JP2023026927A - Close-talking microphone array and setting method for close-talking microphone array - Google Patents

Abstract

To provide a close-talking microphone array and a setting method for the close-talking microphone array capable of preferentially detecting sound from a near sound source.SOLUTION: The sensitivity and terminal polarity of each of a plurality of microphones 20-1, 20-2, 20-3, ..., and 20-N are set on the basis of the calculated eigenvalues and eigenvectors calculated from a generalized eigenvalue problem for a matrix whose elements are real parts of the specific acoustic impedance on the surface of a microphone array 20, which is a set of the plurality of microphones 20-1, 20-2, 20-3, ..., and 20-N, and a matrix whose elements are the imaginary parts of the specific acoustic impedance.SELECTED DRAWING: Figure 1

Description

The present invention relates to a close-talking microphone array and a method of setting a close-talking microphone array.

When talking on a cell phone or the like in a noisy place, the speaker's voice is mixed with ambient noise, making it difficult for the other party to hear the speaker's voice. ing. For example, there is signal processing of estimating a frequency spectrum related to noise from a microphone measurement signal in which noise is mixed in speech, and subtracting the estimated frequency spectrum value from the microphone measurement signal. Such signal processing removes noise from speech by subtracting a "noise estimate" from a microphone measurement signal that is "speech plus noise".

However, it is generally not easy to estimate the noise-related frequency spectrum from a signal containing both speech and noise, and an estimation error may occur. If the estimated value including the estimation error is subtracted from the measured signal of the microphone, the speech may be distorted and the clarity of the call may be deteriorated. Moreover, if the removal of the noise component from the measurement signal of the microphone is suppressed for fear of degrading the clarity of speech, the noise may not be sufficiently removed. Therefore, when clarifying speech by removing the noise component from the measurement signal of the microphone, it is a technical problem to improve the accuracy of estimating the frequency spectrum associated with the noise.

In Patent Document 1, the direct ratio, which is the ratio of the direct sound and the indirect sound (reverberant sound) included in the received sound, changes monotonously according to the distance between the microphone and the sound source. An invention is disclosed relating to a microphone array that estimates the distance to a sound source and eliminates noise by filtering that emphasizes or suppresses the amplitude of the component of the sound source determined to be within a certain distance range.

JP 2011-53062 A

However, the invention described in Patent Document 1 has the problem that the process of estimating the distance to the microphone for each sound source and emphasizing or suppressing the amplitude of the signal associated with each sound source according to the estimated distance is complicated. was there.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a close-talking microphone array and a setting method of the close-talking microphone array that can preferentially detect sounds from nearby sound sources.

In order to solve the above-mentioned problems, the invention according to claim 1 provides a matrix whose elements are real parts of specific acoustic impedance on the surface of a microphone array, which is a set of a plurality of microphones, and whose elements are imaginary parts of the specific acoustic impedance. Based on the eigenvalues and eigenvectors calculated from the generalized eigenvalue problem for the matrix and the polarities of the terminals of each of the plurality of microphones, the sensitivities of the plurality of microphones are set.

According to the present invention, the sensitivity and terminal polarity of each of the multiple microphones can be optimized based on the eigenvalues and eigenvectors calculated from the generalized eigenvalue problem.

In addition, as described in claim 2, the sensitivity and terminal polarity of each of the plurality of microphones is the product of a matrix whose elements are the imaginary parts of the specific acoustic impedance and a real matrix whose columns are the eigenvectors. , the element of the eigenvector whose eigenvalue is the maximum when it is equal to the product of a matrix whose elements are the real part of the specific acoustic impedance, a real matrix whose columns are the eigenvectors, and a diagonal matrix whose elements are the eigenvalues may be set based on

In addition, as described in claim 3, the sensitivity and terminal polarity of each of the plurality of microphones are the product of a matrix having the real part of the specific acoustic impedance as an element and a real matrix having the eigenvectors as columns. , the element of the eigenvector whose eigenvalue is the minimum value when it is equal to the product of the matrix whose elements are the imaginary part of the specific acoustic impedance, the real matrix whose columns are the eigenvectors, and the diagonal matrix whose elements are the eigenvalues may be set based on

In the invention according to claim 4, the plurality of microphones are linearly arranged.

According to the fifth aspect of the invention, the plurality of microphones are arranged at intervals of half the wavelength or less of the upper limit frequency to be measured by each of the plurality of microphones.

According to the sixth aspect of the invention, there is provided a method for setting a close-talking microphone array, wherein a matrix whose elements are real parts of the specific acoustic impedance on the surface of the microphone array, which is a set of a plurality of microphones, and an imaginary part of the specific acoustic impedance calculating eigenvalues and eigenvectors from a generalized eigenvalue problem related to a matrix having and includes

Advantageous Effects of Invention According to the present invention, it is possible to provide a close-talking microphone array and a setting method of the close-talking microphone array capable of preferentially detecting sounds of a nearby sound source.

2 is a block diagram showing an example of the configuration of a microphone array; FIG. FIG. 4 is a block diagram showing a microphone array with five microphones; (A) is two microphones, (B) is three microphones, (C) is four microphones, and (D) is five microphones. be. Eigenvalue for each frequency of (A) when there are two microphones, (B) when there are three microphones, (C) when there are four microphones, and (D) when there are five microphones. is an explanatory diagram showing a change in . (A) is for two microphones, (B) is for three microphones, (C) is for four microphones, and (D) is for five microphones. FIG. 10 is an explanatory diagram showing an example of changes in next eigenvectors; An example of the sensitivity distribution of a microphone array in which two microphones are linearly arranged, (A) at a frequency of 300 Hz, (B) at a frequency of 1850 Hz, and (C) at a frequency of 3400 Hz. It is an explanatory diagram shown. An example of the sensitivity distribution of a microphone array in which three microphones are linearly arranged, (A) at a frequency of 300 Hz, (B) at a frequency of 1850 Hz, and (C) at a frequency of 3400 Hz. It is an explanatory diagram shown. An example of the sensitivity distribution of a microphone array in which four microphones are linearly arranged, (A) at a frequency of 300 Hz, (B) at a frequency of 1850 Hz, and (C) at a frequency of 3400 Hz. It is an explanatory diagram shown. An example of the sensitivity distribution of a microphone array in which five microphones are linearly arranged, (A) at a frequency of 300 Hz, (B) at a frequency of 1850 Hz, and (C) at a frequency of 3400 Hz. It is an explanatory diagram shown. FIG. 4 is an explanatory diagram showing an example of sensitivity distribution of one microphone; (A) is 300 Hz, (B) is 1850 Hz, and (C) is 3400 Hz. It is an explanatory diagram. FIG. 4A is an explanatory diagram showing the frequency sensitivity characteristic of one microphone, and FIG. 4B showing the frequency sensitivity characteristic of a microphone array in which five microphones are linearly arranged;

Embodiments of the present invention will be described below. In this embodiment, a case where the present invention is applied to a close-talking microphone array will be described.

This embodiment proposes an approach of constructing a microphone that is less likely to pick up noise, rather than removing noise components from the measurement signal of the microphone by post-processing. In close-talking devices such as mobile phones and headsets, the distance between the speaker's mouth and the microphone is short, but the distance between the surrounding noise sources and the microphone is long. Therefore, a microphone that measures sound emitted from a nearby sound source with high sensitivity and measures sound emitted from a distant sound source with low sensitivity can preferentially detect a speaker's voice.

Gradient mics (differential mics) are examples of technologies already in practical use. A gradient microphone arranges two microphones and detects the difference between the two measurement signals. The detected difference is proportional to the sound pressure gradient. The gradient of the sound pressure is the sensitivity characteristic that emphasizes the nearby sound source rather than the sound pressure itself. Also, by providing more microphones and obtaining higher-order gradients, nearby sound sources can be further emphasized. However, while sound pressure gradients do enhance nearby sound sources, sound pressure gradients do not always maximize nearby sound sources.

In this embodiment, eigenvalues and eigenvectors relating to the specific acoustic impedance of the surface of the microphone array are calculated, and based on the calculated eigenvalues and eigenvectors, amplitudes (hereinafter referred to as “sensitivities”) indicating the sensitivity of each microphone constituting the microphone array ) and the phase indicating the electrical polarity of each microphone terminal (hereinafter referred to as “polarity”) to effectively detect sounds from nearby sound sources and make it difficult to pick up ambient noise.

FIG. 1 is an explanatory diagram showing the configuration of a sound input device 10 according to this embodiment. The acoustic input device 10 has a microphone array 20 and an amplifier array 30 . The microphone array 20 includes N microphones 20-1, 20-2, 20-3, . . . , 20-N linearly arranged. Here, N is an integer of 2 or more. As a preparation for driving the microphone array 20, the predetermined eigenvectors of the microphone array 20 are assumed to have been calculated. The eigenvector has N vector elements corresponding to the sensitivity and polarity of each of the N microphones 20-1, 20-2, 20-3, . . . , 20-N. For example, the i-th vector element of the N vector elements of the eigenvector indicates the sensitivity and polarity of the i-th microphone 20-i in the microphone array 20. FIG. Here, i is an integer of 1 or more and N or less. The amplifier array 30 includes N amplifiers 30-1, 30-2, 30-3, . . . , 30-N. The i-th amplifier 30-i of the amplifier array 30 controls the sensitivity and polarity of the i-th microphone 20-i of the microphone array 20 according to the i-th vector element of the N vector elements of the eigenvector.

In this embodiment, the amplifier array 30 adjusts the sensitivity and polarity of each of the microphones 20-1, 20-2, 20-3, . . . , 20-N. The sensitivity and polarity of each of 1, 20-2, 20-3, . . . , 20-N may be arbitrarily set.

Each of the microphones 20-1, 20-2, 20-3, . . . , 20-N is either a dynamic microphone, a condenser microphone, a carbon microphone, or a piezoelectric microphone.

First, consider a speaker array instead of a microphone array. The sound (complex acoustic power) emitted from the speaker array is divided into a long-range component (real part of complex acoustic power) and a short-range component (imaginary part of complex acoustic power). The real part of the complex acoustic power is called active acoustic power. On the other hand, the imaginary part of the complex acoustic power is called reactive acoustic power. If the ratio of reactive sound power to active sound power can be increased, sound can be delivered only to nearby areas. The eigenvalue of the generalized eigenvalue problem described below corresponds to the ratio of reactive acoustic power to active acoustic power. Eigenvectors of the generalized eigenvalue problem, which will be described later, correspond to the amplitude and polarity of each speaker that constitutes the speaker array. In other words, by setting the amplitude and polarity of each speaker according to the eigenvector having the maximum eigenvalue, it is possible to realize a speaker array that easily delivers sound only to nearby areas. Due to reciprocity, this idea can be applied to the design of microphone arrays. That is, by setting the sensitivity and polarity of each microphone according to the eigenvector having the maximum eigenvalue, it is possible to realize a microphone array that can easily measure only sounds emitted from nearby sound sources.

Now assume N loudspeakers instead of microphones. Let p be the complex sound pressure vector, which is a column vector whose vector element is the complex sound pressure on the surface of each of the N speakers, and let the complex velocity of each of the N speakers (vibration velocity in the normal direction of the vibration surface) be If v is a complex velocity vector that is a column vector with vector elements, and Z is a specific acoustic impedance matrix that has a specific acoustic impedance of the speaker surface as a matrix element, the following equation (1) holds. The specific acoustic impedance matrix Z becomes a symmetric matrix due to acoustic reciprocity. Here, if p is an N-by-1 vector and v is an N-by-1 vector, then Z is an N-by-N matrix. The i-th row and j-th column element of Z is then the specific acoustic impedance between the i-th and j-th speaker surfaces.

The complex acoustic power W is represented by the following formula (2). s in Equation (2) is the area of the vibration surface of each speaker. For convenience of explanation, in equation (2), it is assumed that the areas of the vibration surfaces of the speakers are the same, but the areas of the vibration surfaces of the speakers do not necessarily have to be the same.
Also, v ^H is the conjugate transpose of the complex velocity vector v.

The complex acoustic power W includes a real part W _r and an imaginary part W _i as follows. In the equations below, W _r denotes active sound power and W _i denotes reactive sound power.

The specific acoustic impedance matrix Z also includes a real part Z _r and an imaginary part Z _i as follows. In the equations below, Z _r denotes an acoustic resistance matrix and Z _i denotes an acoustic reactance matrix.

Therefore, the real part W _r and the imaginary part W _i of the complex acoustic power W are given by the following equations (3) and (4).

In a speaker array, a configuration suitable for listening in the near field can be obtained by increasing the RA ratio, which is the ratio of reactive acoustic power W _i to active acoustic power W _r (W _i /W _r ). . As will be described later, the RA ratio matches the eigenvalues in the generalized eigenvalue problem, and sets the sensitivity and polarity of the individual microphones that make up the microphone array 20 based on the eigenvector that maximizes the eigenvalue.

Define a generalized eigenvalue problem on the acoustic resistance matrix and the acoustic reactance matrix, as shown in equation (5) below. In the following equations, Λ is a diagonal matrix whose elements are eigenvalues, and Φ is a real matrix whose columns are eigenvectors. The acoustic resistance matrix Z _r is a real positive definite symmetric matrix, and the acoustic reactance matrix Z _i is a real symmetric matrix. Also, in the generalized eigenvalue problem, eigenvalues are real numbers and eigenvectors are real vectors. The generalized eigenvalue problem is established even if the acoustic reactance matrix Z _i and the acoustic resistance matrix Z _r are interchanged on the left side and the right side in the following equation (5). However, the eigenvalue of the generalized eigenvalue problem when the acoustic reactance matrix Z _i and the acoustic resistance matrix Z _r are exchanged in Equation (5) is the reciprocal of the RA ratio in the original Equation (5). Therefore, in such a case, the sensitivity and polarity of the individual microphones forming the microphone array 20 are set based on the eigenvector that minimizes the eigenvalue.

In solving the generalized eigenvalue problem, we first solve the standard eigenvalue problem for the acoustic resistance matrix Z _r . The diagonal matrix Λ _r whose elements are real and positive eigenvalues and the orthogonal matrix Φ _r whose columns are real eigenvectors satisfy the following equation (6).

In equation (6) above, Φ ^T _r is the transposed matrix of Φ _r . By using Φ _r Λ _r ^−1/2 , the acoustic reactance matrix Z _i is transformed as shown in Equation (7) below.

In equation (7) above, -1/2 means the reciprocal of the square root of the diagonal element. Z _i ' is a symmetric matrix that is real. Then solve the standard eigenvalue problem for Z _i '. A diagonal matrix Λ in which real-number eigenvalues are arranged in each diagonal element and an orthogonal matrix Φ _i in which real-number eigenvectors are arranged in each column satisfy the following equation (8).

Defining Φ, which is a real matrix, as in Equation (9) below enables diagonalization as shown in Equations (10) and (11).

I in Equation (10) is a unit matrix. Multiplying both sides of equation (10) by Λ from the right side and comparing the result with equation (11) yields equation (5). The complex acoustic power W can be expressed as in Equation (13) by the diagonalization shown in Equations (10) and (11).

A speaker array composed of a plurality of speakers has a plurality of eigenvalues and eigenvectors, but when designing a speaker array based only on the eigenvector _φm of a specific order m (mth order), the complex velocity vector v is as follows: It is shown by Formula (12). However, u _m is an arbitrary coefficient.

At this time, the complex acoustic power W is given by the following equation (13).

The mth-order eigenvalue λ _m means the RA ratio, which is the ratio of the reactive acoustic power W _i to the active acoustic power W _r . Also, the eigenvector φ _m means the amplitude and phase of each loudspeaker (this eigenvector is also called generalized radiation mode). Therefore, to maximize the RA ratio, set the amplitude and phase of each speaker according to the eigenvector with the largest eigenvalue. Since both the eigenvalue λ _m and the eigenvector φ _m are real numbers, setting the phase only requires setting the positive or negative electrical polarity of the speaker terminal.

We will apply the above speaker array theory to a close-talking microphone array. First, considering equation (12) in the speaker array, maximizing the RA ratio means setting the amplitude and phase of each speaker according to the eigenvector with the largest eigenvalue.

Here, the complex sound pressure p(r _o ) observed at an arbitrary position r _o is represented by the following equation (14). In Equation (14) below, z is the row vector of specific acoustic impedance, which is the acoustic transfer function between the sound source and the observation point. The sound pressure level is expected to be higher when closer to the speaker array and lower when farther away.

Next, when the sound source and the observation point are exchanged, the following formula (15) holds due to reciprocity. q(r _o )=Sv in equation (15) is the complex volume velocity (m ³ /s) of the sound source located at r _o , where v corresponds to S being the surface area of this sound source. is the complex velocity for

p in Equation (15) is a column vector whose vector elements are the complex sound pressures observed by the microphones forming the microphone array. Then, filter processing represented by the following equation (16) is performed.

φ ^T _m p on the left side of Equation (16) is proportional to φ ^T _m z ^T and φ ^T _m z ^T =zφ _m , and from Equation (14), the complex sound pressure p(r _o ) is proportional to zφ _m . Therefore, the sensitivity of the microphone array is expected to be high when the microphone array is close to the sound source and low when far from the sound source.

In this embodiment, the sensitivity and polarity of each microphone constituting the microphone array are determined by solving a generalized eigenvalue problem related to the acoustic resistance matrix Z _r and the acoustic reactance matrix Z _i . The optimal filter φ ^T _m in the microphone array is obtained as the eigenvector with the maximum eigenvalue. The specific acoustic impedance is frequency dependent, and the eigenvalues and eigenvectors are frequency dependent.

In this embodiment, the eigenvector associated with the maximum eigenvalue indicates the sensitivity (amplitude) and polarity (phase) of each microphone that constitutes the microphone array. As mentioned above, since the eigenvalues and eigenvectors consist of real numbers, adjusting the phase of an individual microphone only requires setting either positive or negative sign.

Next, the operation of the microphone array 20 according to this embodiment will be described. In this embodiment, the characteristics of the microphone array 20 are analyzed using a simulation model. In the simulation, the specific acoustic impedance z of the surface of the microphone array 20 is calculated as the specific acoustic impedance of the surface of the speaker array having the same geometrical shape as the microphone array 20 by the following formula (17).

In equation (17) above, ρ is the air density, ρ = 1.21 kg/m ³ , k is the wavenumber, k = ω/c, ω is the angular frequency, and c is the speed of sound in air. , where c=340 m/s and r is the distance between the sound source and the observation point.

As mentioned above, the i-th row, j-th column element of the specific acoustic impedance matrix Z is the specific acoustic impedance between the i-th speaker surface and the j-th speaker surface. The z on the left side of the equation (17) indicates a matrix element of the specific acoustic impedance matrix Z. The r on the right side of equation (17) is the distance between the i-th speaker surface and the j-th speaker surface. When calculating the off-diagonal elements (i≠j) of the matrix Z, since r≠0, Equation (17) having r as the denominator on the right side can be applied. However, when calculating the diagonal elements (i=j) of the matrix Z, since r=0, the calculation result diverges in Equation (17) having r in the denominator on the right side. In such a case, z is calculated using the following equation (18) instead of the above equation (17).

As a peculiar case, Equation (18) for calculating the specific acoustic impedance z at r=0 is as follows. In equation (18), l _p is the length of one side of the microphone diaphragm assumed to be square, and S=l _p ² (for example, l _p =0.005 m).

FIG. 2 is an explanatory diagram showing an example of the microphone array 20 according to the embodiment of the invention. The microphone array 20 shown in FIG. 2 is a simulation model for numerical analysis of eigenvalues and eigenvectors. In this example, the microphone array 20 is arranged on the X axis of the XY plane (infinite baffle) in the XYZ space. It has five microphones 20-1, 20-2, . . . , 20-5.

The spacing of each microphone in the X direction is L _x . As will be described later, in the numerical analysis simulation, L _x =0.015 m.

FIG. 3 is an explanatory diagram showing an example of the configuration of the microphone array 20 in which the respective microphones are linearly arranged in the simulation of this embodiment. 3(A) shows the case of two microphones, FIG. 3(B) shows the case of three microphones, FIG. 3(C) shows the case of four microphones, and FIG. 3(D) shows the case of five microphones. Each case is shown.

In any case of FIGS. 3A to 3D, the center of the microphone array coincides with the origin O of the x-axis. The interval between adjacent microphones is 0.015 m, which is the half-wave length of the sound wave at 3400 Hz, which is the upper limit frequency to be measured. Generally, the frequency band for voice calls is from 300 Hz to 3400 Hz, and 3400 Hz corresponds to the upper limit of the frequency band.

FIG. 4 is an explanatory diagram showing changes in eigenvalues with respect to frequencies of the microphone array 20 in which the respective microphones are linearly arranged in a simulation according to this embodiment. 4A shows the case of two microphones, FIG. 4B shows the case of three microphones, FIG. 4C shows the case of four microphones, and FIG. 4D shows the case of five microphones. , respectively.

The generalized radiation mode is composed of radiation modes of orders corresponding to the number of microphones forming the microphone array 20 . For example, if the microphone array 20 has two microphones, it has a primary radiation mode and a secondary radiation mode, and if there are three microphones, it has a primary radiation mode and a secondary radiation mode. and a third-order radiation mode.

In FIG. 4, numbers are assigned to primary, secondary, tertiary, . . . in descending order of eigenvalues. Therefore, the smaller the order, the larger the eigenvalue, and the first-order eigenvalue is the maximum eigenvalue. Also, the greater the number of microphones forming the microphone array 20, the greater the maximum eigenvalue.

Therefore, as the number of microphones forming the microphone array 20 increases, the RA ratio in the microphone array 20 can be increased.

FIG. 5A shows the case with two microphones, FIG. 5B shows the case with three microphones, FIG. 5C shows the case with four microphones, and FIG. 5D shows the case with five microphones. FIG. 10 is an explanatory diagram showing an example of changes in eigenvectors at frequencies of 300 Hz, 1850 Hz, and 3400 Hz in each of cases 1 and 2;

Each of FIGS. 5A-5D is normalized by the largest absolute value in the vector element of each eigenvector. Each of the eigenvectors also indicates the amplitude (sensitivity) and phase (polarity) of each individual microphone that constitutes the microphone array 20 . Specifically, the numerical values of the vector elements of the eigenvectors indicate the amplitudes (sensitivities) of the individual microphones, and the signs of the vector elements of the eigenvectors indicate the phases (polarities) of the individual microphones. For example, if the absolute value of the numerical value of the vector element of the eigenvector is large, the microphone is required to have high sensitivity. This indicates that the polarity of the positive terminal is reversed.

FIG. 5A shows that a two-microphone microphone array 20 results in a first-order gradient microphone.

FIG. 5B shows that a microphone array 20 consisting of three microphones results in a quadratic gradient microphone consisting of three microphones.

FIG. 5C shows that the microphone array 20 consisting of four microphones is different from the quadratic gradient microphone consisting of four microphones. The output p of the quadratic gradient microphone is given by the following equation. x ₁ , x ₂ , x ₃ , and x ₄ in the equations below are the x-coordinates of each microphone as shown in FIG. 3(C). The difference in sensitivity between the microphone array 20 composed of four microphones and the quadratic gradient microphone will be described later. p=p(x ₁ ,0,0)−p(x ₂ ,0,0)−p(x ₃ ,0,0)+p(x ₄ ,0,0)

FIG. 5(D) means the sensitivity and polarity of each microphone constituting the microphone array 20 composed of five microphones.

As shown in FIGS. 5A to 5D, the eigenvector with the maximum eigenvalue is almost frequency-independent. Therefore, the eigenvector at a certain frequency can be used to preferentially detect the sound of a near sound source at a practically sufficient level in a certain frequency range including that frequency. This eliminates the need to control the amplitude (sensitivity) and phase (polarity) of each microphone according to an eigenvector that differs for each frequency, thereby simplifying the procedure for constructing the microphone array 20 .

6, 7, 8 and 9 each show the sensitivity distribution at frequencies 300 Hz, 1850 Hz and 3400 Hz of the microphone array 20 on the xz plane with y=0 in the positional relationship shown in FIG. It is an explanatory diagram. In each of FIGS. 6, 7, 8, and 9, darker areas have lower sensitivity, and brighter areas have higher sensitivity.

Each of FIGS. 6, 7, 8 and 9 plots the numerical value in dB normalized by the maximum value of |φ ^T _m p| in the xz plane. ^Specifically _{, if} ^the _maximum ^value _of |φ ^T _m p| values are plotted.

6A shows a microphone array with a frequency of 300 Hz, FIG. 6B shows a frequency of 1850 Hz, and FIG. 6C shows a microphone array with a linear arrangement of two microphones. is an explanatory diagram showing an example of the sensitivity distribution of . As shown in FIGS. 6(A) to (C), at each frequency, there is an undesirable knot-like region of low sensitivity at the center 0 m of the x-axis, but the sensitivity is highest at the center region of the x-axis excluding 0 m. is high, and the sensitivity gradually decreases with increasing distance from the central region in the x-axis direction or the z-axis direction. The nodular low-sensitivity regions are also recognized in FIGS. 8A, 8B, and 8C, which will be described later.

7A shows a microphone array with a frequency of 300 Hz, FIG. 7B shows a frequency of 1850 Hz, and FIG. 7C shows a microphone array with a linear arrangement of three microphones. is an explanatory diagram showing an example of the sensitivity distribution of .

8A shows a microphone array with a frequency of 300 Hz, FIG. 8B shows a frequency of 1850 Hz, and FIG. 8C shows a microphone array with a linear arrangement of four microphones. is an explanatory diagram showing an example of the sensitivity distribution of .

9A shows a microphone array with a frequency of 300 Hz, FIG. 9B shows a frequency of 1850 Hz, and FIG. 9C shows a microphone array with a linear arrangement of five microphones is an explanatory diagram showing an example of the sensitivity distribution of .

In FIGS. 7 to 9, there is a slight difference in sensitivity due to the difference in frequency, and especially at a frequency of 300 Hz, the sensitivity at a short distance from the microphone is remarkably high at the position of 0 m where the microphone is arranged. It also shows that as the number of microphones forming the microphone array 20 increases to 3, 4, and 5, the areas with high sensitivity are limited to areas closer to the microphones.

FIG. 10 is an explanatory diagram showing an example of the sensitivity distribution of one microphone arranged at the origin O. FIG. In the case of one microphone, no significant change due to frequency is observed, so only the case of the frequency of 300 Hz is exemplified.

Comparing FIG. 10, which shows the case of one microphone, with FIGS. 10, the cases of FIGS. 6A to 6C are similar to the first-order gradient microphone. However, in the case of FIGS. 6A to 6C, an undesirable knot-shaped low sensitivity region occurs around the origin O as described above. (8), FIG. 8(B) and FIG. 8(C) for the case of .

In the case of an even number of microphones, the regions of low sensitivity around the origin O as shown in FIGS. 6A to 6C and FIGS. As shown in FIG. 5A in the case of four microphones and FIG. Moreover, as described above, as the number of microphones constituting the microphone array 20 increases to 3, 4, and 5, the area with high sensitivity is limited to the area closer to the microphones. An odd number of microphones is desirable, and a greater number of microphones constituting the microphone array 20, such as five rather than three, is expected to provide better characteristics as a close-talking microphone array.

Fig. 11 (A) shows the frequency at 300 Hz, Fig. 11 (B) shows the frequency at 1850 Hz, and Fig. 11 (C) shows the frequency of the second-order gradient microphone composed of four microphones at 3400 Hz. FIG. 4 is an explanatory diagram showing an example of sensitivity distribution; The quadratic gradient microphone has an output of p(x ₁ ,0,0)−p(x ₂ ,0,0)−p(x ₃ ,0,0)+p(x ₄ ,0,0) .

8A to 8C showing the sensitivity characteristics of the microphone array 20 configured by the four microphones according to the present embodiment and FIG. 8A to 8C, the microphone array 20 obtained by this method has a more remarkable contrast in sensitivity near the origin O than the second-order gradient microphones shown in FIGS. 11A to 11C. The microphone array 20 shown has areas of high sensitivity concentrated near the origin O. FIG.

FIG. 12(A) shows the frequency sensitivity characteristics of one microphone placed at the origin O, and FIG. 12(B) shows the third microphone from the left and right as shown in FIG. FIG. 5 is an explanatory diagram showing sensitivity characteristics with respect to frequencies of a microphone array in which five microphones are arranged in a line. FIGS. 12(A) and 12(B) show changes in sensitivity of the microphone array with respect to frequency when one point sound source exists at different distances in the z-axis direction from the microphone array. The coordinates of the point sound source are , 1m). If the output of the sound source is constant, the sound pressure is proportional to the frequency as shown in the above equation (17). Therefore, in this embodiment, the output of the sound source is operated so as to be inversely proportional to the frequency. As a result, in FIG. 12A, the response to frequency at one microphone is flat.

As shown in FIG. 12(A), the sensitivity of a single microphone decreases with the distance to the sound source, indicating distance attenuation of the spherical wave.

In the case of FIG. 12(B), similarly to the case of FIG. 12(A), the operation is performed so that the output of the sound source is inversely proportional to the frequency. As shown in FIGS. 12(A) and 12(B), in FIG. 12(B) showing the characteristics of the microphone array 20 according to the present embodiment, distance attenuation is more pronounced than in the case of FIG. 12(A). It has become. In addition, the response to frequencies is almost flat at a close distance from the sound source, but exhibits characteristics similar to a high-pass filter in which the sensitivity in the low frequency range decreases as the distance from the sound source increases.

An additional equalizer is required to flatten the response over frequency.

As described above, the microphone array 20 according to the present embodiment sets the sensitivity and polarity of each microphone according to the eigenvector that maximizes the eigenvalue. It is possible to realize the microphone array 20 having characteristics suitable for close-talking, in which the sensitivity is high and the sensitivity is low when the microphone array 20 and the sound source are distant.

The generalized eigenvalue problem is established even if the acoustic reactance matrix Z _i and the acoustic resistance matrix Z _r are interchanged on the left side and the right side in the above equation (5). The eigenvalue of the normalized eigenvalue problem is the inverse of the RA ratio in the original equation (5). Therefore, in such a case, the sensitivity and polarity of the individual microphones forming the microphone array 20 are set based on the eigenvector that minimizes the eigenvalue. In this embodiment, even when the generalized eigenvalue problem is handled based on Equation (5), the acoustic reactance matrix Z _i and the acoustic resistance matrix Z _r in Equation (5) are replaced by the left side and the right side. Also in this case, the sensitivity and polarity of each microphone are set based on the eigenvector when the eigenvalue is an extreme value.

In summary, in the generalized eigenvalue problem in Equation (5), the product of the acoustic reactance matrix _Zi , which is a matrix whose elements are the imaginary parts of the specific acoustic impedance, and the real matrix Φ, whose columns are the eigenvectors, is the specific acoustic impedance The element of the eigenvector whose eigenvalue is the maximum when it is equal to the product of the acoustic resistance matrix _Zr , which is a matrix whose elements are real parts, the real matrix Φ whose columns are eigenvectors, and the diagonal matrix Λ whose elements are eigenvalues. Set the sensitivity and polarity of each microphone based on The eigenvalue is maximized when the reactive acoustic power W _i is maximized with respect to the active acoustic power W _r .

In addition, in the generalized eigenvalue problem when the acoustic reactance matrix Z _i and the acoustic resistance matrix Z _r are interchanged on the left side and the right side in Equation (5), the product of the acoustic resistance matrix Z _r and the real matrix Φ is equal to the product of the acoustic reactance matrix Z _i , the real matrix Φ and the diagonal matrix Λ. The eigenvalue is the minimum when the active acoustic power W _r is the minimum with respect to the reactive acoustic power W _i .

A microphone array composed of a plurality of microphones includes primary gradient microphones or secondary gradient microphones. It does not take into account the optimal settings of the individual microphones and does not have the characteristics suitable for close calls as compared to the microphone array 20 according to the present embodiment with four or five microphones.

In the microphone array 20 according to this embodiment, the number of microphones is also important. 6A to 6C, 8B and 8C, the microphone array 20 A phenomenon can occur in which the sensitivity of is significantly reduced.

To avoid such a phenomenon, the microphone array 20 is configured with an odd number of microphones, one microphone is arranged at the origin O, and the remaining microphones are arranged symmetrically with respect to the origin O. FIG. Such an arrangement can avoid the phenomenon that the sensitivity of the microphone array 20 at the origin O is remarkably lowered.

Also, even if the number of microphones constituting the microphone array 20 is the same odd number, five microphones have characteristics more suitable for close-talking than three.

The microphone array 20 according to the present embodiment has properties suitable for close-talking applications such as, but not limited to, telephones or headsets. In addition to close-talking, it also has characteristics as a near-field microphone array for picking up sounds such as the murmuring of a river and the flapping of insects without detecting the noise in the surroundings.

The embodiments described above are for facilitating the understanding of the present invention, and are not intended to limit and interpret the present invention. The present invention may be modified or improved without departing from its spirit, and the present invention also includes equivalents thereof. In other words, any design modifications made by those skilled in the art to the embodiments are also included in the scope of the present invention as long as they have the features of the present invention. Moreover, each element provided in the embodiment can be combined as long as it is technically possible, and a combination of these is also included in the scope of the present invention as long as it includes the features of the present invention.

10 Acoustic input device 20 Microphone array 20-1, 20-2, 20-3, 20-4, 20-5 Microphone 30 Amplifier array 30-1, 20-2, 20-3, 20-4, 20-5 Amplifier

Claims (6)

Eigenvalues and eigenvectors calculated from a generalized eigenvalue problem relating to a matrix whose elements are the real parts of the specific acoustic impedance and a matrix whose elements are the imaginary parts of the specific acoustic impedances on the surface of a microphone array, which is a set of a plurality of microphones A close-talking microphone array in which the sensitivity and terminal polarity of each of the plurality of microphones are set based on. The sensitivity and terminal polarity of each of the plurality of microphones is obtained by multiplying a matrix having the imaginary part of the specific acoustic impedance as an element and a real matrix having the eigenvector as the column, with the real part of the specific acoustic impedance as an element. 2. The eigenvalue is set based on the element of the eigenvector having the maximum value when the eigenvalue is equal to the product of a matrix having the eigenvectors as columns and a diagonal matrix having the eigenvalues as elements. close-talking microphone array. The sensitivity and terminal polarity of each of the plurality of microphones is obtained by multiplying a matrix having the real part of the specific acoustic impedance as an element and a real matrix having the eigenvector as the column, with the imaginary part of the specific acoustic impedance as an element. 2. The eigenvalue is set based on the element of the eigenvector having the minimum value when the eigenvalue is equal to the product of a matrix having the eigenvectors as columns and a diagonal matrix having the eigenvalues as elements. close-talking microphone array. The close-talking microphone array according to any one of claims 1 to 3, wherein the plurality of microphones are linearly arranged. 4. The close-talking microphone array according to claim 3, wherein the plurality of microphones are arranged at intervals equal to or less than half the wavelength of the upper limit frequency to be measured by each of the plurality of microphones. Eigenvalues and eigenvectors are calculated from a generalized eigenvalue problem relating to a matrix whose elements are real parts of specific acoustic impedances on the surface of a microphone array, which is a set of a plurality of microphones, and a matrix whose elements are imaginary parts of said specific acoustic impedances. a step;
setting the sensitivity and terminal polarity of each of the plurality of microphones based on the calculated eigenvalues and eigenvectors;
How to set up a microphone array for close-talk, including

Images