of. The angle θ depends on the first super-directional microphone 1 and the second super-directional microphone 1.
Letting d be the distance between the sound holes and the super-directional microphone 2 of c, and c be the speed of sound, they are expressed as in equation (4).

[0028]

[Equation 4]

Further, when the relationship between θ, d, τ 1 and c is shown in FIG. 2, the sound wave arriving from the direction from the front to the left at the angle θ ° first reaches the second super-directional microphone 2 and the time
First super-directional microphone 1 with a delay of d · cos (θ) / c
Reach Therefore, the output signal from the second super-directional microphone 2 is output by the second signal delay means 12 at time τ1 =
By delaying d · cos (θ) / c and subtracting from the output signal of the first superdirective microphone 1, the sensitivity in the θ ° direction from the front to the left becomes zero. This is a so-called sound pressure gradient microphone directivity forming method, and the directivity added by this calculation is as shown in FIG. Therefore, the directivity characteristic of the output signal from the first signal calculation means 31 is as shown in FIG. 3C, which is obtained by multiplying the superdirectivity of FIG. 3A by the directivity of FIG. 3B. Similarly, the second
The output signal from the signal subtraction means 32 has a superdirective characteristic (FIG. 3 (e)) having a blind spot in the right side θ ° direction. Figure 3
It is possible to stereophonically collect a distant sound by directivity as shown in (c) and (e).

Further, in order to realize the miniaturization of the entire apparatus by arranging the first and second super-directional microphones 1 and 2 as close as possible, the first super-directional microphone as shown in FIG. Assuming that the directional microphone 1 and the second super-directional microphone 2 are line microphones (waveform microphones) provided with sound holes on the side surface of the acoustic tube piece, the sounds of the first and second super-directional microphones 1 and 2 are Rather than arranging the holes so as to face each other (FIG. 4 (b)), the sound holes are arranged on the side surface opposite to the other super-directional microphone (FIG. 4).
By (a)), the distance d is increased and the characteristic in the low frequency range is improved (equal to the relationship between the distance between the sound holes (or between the microphone elements) of the sound pressure gradient type microphone and the low frequency range characteristic).

Further, by varying the signal delay amount τ1 of the first and second signal delaying means, the angle θ in the dead angle direction can be varied according to the above-mentioned (Equation 5). Also,
At this time, the range of τ1 is 0 <τ1 ≦ d / c, considering the range of θ as an appropriate value of 0 ° <θ ≦ 90 °.

A stereo microphone according to a second embodiment of the present invention will be described below with reference to the drawings.

FIG. 5 shows the configuration of a stereo microphone according to the second embodiment of the present invention. In FIG. 5, 1 is a first super-directional microphone. 2 is second
Is a super directional microphone. Reference numeral 11 is a first signal delay means, which receives an output signal from the first superdirective microphone 1. Reference numeral 12 is a second signal delay means, which receives the output signal from the second superdirective microphone 2. Reference numeral 13 is a third signal delay means, which receives an output signal from the first superdirective microphone 1.
Reference numeral 14 is a fourth signal delay means, which receives an output signal from the second superdirective microphone 2. 21 is the first
The amplitude and phase correction means of the first signal delay means 11
The output signal from is used as the input. Reference numeral 22 is a second amplitude / phase correction means, which receives the output signal from the second signal delay means 12. 31 is a first signal subtraction means,
The output signal from the third signal delay unit 13 and the output signal from the second amplitude / phase correction unit 22 are input. A second signal subtracting means 32 receives the output signal from the fourth signal delaying means 14 and the output signal from the first amplitude / phase correcting means 21. Reference numeral 51 denotes a first output terminal, which is provided at a stage subsequent to the first signal subtracting means. Reference numeral 52 denotes a second output terminal, which is provided at a stage subsequent to the second signal subtracting means.

The operation of the stereo microphone configured as described above will be described below with reference to FIG.

In FIG. 5, the difference from the first embodiment is that the third signal delay means 13 is provided between the first superdirective microphone 1 and the first signal subtraction means 31. The fourth signal delay means 14 is provided between the second super-directional microphone 2 and the second signal subtraction means 32, and the first amplitude / phase correction means 21 is provided as the first signal delay means 11. Provided between the second signal subtraction means 32 and
The second amplitude / phase correction means 22 is replaced with the second signal delay means 12
And the first signal subtraction means 31. The purpose is to match the amplitude and phase characteristics of the first and second superdirective microphones 1 and 2 with respect to the left and right angle θ directions, and There is enough cancellation. First and second
Regarding the determination of the transfer characteristics of the amplitude / phase correcting means 21 and 22, the sound pressure frequency characteristics in the right θ ° direction of the first and second super-directional microphones 1 and 2 measured under the same conditions are respectively M
As 1 _R (ω) and M2 _R (ω), the first amplitude / phase correction means 21
The transfer characteristic H1 (ω) of is determined as in (Equation 5).

[0036]

[Equation 5]

The output of the sound wave from the direction of the angle θ ° to the right of the first superdirective microphone 1 is delayed by the time τ 1 by the first signal delay means 11, and then the first amplitude / phase correction means is provided. 21 is multiplied by the characteristic represented by (Equation 5) to equalize the sound pressure frequency characteristic with respect to the sound wave from the right θ ° direction of the second super-directional microphone 2, and the second signal subtracting means 32 The subtraction is performed to cancel the signal of the sound wave in the right θ ° direction. At this time, the fourth signal delay means 14
The first amplitude / phase correction unit 21 is provided for compensation when a signal time delay occurs. Similarly,
The transfer function of the second amplitude / phase correction means 22 is determined as in (Equation 6).

[0038]

[Equation 6]

The output of the sound wave from the direction of the angle θ ° to the left of the second superdirective microphone 2 is delayed by the time τ 1 by the second signal delay means 12, and then the second amplitude / phase correction means is provided. 22 is multiplied by the characteristic represented by the equation 6 to equalize the sound pressure frequency characteristic with respect to the sound wave from the left θ ° direction of the first super-directional microphone 1, and the first signal subtracting means 31 subtracts it. Then, the sound wave signal in the left θ ° direction is canceled. At this time, the third signal delay means 13 is provided in the second amplitude / phase correction means 22 for compensation when a signal time delay occurs.

As described above, in the second embodiment, the first
Even if there is a slight variation in characteristics between the second super-directional microphones 1 and 2, at least the left and right blind spots in the direction of the angle θ ° are sufficiently formed.

A stereo microphone according to a third embodiment of the present invention will be described below with reference to the drawings.

FIG. 6 shows the configuration of a stereo microphone according to the third embodiment of the present invention. In FIG. 6, a first super-directional microphone 1, a second super-directional microphone 2, a first signal delay means 11, a second signal delay means 12, a third signal delay means 13, and a fourth signal. The delay means 14, the first signal subtraction means 31, the second signal subtraction means 32, the first output terminal 51, and the second output terminal 52 have the same configurations as those in the second embodiment.

Differences from the second embodiment will be described below. First, reference numeral 40 denotes a cross-correlation calculating means, which is a first super-directional microphone 1 and a second super-directional microphone 2.
The output signal from is used as the input. Reference numeral 23 denotes a first adaptive amplitude / phase correction means, which is replaced with the amplitude / phase correction means 21 in the second embodiment, receives an output signal from the first signal delay means 11 as an input, and a second signal subtraction means. A signal is output to 32, and the transfer characteristic of itself is adaptively updated by using the output signal from the second signal subtracting means 32 and the output signal from the cross-correlation calculating means 40 as inputs. Reference numeral 24 denotes a second adaptive amplitude / phase correcting means, which is replaced with the amplitude / phase correcting means 22 in the second embodiment, receives the output signal from the second signal delay means 12, and receives the first signal subtracting means. A signal is output to 31 and the output signal from the first signal subtracting means 31 and the output signal from the cross-correlation calculating means 40 are used as inputs to adaptively update its own transfer characteristic.

The operation of the stereo microphone configured as described above will be described below with reference to FIG.

In FIG. 6, the operation different from that of the second embodiment is that the first and second adaptive amplitude / phase correcting means 23 and 24 are the first and the second with respect to the sound wave from the blind angle direction of the right and left θ °. This is a point to adaptively equalize variations in characteristics of superdirective microphones 1 and 2. Here, as a specific example for realizing the first and second adaptive amplitude / phase correcting means 23 and 24, the normalized L
MS algorithm (learning identification method) (for example, S. Heikin, Takebe, Seki: "Introduction to Adaptive Filters", P.119-)
The case where the adaptive equalizer using P.122) is applied is shown.

The impulse response (filter coefficient) giving the transfer characteristic of the first adaptive amplitude / phase correcting means 23 is represented by h _L (n),
The output signal from the first signal delay means 11 is u _L (n),
When the output signal from the signal delay means 14 of the above is d _L (n) and the output signal from the second signal subtraction means 32 is e _L (n), the learning identification method is In the first adaptive amplitude / phase correction means 23, the filter coefficient of (Equation 7) is updated and the second term on the right side of (Equation 8) is calculated. (Equation 8) The minus on the right side corresponds to the second signal subtraction means 32. If d _L (n) and u _L (n) are independent of each other,
Since (Equation 7) cannot converge, it is necessary to update the filter coefficient of (Equation 7) only when the sound wave from the direction to be canceled is strong in order to operate normally as the equalizer. Therefore, the cross-correlation calculation means 40 sends a signal to the first adaptive amplitude / phase correction means 23 to the right indicating whether or not the correlation of the sound waves in the angle θ ° direction is high, and only when the correlation is high, the update of (Equation 7) is performed. I do. The fourth signal delay means 14 is provided to satisfy the causality of each signal with respect to time, and sets the time delay τ 2 corresponding to the time length of the filter impulse response h _L (n). With such a configuration, u _L (n) is d for the sound wave from the direction of the angle θ ° from the front to the right.
_The filter coefficient h _L (n) is updated so that e _L (n) approaches zero in agreement with _L (n), and a clear directional blind spot is formed in the right θ ° direction.

[0047]

[Equation 7]

[0048]

[Equation 8]

The impulse response (filter coefficient) giving the transfer characteristic of the second adaptive amplitude / phase correcting means 24 is expressed by h _R (n),
The output signal from the second signal delay means 12 is u _R (n), the third
When the output signal from the signal delaying means 13 of the above is d _R (n) and the output signal from the first signal subtracting means 31 is e _R (n), the learning identification method is In the second adaptive amplitude / phase correcting means 24, the filter coefficient of (Equation 9) is updated and the second term on the right side of (Equation 10) is calculated.

(Equation 10) The minus on the right side is the first signal reduction.
It corresponds to the calculating means 31. Also, d_R(n) and u_R(n) are
If they are independent, (Equation 9) cannot converge
Therefore, in order to operate normally as an equalizer,
Filter of (Equation 7) only when the sound wave from the desired direction is strong
It is necessary to update the coefficient. Therefore, the cross-correlation calculator
Step 40 indicates whether or not the correlation of the sound waves in the angle θ ° direction to the left is high.
Is sent to the second adaptive amplitude / phase correction means 24,
Only when the correlation is high, the update of (Equation 9) is performed. Third signal
The delay means 13 satisfies the causality regarding the time of each signal.
For filter impulse response h_R(n)
Set the time delay τ2 corresponding to the time length. Structure like this
Depending on the composition, sound waves from the direction of angle θ ° from the front to the left
For u _R(n) is d_Re in agreement with (n)_R(n) approaches zero
Sea urchin filter coefficient h_R(n) is updated, and it is clear in the left θ ° direction
You can make a blind spot with various directivities.

However, h _L (n), h _R (n) is a vector showing a filter coefficient sequence at time n, and u _L (n), u _R (n) is a tap input vector (u _L (n) = (u _L (n), u _L (n-1), u _L (n-2), ...
・)), The number of dimensions of each vector is equal.

[0052]

[Equation 9]

[0053]

[Equation 10]

[0054]

According to the present invention, the first super-directional microphone and the second super-directional microphone are provided parallel to each other with their main axes facing the front direction, and the first super-directional microphone is provided in the latter stage. 1 signal delaying means, and 2nd signal delaying means provided after the 2nd super-directional microphone,
The output signal from the first super-directional microphone and the second signal
First signal subtracting means that receives the output signal from the signal delaying means, and a second signal that receives the output signal from the second superdirective microphone and the output signal from the first signal delaying means By providing subtraction means and obtaining output signals from the first signal subtraction means and the second signal subtraction means, respectively,
It is possible to realize a stereo microphone that distributes superdirective directional characteristics to the left and right and collects distant sound in stereo.

Furthermore, by varying the signal delay amounts of the first signal delay means and the second signal delay means, it is possible to adjust the stereo feeling according to the sound field. The range of the signal delay amount τ 1 at this time is the range of the first super-directional microphone and the second
The distance between the sound holes of the super directional microphone of
Then, 0 <τ1 ≦ d / c.

Further, the first super-directional microphone and the second super-directional microphone are waveform type super-directional microphones having sound holes on the side surface of the acoustic tube, and By providing the sound holes of the second super-directional microphone in opposite directions so that the distance between the sound holes is maximized, the distance between the sound holes can be increased in a limited size, and the low sound distance can be reduced. The characteristics of the range can be improved.

Furthermore, the first super-directional microphone and the second super-directional microphone are provided in parallel with each other with their main axes facing the front direction, and the first super-directional microphone receives the output signal from the first super-directional microphone as an input. Signal delaying means, second signal delaying means for inputting the output signal from the second super-directional microphone, and third signal delaying means for inputting the output signal from the first super-directional microphone. A fourth signal delay means for receiving an output signal from the second superdirective microphone, a first amplitude / phase correcting means for receiving an output signal from the first signal delay means, and a second Second amplitude / phase correction means that receives the output signal from the signal delay means, and the first amplitude-phase correction means that receives the output signal from the third signal delay means and the output signal from the second amplitude / phase correction means. A signal subtracting means and a fourth Compensating for variations in the characteristics of the first and second superdirective microphones, the second signal subtracting means having the output signal from the signal delay means and the output signal from the first amplitude / phase correcting means as inputs. Thus, the signals from the specific direction are surely subtracted in the first and second signal subtraction means, and the directivity can be improved.

Further, in addition to the above-mentioned structure, a cross-correlation calculating means for inputting an output signal from the first super-directional microphone and an output signal from the second super-directional microphone is provided, and the first amplitude phase The correcting means is replaced with the first adaptive amplitude / phase correcting means, and the amplitude / phase characteristic thereof is updated based on the output signal from the second signal subtracting means and the output signal from the cross-correlation calculating means to obtain the second amplitude. The phase correcting means is replaced with the first adaptive amplitude / phase correcting means, and the amplitude / phase characteristic is updated based on the output signal from the first signal subtracting means and the output signal from the cross-correlation calculating means. As a result, the first adaptive amplitude phase correcting means and the second adaptive amplitude phase correcting means can realize appropriate characteristics by self-learning.

[Brief description of drawings]

FIG. 1 is a configuration diagram of a stereo microphone according to a first embodiment of the present invention.

FIG. 2 is a layout diagram of first and second superdirective microphones.

FIG. 3 is a directional characteristic diagram of first and second superdirective microphones and a stereo microphone of the present invention.

FIG. 4 is a layout diagram of sound holes of the first and second superdirective microphones.

FIG. 5 is a configuration diagram of a stereo microphone according to a second embodiment of the present invention.

FIG. 6 is a configuration diagram of a stereo microphone according to a third embodiment of the present invention.

FIG. 7 is a configuration diagram and directional characteristic diagram of a conventional example.

[Description of Reference Signs] 1 first super-directional microphone 2 second super-directional microphone 11 first signal delay means 12 second signal delay means 13 third signal delay means 14 fourth signal delay means 21 First amplitude / phase correction means 22 Second amplitude / phase correction means 23 First adaptive amplitude / phase correction means 24 Second adaptive amplitude / phase correction means 31 First signal subtraction means 32 Second signal subtraction means 40 Cross-correlation Computing means 51 First output terminal 52 Second output terminal

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Hiroki Furukawa 1006 Kadoma, Kadoma City, Osaka Prefecture Matsushita Electric Industrial Co., Ltd.

Claims (8)

[Claims] 1. A first super-directional microphone and a second super-directional microphone are provided in parallel with each other with their main axes facing the front direction, and a first super-directional microphone is provided after the first super-directional microphone. A signal delaying unit, and a second signal delaying unit provided after the second superdirective microphone,
First signal subtraction means for inputting an output signal from the first super-directional microphone and an output signal from the second signal delay means; an output signal from the second super-directional microphone; A stereo characterized by comprising a second signal subtracting means for receiving an output signal from the first signal delaying means, and obtaining an output signal from each of the first signal subtracting means and the second signal subtracting means. Microphone. 2. The stereo microphone according to claim 1, wherein the signal delay amounts of the first signal delay means and the second signal delay means are variable. 3. A signal delay amount .tau.1 of the first signal delay means and the second signal delay means, where d is the distance between the sound holes of the first superdirective microphone and the second superdirective microphone.
Is 0 <τ1 ≦ d / c, where c is the speed of sound. 4. The stereo microphone according to claim 1, wherein the first superdirective microphone and the second superdirective microphone are waveform-type superdirective microphones in which sound holes are provided on the side surface of the acoustic tube. 5. The stereo microphone according to claim 4, wherein the sound holes of the first super-directional microphone and the second super-directional microphone are provided on opposite sides of the other super-directional microphone, respectively. 6. A first super-directional microphone and a second super-directional microphone are provided in parallel with each other with their main axes facing forward, and the output signal from the first super-directional microphone is input. One signal delay means, a second signal delay means for receiving an output signal from the second super-directional microphone, and a third signal delay means for receiving an output signal from the first super-directional microphone. A signal delay means, a fourth signal delay means for receiving an output signal from the second superdirective microphone, and a first amplitude / phase correction for receiving an output signal from the first signal delay means. Means, a second amplitude / phase correction means for receiving the output signal from the second signal delay means, an output signal from the third signal delay means and an output from the second amplitude / phase correction means Signal input Signal subtraction means, and second signal subtraction means to which the output signal from the fourth signal delay means and the output signal from the first amplitude / phase correction means are input, and the first signal subtraction means And a stereo microphone, wherein an output signal is obtained from each of the second signal subtracting means. 7. A first super-directional microphone and a second super-directional microphone are provided with their main axes parallel to each other in the front direction, and a first super-directional microphone receives an output signal from the first super-directional microphone as an input. Signal delay means, a second signal delay means for receiving an output signal from the second super-directional microphone, and a third signal for receiving an output signal from the first super-directional microphone. A delay means and a fourth means for receiving an output signal from the second superdirective microphone
Signal delay means, a first adaptive amplitude / phase correction means for receiving an output signal from the first signal delay means, and a second adaptive for receiving an output signal from the second signal delay means. Amplitude and phase correction means, first signal subtraction means that receives the output signal from the third signal delay means and the output signal from the second adaptive amplitude and phase correction means, and the fourth signal delay means. Output signal from the first adaptive amplitude phase correction means and second signal subtraction means that receives the output signal from the first adaptive amplitude and phase correction means, and the output signal from the first superdirective microphone and the second superdirectivity. And a cross-correlation calculation means using the output signal from the microphone as an input signal,
The characteristic of the first adaptive amplitude / phase correcting means is sequentially updated by the output signal from the cross-correlation calculating means and the output signal from the second signal subtracting means, and the characteristic of the second adaptive amplitude / phase correcting means is the mutual signal. A stereo microphone which is sequentially updated by an output signal from the correlation calculating means and an output signal from the first signal subtracting means. 8. The signal delay amounts of the third signal delay means and the fourth signal delay means are equal to each other at .tau.2, and are equal to the signal delay amounts of the first amplitude phase correction means and the second amplitude phase correction means. Τ 2 is set so that the signal delay amounts of the first signal delay means and the second signal delay means are equal at τ 1, and τ 1 is 0 <τ 1 ≦ d / c, where d is the first superdirective microphone. Sound hole and second
8. The stereo microphone according to claim 6 or 7, wherein the distance between the sound holes of the super-directional microphone is c: the speed of sound.