JP5488679B1 - Microphone array selection device, microphone array selection program, and sound collection device - Google Patents

Abstract

Even if a large number of microphone arrays are arranged in a space, a candidate for a combination of microphone arrays with a reduced amount of calculation and high area sound collection performance is selected.
The present invention provides a minimum overlap angle calculation unit that calculates a value obtained by dividing 180 ° by the number of microphone arrays used for area sound collection processing as a minimum overlap angle, position information of each microphone array, and information on a target area. Based on the position information, a directional intersection angle calculation unit that calculates a directional intersection angle between each microphone array with respect to the target area, calculates a difference between the minimum overlap angle and each directional intersection angle, and the value of the difference And a microphone array selection unit that calculates selection evaluation values of all combinations in all microphone arrays and selects a combination of microphone arrays based on each selection evaluation value.
[Selection] Figure 1

Description

  The present invention relates to a microphone array selection device, a microphone array selection program, and a sound collection device. For example, the present invention is applied to an area sound collection technique that uses a plurality of microphone arrays to emphasize sounds in a specific area and suppress other sounds. It is possible.

  There is a technique using a plurality of microphone arrays as a technique for emphasizing sound in a specific area (speech and sound; sound and sound may be collectively referred to as sound) and suppressing sound in other areas (patent) Reference 1, Japanese Patent Application No. 2012-046989).

  In the method described above, noise sources exist around the target area by directing the directivity of each microphone array from different directions to the area where the target sound source exists (hereinafter also referred to as the target area). Even so, it is possible to collect only the sound of the target area.

  In order to form the directivity of the microphone array, a technique called a beam former (hereinafter also referred to as BF) is used. BF is a technique for forming directivity by using a time difference between signals reaching each microphone of a microphone array (see Non-Patent Document 1).

The basic technique in the beamformer is the delay sum method. FIG. 6 is a block diagram showing the configuration of the delay sum method. In the delay sum method, a plurality (M in FIG. 6) of microphones 21-1, 21-2,..., 21-M are arranged on a straight line at equal intervals (distance d), and each microphone 21- 1, 21-2,..., 21 -M, and the captured signals x ₁ (t) and x ₂ (t) by the corresponding microphones 21-1, 21-2,. _, ..., _x M (t) the delay time with respect to _{(delay) D 1, D 2, ...} , delay device 22-1 and 22-2 which imparts _{D M,} ..., and 22-M, all the delay The sum of output signals x ₁ (t−D ₁ ), x ₂ (t−D ₂ ),..., X _M (t−D _M ) from the devices 22-1, 22-2,. The summing device 23 is functioning.

The angle from the front of the microphone 21-i (i is 1 to M) to the target direction is θ _L , and the sound speed is c. The sound from the sound source in the target direction reaches the adjacent microphones (for example, the microphones 21-1 and 21-2) with a timing shifted by the propagation delay time τ _L shown in the equation (2). Therefore, when each delay amount D _i is selected as shown in equation (1), output signals x ₁ (t−D ₁ ), x ₂ from all delay devices 22-1, 22-2,. _{(t-D 2), ...} , x M (t-D M) is a one phase are aligned with respect to the acoustic component in the intended direction theta _L. As shown in the equation (3), by calculating the sum of the acoustic components from the target direction θ _{L in} which the phases are aligned as described above, the output signal y (t) from the summer 23 becomes the acoustic in the target direction. Will be emphasized. Note that sounds in other directions are not emphasized because the phases are not aligned even through the delay group 22, so that sounds in directions other than the target direction can be suppressed. Note that by applying a delay device 21-i that can change the delay amount _Di , it is possible to easily cope with a change in the target direction. The above processing can be performed not only in the time domain but also in the frequency domain.

  However, there is a problem that noise existing in the same direction as the target area cannot be suppressed in the BF of the microphone array alone. To solve this problem, the inventor of the present application uses a plurality of microphone arrays, directs the directivity formed by the BF in all microphone arrays toward the target area, and compares the BF outputs in the frequency domain. A method for estimating and enhancing the sound of an area has been proposed (see Japanese Patent Application No. 2012-046989).

  FIG. 7 shows a state in which two microphone arrays 1 and 2 are used and directional beams are directed from different locations toward the target area. In the state of FIG. 7, the directivity of the BF of each microphone array 1, 2 includes a non-target area sound in the target area sound direction.

  However, due to the sparseness of speech, there is a low probability that the target area sound and the noise component overlap at the same time in the frequency domain. Therefore, the amplitude spectrum of the target area sound and the noise component in the BF output is increased or decreased independently. FIG. 8 is an image diagram showing the change before and after the BF in each of the microphone arrays 1 and 2 in the frequency domain.

  In FIG. 8, when the amplitude spectra after BF of the microphone arrays 1 and 2 are compared, both the components of the target area sound (blacked portions) are amplified. On the other hand, the noise A (shaded portion) is amplified by the BF of the microphone array 1 located in the same direction as the target area direction, but attenuated by the microphone array 2 located in another direction. Conversely, noise B (outlined portion) is attenuated by the microphone array 1 but amplified by the microphone array 2. That is, the amplitude spectrum of the target area sound component is amplified after BF in all the microphone arrays 1 and 2, but the noise component is increased or decreased for each microphone array. From this difference in change, the frequency at which the power is amplified after BF in all the microphone arrays 1 and 2 is estimated as the component of the target area sound. The target area sound can be emphasized by attenuating frequency components other than the target area sound with respect to the BF outputs of the microphone arrays 1 and 2.

JP 2007-235358 A

Asano Tadashi, “Sound Array Signal Processing-Localization / Tracking and Separation of Sound Sources”, published by Corona Inc., 2011. Takashi Yagami, Makoto Morito, Satoshi Yamada, Tetsuji Ogawa, “Special Issue: Approach to Practical Use of Speech Recognition Technology, Sound Source Separation Technology Using Square Microphone Array”, Information Processing Vol. 51, no. November, November, 2010, P.1410 to P.1416.

  When the above-described conventional area sound collection technique is used, an arbitrary area can be picked up by changing the direction of directivity of the installed microphone array. However, the directivity formed by BF has the property of becoming weaker as the distance increases. Therefore, in order to emphasize the target area sound with the same performance regardless of which area is selected, it is necessary to arrange a large number of microphone arrays in the space including the target area.

  In the conventional area sound collection technique described above, at least two microphone arrays are used for area sound collection processing. When a large number of microphone arrays are arranged, the area sound collection data finally output uses a combination of microphone arrays that can emphasize the target area sound most. In order to specify which combination of microphone arrays is optimal, it is necessary to perform area sound collection processing for all combinations and compare the results. However, if the number of arranged microphone arrays is large, the amount of calculation becomes enormous, which may affect real-time processing.

  Furthermore, combinations of microphone arrays may include combinations with extremely low area sound collection performance. For example, even if area sound collection processing is performed using a microphone array arranged at a position facing the target area, noise in the target area direction cannot be suppressed. Further, even if a microphone array arranged at a location away from the target area is used, a sufficient target area sound enhancement effect cannot be obtained. Accordingly, the amount of calculation can be reduced by excluding combinations that are predicted to have low target area sound enhancement performance in advance and narrowing down candidates and performing area sound collection processing. However, for that purpose, a criterion for selecting a combination of microphone arrays is required.

  For this reason, a plurality of microphone arrays are arranged in the space, a combination of microphone arrays with high area sound collection performance can be selected, and a microphone array selection device and microphone array selection that can reduce the amount of calculation. There is a need for programs and sound collection devices.

  In order to solve this problem, according to the first aspect of the present invention, (1) the number of microphone arrays used for area sound collection processing is divided by 180 °, and the value is determined as the overlapping range of directivity between the microphone arrays. Each microphone for the target area based on the minimum overlap angle calculation unit that calculates as the smallest minimum overlap angle, (2) the position information of each microphone array used for area sound collection processing, and the position information of the target area A directivity intersection angle calculation unit for calculating a directivity intersection angle between the arrays; (3) a minimum overlap angle calculated by the minimum overlap angle calculation unit; and a directivity intersection angle calculated by the directivity intersection angle calculation unit. Difference, and for each combination in all microphone arrays used for area sound collection processing, a selection evaluation value using the difference value is calculated and calculated. Based on the selected evaluation value, a microphone array select device, characterized in that it comprises a microphone array selecting unit that selects the combination of the microphone array.

  According to the second aspect of the present invention, the computer is divided into (1) the number of microphone arrays used for area sound collection processing and 180 ° is divided, and the value is the minimum overlap that minimizes the overlapping range of directivity between the microphone arrays. A minimum overlap angle calculation unit for calculating an angle; (2) directivity intersection between each microphone array with respect to the target area based on the position information of each microphone array used for area sound collection processing and the position information of the target area; A directivity intersection angle calculation unit for calculating an angle; (3) calculating a difference between each of the minimum overlap angle calculated by the minimum overlap angle calculation unit and the directivity intersection angle calculated by the directivity intersection angle calculation unit; For each combination in all microphone arrays used for sound collection processing, a selection evaluation value using the difference value is calculated, and each calculated selection evaluation value is calculated. Based on a microphone array selection program, characterized in that the function as a microphone array selecting unit that selects the combination of the microphone array.

  According to a third aspect of the present invention, there is provided (1) a plurality of microphone arrays, and (2) a microphone array according to the first aspect of the present invention, wherein a combination of microphone arrays used for sound collection processing of a target area is selected from the plurality of microphone arrays. A microphone array selection unit corresponding to the selection device; and (3) a sound collection processing unit that performs predetermined sound collection processing based on signals from the combination of microphone arrays selected by the microphone array selection unit. The sound collecting device.

  According to the present invention, even if a large number of microphone arrays are arranged in a space, a combination candidate of microphone arrays having high area sound collection performance can be selected, and the amount of calculation can be reduced.

It is a block diagram which shows the structure of the sound collection device which concerns on 1st Embodiment. It is a flowchart which shows the process of the microphone array selection apparatus which concerns on 1st Embodiment. It is the figure which showed an example of the arrangement | positioning from which the directional crossing angle between each microphone array becomes the same. It is a block diagram which shows the structure of the sound collection device which concerns on 2nd Embodiment. It is a flowchart which shows the process of the microphone array selection apparatus which concerns on 2nd Embodiment. It is a block diagram which shows the structure concerning a delay sum method. It is explanatory drawing which shows the state which used two microphone arrays and directed the directional beam from the different places to the direction of the target area. The figure which showed the change of the amplitude spectrum before and after forming the directivity in each microphone array.

(A) Technical idea common to the first embodiment and the second embodiment As described above, the microphone array selection device, the microphone array selection program, and the sound collection device according to the present invention are arranged with a plurality of microphone arrays. Even if the microphone array is used, the microphone array that can emphasize the target area sound using the degree of directivity overlap when each microphone array directs the directivity toward the target area and the distance between each microphone array and the target area. Select a combination candidate.

  The area sound collection technology using a microphone array cannot suppress the noise when the directivity of each microphone array includes the same noise. That is, in order to emphasize only the target area sound, the directivities of the microphone arrays need to include only the target area sound component in common and the noise components must be different. From this feature, if a combination of microphone arrays with a small directivity overlap range is selected, there is a high possibility that good performance can be obtained. However, it is difficult to actually obtain the degree of directivity overlap from the area and volume of directivity.

  Therefore, in the present invention, the angle connecting the center of the target area and each microphone array is defined as the degree of directivity overlap. The angle with the smallest overlapping range of directivity of each microphone array (hereinafter also referred to as the minimum overlapping angle) is a number obtained by dividing 180 ° by the number of microphone arrays used for area sound collection processing. That is, as shown in FIG. 7, when two microphone arrays existing on the same plane are used, an angle at which the directivity of the two microphone arrays with respect to the target area intersects (hereinafter also referred to as a directivity intersection angle). When the angle is 90 °, the directivity overlap is minimized. That is, a higher priority is given to a combination of microphone arrays in which the directional intersection angle forms an angle close to the minimum overlap angle.

  Furthermore, as the distance between the target area and the microphone array is increased, the acoustic signal is attenuated and the directivity by BF is also weakened. In consideration of this characteristic, the present invention performs weighting according to the distance between the target area and each microphone array. Thereby, when the directivity crossing angle is the same, the combination of microphone arrays used for the target area is preferentially selected. If one microphone array is far from the target area even if the directional crossing angle is close to the minimum overlap angle, the microphone array closest to the target area is the second closest directional crossing angle to the minimum overlap angle. Selected. For example, when three microphone arrays are installed and two microphone arrays are selected from among them, the minimum overlap angle is 90 °. For example, even if the directional intersection angle of the microphone arrays 1 and 2 is 90 ° and the directional intersection angle of the microphone arrays 2 and 3 is 60 °, the microphone array 1 is far from the target area and the microphone array 3 is If the target area is close, the microphone arrays 2 and 3 are selected.

(B) First Embodiment Hereinafter, a first embodiment of a microphone array selection device, a microphone array selection program, and a sound collection device of the present invention will be described in detail with reference to the drawings.

(B-1) Configuration of First Embodiment FIG. 1 is a block diagram illustrating a configuration of a sound collection device 100A according to the first embodiment. In FIG. 1, the sound collection device 100 </ b> A is a circuit device such as a CPU, ROM, RAM, EEPROM, and input / output interface, and can be realized by the CPU executing a program stored in the ROM. . Even in this case, the sound collecting device 100A can functionally have a configuration as shown in FIG.

  1, a sound collection device 100A according to the first embodiment includes a plurality (n in FIG. 1) of microphone array 1, microphone array 2,..., Microphone array n, microphone array selection device 10A, and sound collection processing unit 17. It is comprised.

  The microphone array selection apparatus 10A according to the first embodiment includes a data input unit 11, a minimum overlap angle calculation unit 12, a directivity intersection angle calculation unit 13, a microphone array selection unit 14, and a spatial coordinate data storage unit 15. .

  The microphone arrays 1, 2,..., N are composed of a plurality of microphones, and an acoustic signal picked up (captured) by each microphone is given to the microphone array selection device 10A.

  The microphone arrays 1, 2,..., N are arranged at positions where the target area can be pointed in the space where the target area exists. The microphone arrays 1, 2,..., N are arranged at positions different from those of the other microphone arrays. The microphone arrays 1, 2,..., N may be arranged on the same plane (for example, on the same horizontal plane) in the space, or may be arranged at three-dimensionally different positions.

  The plurality of microphones constituting the microphone arrays 1, 2,..., N may be arranged in a straight line (horizontal row, vertical row), or arranged in a cross shape or a lattice shape. It may be a thing, and it may be arranged at the vertex of a polygon (for example, a rectangle, a hexagon, etc.).

  The microphone array selection apparatus 10A selects a combination of microphone arrays having a small directivity overlap range from a plurality of microphone arrays 1, 2,..., N.

  The data input unit 11 converts acoustic signals from the microphone arrays 1, 2,..., N from analog signals to digital signals. The data input unit 11 supplies acoustic signals (digital signals) from the respective microphone arrays 1, 2,..., N to the microphone array selection unit 14.

  The spatial coordinate data storage unit 15 stores the coordinates of the center of the target area and the arrangement coordinates of all the microphone arrays 1, 2,..., N arranged in the space. The spatial coordinates stored in the spatial coordinate data storage unit 15 can be spatial coordinates with respect to an arbitrary spatial coordinate axis. In addition, the arrangement coordinates of the microphone arrays 1, 2,..., N may be representative spatial coordinates of the microphone arrays 1, 2,. For example, the representative spatial coordinates of each microphone array 1, 2,..., N may be the spatial coordinates of the center of a line segment connecting a plurality of microphones included in the microphone arrays 1, 2,. Alternatively, the spatial coordinates of the center of each microphone array itself may be used.

  The minimum overlap angle calculator 12 calculates an angle (minimum overlap angle) having the smallest directivity overlap range from the number of microphone arrays. That is, the minimum overlap angle calculation unit 12 calculates the minimum overlap angle by dividing 180 ° by the number of the plurality of microphone arrays used for the area sound collection processing.

  Note that the minimum overlap angle calculation unit 12 may consider the same plane in which a plurality of microphone arrays exist even when a plurality of microphone arrays are arranged in a three-dimensional space. In this case, for example, the minimum overlap angle calculation unit 12 refers to the spatial coordinate data storage unit 15 to obtain the number of microphone arrays existing on the same plane, and minimizes by dividing 180 ° by the number of microphone arrays. Calculate the overlap angle.

  In addition, the minimum overlap angle calculation unit 12 gives the calculated minimum overlap angle to the microphone selection unit 14.

  The directivity intersection angle calculation unit 13 determines the center of the target area based on the center coordinates of the target area stored in the spatial coordinate data storage unit 15 and the arrangement coordinates of all the microphone arrays 1, 2,. The directional intersection angle is calculated. That is, the directivity intersection angle calculation unit 13 obtains the depression angle between the center of the target area and two different microphone arrays, and sets this as the directivity intersection angle. Further, the directivity intersection angle calculation unit 13 gives all the directivity intersection angles with respect to the center of the target area to the microphone array selection unit 14.

  The microphone array selection unit 14 calculates a difference between the minimum overlap angle calculated by the minimum overlap angle calculation unit 12 and the directivity cross angle calculated by the directivity intersection angle calculation unit 13 and adds the differences. A combination of microphone arrays having a small value is selected. The microphone array selection unit 14 gives the acoustic signal of the selected combination of microphone arrays to the sound collection processing unit 17.

  The sound collection processing unit 17 acquires the acoustic signals of the combination of microphone arrays selected by the microphone array selection unit 14 and performs sound collection processing based on the acquired acoustic signals. The sound collection processing unit 17 emphasizes the target area sound component and suppresses the sound component other than the target area based on the acquired acoustic signal of the microphone array. In addition, the method of the sound collection process by the sound collection process part 17 can apply various methods widely, and is not specifically limited.

(B-2) Operation of the First Embodiment Next, the operation of the microphone array selection device 10A according to the first embodiment will be described.

  FIG. 2 is a flowchart showing a processing operation in the microphone array selection apparatus 10A according to the first embodiment.

  Sounds from various sound sources in the space where the target area exists are collected (captured) by the microphones constituting the microphone arrays 1, 2,... It is converted into a digital signal (S101).

The minimum overlap angle calculation unit 12 calculates an angle (minimum overlap angle) having the smallest directivity overlap range based on the number of microphone arrays used for the area sound collection process (S102). The minimum overlap angle calculator 12 calculates the minimum overlap angle θ _moa according to the equation (4).

Here, N is the number of microphone arrays used in the area sound collection process. The unit of the minimum overlap angle θ _moa may be radians by the arc method or angles by the power method. Formula (4) exemplifies a case where the minimum overlap angle θ _moa is calculated in radians by the arc degree method.

  The directivity intersection angle calculation unit 13 uses the coordinate data stored in the spatial coordinate data storage unit 15 to combine the two microphone arrays of all the microphone arrays 1, 2,. A crossing angle that minimizes the overlapping range of directivity with respect to the center is calculated (S102).

First, the directivity intersection angle calculation unit 13 acquires the coordinates of the arranged microphone array and the coordinates of the center of the target area from the spatial coordinate data storage unit 15, and the center of the target area and two microphone arrays (for example, A depression angle θ _ij with respect to the microphone arrays i and j is calculated (S103). At this time, the depression angle θij is set to π [rad] or less.

Next, the directivity intersection angle calculation unit 13 calculates θ (bar) _ij that is a value obtained by subtracting the depression angle θ _ij from 180 ° (π [rad]) according to the equation (5) (S104). The directivity intersection angle calculation unit 13 compares θ _ij and θ (bar) _ij, and sets the smaller one as the directivity intersection angle ω _ij (S105).

Here, the meaning of the equation (6) will be described. Since the directivity is formed linearly, the two microphone arrays arranged toward the front side across the target area have the same directivity. For example, as shown in FIG. 3, the microphone array 2 and the microphone array 3 are arranged to face each other across the target area, and the microphone array 2 and the microphone array 3 form the same directivity. In this case, the directivity intersection angles ω ₁₂ and ω ₁₃ with the microphone array 1 must be the same value. Therefore, the directivity intersection angle calculation unit 13 calculates the directivity intersection angle having the same overlap degree according to the equation (6).

The microphone array selection unit 14 includes a minimum overlap angle θ _moa calculated by the minimum overlap angle calculation unit 12 and two microphone arrays 1, 2,..., N calculated by the directivity intersection angle calculation unit 13. A difference from the directivity intersection angle ω _ij between the microphone arrays is calculated, and these differences are added together to select a combination of microphone arrays having a small value. Then, the microphone array selection unit 14 outputs an acoustic signal from the microphone array related to the selected combination to the sound collection processing unit 17.

First, the microphone array selection unit 14 calculates a selection evaluation value Λ _m for selecting a combination of microphone arrays according to Expression (7).

Here, when the microphone array selection unit 14 calculates the selection evaluation value Λ _m , the combination of microphone arrays must be continuous. That is, in equation (7), the minimum overlap angle θ as subscripts directional intersection angle omega obtaining a difference between _moa are consecutive, for example, the following directional intersection angle omega ₁₂ is directional intersection angle omega ₂₃ or The directivity intersection angle ω ₂₄ is set.

For example, when there are four microphone arrays 1, 2, 3, 4 on the same plane, the microphone array selection unit 14 selects one of the four microphone arrays 1, 2, 3, 4, for example, Λ ₁ = | Θ _moa −ω ₁₂ | + | θ _moa −ω ₂₃ | + | θ _moa −ω ₃₄ |, Λ ₂ = | θ _moa −ω ₁₃ | + | θ _moa −ω ₃₄ |, Λ ₃ = | θ _moa −ω ₁₄ |, Λ ₄ = | θ _moa −ω ₂₃ | + | θ _moa −ω ₃₄ | + | θ _moa −ω ₄₁ |, Λ ₅ = | θ _moa −ω ₂₄ | + | θ _moa −ω ₄₁ | , Λ ₆ = | θ _moa −ω ₂₁ |,..., Etc., the selection evaluation values Λ _m of all combinations are calculated (S106).

From equation (7), it can be seen that when the directional intersection angle ω _ij is the same as the minimum overlap angle θ _moa , the parameter related to the angle in parentheses (within the absolute value) is the minimum value “0”. As a result, the value of the selection evaluation value Λ _m is also reduced, so that the microphone array selection unit 14 can select a combination of microphone arrays having a small directivity overlap range by using the selection evaluation value Λ _m .

Then, the microphone array selection unit 14 compares the combination selection evaluation values Λ _m of all the microphone arrays, and the priority is higher as the selection evaluation value Λ _m is smaller. The degree is determined (S107). Thereafter, the microphone array selection unit 14 selects a combination of microphone arrays having high priority, and outputs an acoustic signal (data) of the selected microphone array to the sound collection processing unit 17 (S108).

(B-3) Effects of First Embodiment As described above, according to the first embodiment, even when a large number of microphone arrays are installed in a space including a target area, the distance between each microphone array is By using the overlapping degree of directivity as a selection criterion, it is possible to select a combination of microphone arrays having high area sound collection performance.

(C) Second Embodiment Next, a microphone array selection device, a microphone array selection program, and a sound collection device according to a second embodiment of the present invention will be described in detail with reference to the drawings.

  The acoustic signal attenuates as the microphone array moves away from the target area. In addition, the directivity of the microphone array becomes weaker as the distance from the target area to the microphone array increases. Therefore, the second embodiment exemplifies a case where a combination of microphone arrays is selected by weighting in consideration of the influence of distance in addition to the degree of directivity overlap.

(C-1) Configuration of Second Embodiment FIG. 4 is a block diagram illustrating a configuration of a sound collection device according to the second embodiment. In FIG. 4, the sound collection device 100B according to the second embodiment is a circuit device such as a CPU, ROM, RAM, EEPROM, input / output interface, etc., and the CPU executes a program stored in the ROM. Can also be realized. Even in this case, the sound collection device 100B can functionally have a configuration as shown in FIG.

  4, the sound collection device 100B according to the second embodiment includes a plurality (n in FIG. 4) of microphone array 1, microphone array 2,..., Microphone array n, microphone array selection device 10B, and sound collection processing unit 17. It is comprised.

  In addition, the microphone array selection device 10B according to the second exemplary embodiment includes a data input unit 11, a minimum overlap angle calculation unit 12, a directivity intersection angle calculation unit 13, a microphone array selection unit 24, a spatial coordinate data storage unit 15, and a distance. A coefficient calculation unit 26 is provided.

  In the second embodiment, the configuration of the first embodiment is the same as that of the first embodiment in that a microphone array selection unit 24 is provided instead of the microphone array selection unit 14 of the first embodiment, and a distance coefficient calculation unit 26 is newly provided. Different.

  The distance coefficient calculation unit 26 receives each microphone array 1, 2,..., N from the spatial coordinate data storage unit 15 based on the coordinates of the center of the target area and the coordinates of each microphone array 1, 2,. The distance between the target area and the center of the target area is obtained, and a weighting coefficient weighted according to the obtained distance is calculated. In addition, the distance coefficient calculation unit 26 calculates a distance coefficient of each microphone array and gives the calculated distance coefficient to the microphone array selection unit 24.

  The microphone array selection unit 24 calculates a difference between the minimum overlap angle calculated by the minimum overlap angle calculation unit 12 and the directivity intersection angle calculated by the directivity intersection angle calculation unit 13, and a distance coefficient corresponding to each difference. A combination of microphone arrays having a small value obtained by multiplying the respective values by multiplying the distance coefficient from the calculation unit 26 is selected. Further, the microphone array selection unit 24 provides the sound collection processing unit 17 with the acoustic signals of the selected combination of microphone arrays.

(C-2) Operation of Second Embodiment Next, the operation of the microphone array selection device 10B according to the second embodiment will be described.

  FIG. 5 is a flowchart showing a processing operation in the microphone array selection apparatus 10B according to the second embodiment.

  Since each process of S201-S205 shown in FIG. 5 is the same as or corresponds to each process of S101-S105 shown in FIG. 2, detailed description thereof is omitted here.

  The distance coefficient calculation unit 26 calculates a coefficient (value) for weighting the influence of the distance on the directivity based on the distance between the target area and each of the microphone arrays 1, 2,.

First, the distance coefficient calculation unit 26 acquires the coordinates of the arranged microphone arrays 1, 2,..., N and the coordinates of the center of the target area from the spatial coordinate data storage unit 15. Then, the distance coefficient calculation unit 26 determines the target area from the respective microphone arrays 1, 2,..., N based on the coordinates of the center of the target area and the coordinates of the respective microphone arrays 1, 2,. The distance l _i (1 ≦ i ≦ n) to the center is calculated.

Next, the distance coefficient calculation unit 26 calculates the distance l _i between each of the microphone arrays 1, 2,..., N and the center of the target area according to Expression (8) as the distance between all the microphone arrays and the target area. And a value L _i weighted by the distance is calculated (S206).

Here, in Equation (8), M is the total number of microphone arrays. α (l _i ) is a weighting coefficient that takes into account the attenuation of sound due to the microphone array moving away from the target area and the influence of the weak directivity.

Next, the distance coefficient calculation unit 26 calculates an average distance to two microphone arrays (for example, microphone array i and microphone array j) as a distance coefficient L _ij according to Expression (9) (S207). This distance coefficient L _ij has a characteristic that becomes larger as the distance between the target area and the microphone array related to the combination becomes larger.

  The microphone array selection unit 24 calculates the difference between the minimum overlap angle calculated by the minimum overlap angle calculation unit 12 and the directional cross angle of the combination of each microphone array calculated by the directivity cross angle calculation unit 13. Further, the microphone array selection unit 24 calculates a value obtained by multiplying the difference between the minimum overlap angle and each directivity intersection angle by the corresponding distance coefficient. Then, the microphone array selection unit 24 calculates a value obtained by adding the values obtained by multiplying the combinations of the microphone arrays by the distance coefficient, and selects a combination of microphone arrays having a small calculated value. Then, the microphone array selection unit 24 outputs the acoustic signal (data) of the selected combination of microphone arrays to the sound collection processing unit 17.

Here, the microphone array selection unit 24 calculates a selection evaluation value Λ _m for selecting a combination of microphone arrays according to the equation (10).

Here, Expression (10) is an extension of Expression (7). For example, when the directivity intersection angle ω _ij is the same as the minimum overlap angle θmoa, the parameter related to the angle in parentheses is “1”. It has changed. Then, the microphone array selection unit 24 calculates the selection evaluation value Λ _m by multiplying the parameter (value) related to the angle in each parenthesis of Expression (10) by the corresponding distance coefficient L _ij (S208). This makes it possible to select a combination of microphone arrays that have a small directivity overlap range and are arranged at positions close to the target area.

Then, the microphone array selection unit 24 compares the combination selection evaluation values Λ _m of all the microphone arrays. The smaller the selection evaluation value Λ _m is, the higher the priority is. The degree is determined (S209). Thereafter, the microphone array selection unit 24 selects a combination of microphone arrays having high priority, and outputs an acoustic signal (data) of the selected microphone array to the sound collection processing unit 17 (S210).

(C-3) Effects of the Second Embodiment As described above, according to the second embodiment, even when a large number of microphone arrays are installed in a space including a target area, between each microphone array. By using as a selection criterion a value obtained by weighting the degree of overlap of directivity with a distance, a combination of microphone arrays having high area sound collection performance can be selected.

(D) Other Embodiments In the first and second embodiments described above, various modified embodiments of the present invention have been described. However, the present invention can also be applied to the following other embodiments. it can.

  In the first and second embodiments described above, the arithmetic expression for obtaining the selection evaluation value is not limited to Expression (7) and Expression (10). For example, in Expression (7), the calculation is performed by adding the terms related to the combination of the microphone arrays. However, an arithmetic expression for obtaining the selection evaluation value may be calculated by multiplying the terms. As a result, when the directivity intersection angle is the same as the minimum overlap angle, the parameter related to the angle in parentheses (within the absolute value) includes the minimum value “0”, and the selection evaluation value is also “0”. Therefore, a combination of microphone arrays can be selected.

  In the first and second embodiments described above, when selecting a combination of macrophone arrays based on a selection evaluation value, if a plurality of selection evaluation values are the same value, based on a predetermined priority criterion, Any combination of microphone arrays may be selected. For example, from the viewpoint of the enhancement accuracy of the target area sound, it may be possible to select one having a larger number of microphone arrays that use 180 ° for area sound collection processing, or from the viewpoint of reducing the amount of calculation, 180 °. You may make it select the one where the number of microphone arrays used for area sound collection processing is smaller.

  The selection result of the combination of the microphone arrays selected by the first and second embodiments described above is stored (ie, learned), and the sound collection processing is performed with the combination stored in the past for the same target area sound. May be.

100A and 100B ... sound collecting device,
10A and 10B ... microphone array selection device,
DESCRIPTION OF SYMBOLS 11 ... Data input part, 12 ... Minimum overlap angle calculation part, 13 ... Directionality cross angle calculation part, 14 and 24 ... Microphone array selection part, 15 ... Spatial coordinate data storage part, 26 ... Distance coefficient calculation part,
1, 2,..., N ... microphone array, 17 ... sound collection processing unit.

Claims (4)

Divide 180 ° by the number of microphone arrays used for area sound collection processing, and calculate the value as the minimum overlap angle that minimizes the overlapping range of directivity between the microphone arrays;
A directional intersection angle calculation unit that calculates a directional intersection angle between each microphone array with respect to the target area, based on the position information of each microphone array used for the area sound collection processing and the position information of the target area;
The difference between the minimum overlap angle calculated by the minimum overlap angle calculation unit and the directivity cross angle calculated by the directivity intersection angle calculation unit is calculated, and all the microphone arrays used for the area sound collection processing are calculated. And a microphone array selection unit that calculates a selection evaluation value using the difference value for all of the combinations, and selects a combination of microphone arrays based on the calculated selection evaluation values. A microphone array selection device. Based on the position information of the target area and the position information of each microphone array used for area sound collection processing, the distance between the target area and each microphone array is calculated, given a predetermined weight, and between the microphone arrays. A distance coefficient calculation unit for calculating the distance coefficient of
All microphones used by the microphone array selection unit for the area sound collection processing using values obtained by multiplying the difference between the minimum overlap angle and the directivity intersection angles by the corresponding distance coefficients. 2. The microphone array according to claim 1, wherein selection evaluation values for all combinations in the array are calculated, and combinations of microphone arrays are selected based on the calculated selection evaluation values. Selection device. Computer
Divide 180 ° by the number of microphone arrays used for area sound collection processing, and calculate the value as the minimum overlapping angle that minimizes the overlapping range of directivity between the microphone arrays,
A directional intersection angle calculation unit for calculating a directional intersection angle between each microphone array with respect to the target area, based on the position information of each microphone array used for the area sound collection processing and the position information of the target area;
The difference between the minimum overlap angle calculated by the minimum overlap angle calculation unit and the directivity cross angle calculated by the directivity intersection angle calculation unit is calculated, and all the microphone arrays used for the area sound collection processing are calculated. It functions as a microphone array selection unit that calculates a selection evaluation value using the difference value for all the combinations, and selects a combination of microphone arrays based on the calculated selection evaluation values. A microphone array selection program. Multiple microphone arrays,
A microphone array selection unit corresponding to the microphone array selection device according to claim 1 or 2, wherein a combination of microphone arrays used for sound collection processing of a target area is selected from the plurality of microphone arrays.
A sound collection device comprising: a sound collection processing unit that performs predetermined sound collection processing based on signals from a combination of microphone arrays selected by the microphone array selection unit.

Images