Microphone Array

Abstract

Even with a small number of microphone elements, orthogonality of high-order spherical harmonics is guaranteed. The microphone array includes a sphere and a plurality of microphone elements arranged on the surface of the sphere according to coordinate positions of a spherical t-design.

Description

[Technical Field]

[0001] The present disclosure relates to microphone arrays. [Background technology]

[0002] A microphone array in which multiple microphone elements are arranged on the surface of a sphere is generally used for measuring or analyzing a three-dimensional sound field. For example, Patent Document 1 describes a microphone array having multiple microphone elements arranged along a spiral trajectory on the spherical surface. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 2018-157309 Summary of the Invention [Problem to be solved by the invention]

[0004] The sound pressure on a sphere is expressed as a linear combination of spherical harmonics. In measuring or analyzing a three-dimensional sound field, it is necessary to obtain the coefficients of the spherical harmonics from the sound pressure signals from the microphone elements by utilizing the orthogonality of the spherical harmonics.

[0005] However, the microphone array described in Patent Document 1 does not take into consideration the orthogonality of higher-order spherical harmonic functions, which makes it difficult to measure or analyze a sound field with high precision.

[0006] In consideration of the above circumstances, one aspect of the present disclosure aims to ensure the orthogonality of high-order spherical harmonics even when the number of microphone elements is small. [Means for solving the problem]

[0007] In order to solve the above problems, a microphone array according to a preferred embodiment of the present disclosure comprises a sphere and a plurality of microphone elements arranged on the surface of the sphere according to the coordinate positions of a spherical t-design. [Brief explanation of the drawings]

[0008] [Figure 1] 1 is an overall view of a microphone array according to an embodiment. [Figure 2] 1A and 1B are diagrams illustrating an example of the arrangement of a plurality of microphone elements according to an embodiment. [Figure 3] FIG. 3 is a diagram showing coordinate positions in the arrangement example shown in FIG. 2. [Figure 4] 10 is a graph showing the evaluation results of orthogonality between modes of spherical harmonics in a Fibonacci spiral arrangement. [Figure 5] 10 is a graph showing the results of an evaluation of orthogonality between modes of spherical harmonics in a spherical t-design arrangement. [Figure 6] FIG. 10 is a diagram showing the results of sound collection by beamforming at horizontal angles in a Fibonacci spiral arrangement. [Figure 7] This figure shows the results of sound collection using beamforming at horizontal angles in a spherical t-design arrangement. [Figure 8] FIG. 10 is a diagram showing the results of sound collection by beamforming at an elevation angle in a Fibonacci spiral arrangement. [Figure 9] This figure shows the results of sound collection using beamforming at an elevation angle in a spherical t-design configuration. [Figure 10] FIG. 10 shows the signal-to-noise ratio of microphone arrays in a Fibonacci spiral configuration and a spherical t-design configuration. DETAILED DESCRIPTION OF THE INVENTION

[0009] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings. Note that the dimensions and scale of each part in the drawings are appropriately different from those in reality. Furthermore, the embodiments described below are preferred specific examples of the present disclosure. Therefore, various technically preferable limitations are applied to the present embodiments. However, the scope of the present disclosure is not limited to these embodiments unless otherwise specified in the following description to the effect that the present disclosure is limited.

[0010] 1. Embodiment 1-1. Microphone array overview FIG. 1 is an overall view of a microphone array 10 according to an embodiment. FIG. 1 illustrates a system 100 using the microphone array 10. For ease of explanation, FIG. 1 also illustrates an X-axis, a Y-axis, and a Z-axis, which are orthogonal to each other. The Z-axis is an axis parallel to the vertical axis. The direction vertically upward along the Z-axis is the Z1 direction, and the direction opposite to the Z1 direction is the Z2 direction. Hereinafter, the Z1 direction may be referred to as "up" and the Z2 direction as "down." The Y1 direction may also be referred to as "front."

[0011] In system 100, a sound pressure signal indicating the sound pressure of sound picked up by microphone array 10 is processed by processing device 20. Here, microphone array 10 and processing device 20 are connected via cable 30, and the sound pressure signal is input from microphone array 10 to processing device 20 via cable 30. Processing device 20 performs processing to convert the sound pressure signal into a multi-channel audio signal of a predetermined standard. Note that processing device 20 may have a function to analyze a three-dimensional sound field based on the sound pressure signal.

[0012] The microphone array 10 comprises a sphere 11 and a plurality of microphone elements 12 .

[0013] Sphere 11 is a hollow support that supports multiple microphone elements 12, and is made of, for example, resin or a composite material. The diameter of sphere 11 is not particularly limited, but is, for example, in the range of 5 cm to 50 cm, and preferably in the range of 5 cm to 20 cm.

[0014] Sphere 11 is supported by support 13. Support 13 is a structure that supports sphere 11 and is made of, for example, metal. In the example shown in FIG. 1, support 13 is a tripod-shaped structure that has a shaft that extends in the Z2 direction from the end, i.e., the lower end, of sphere 11, and three legs that are connected to the lower end of the shaft. Note that the shape of support 13 is not limited to the example shown in FIG. 1 and can be any shape.

[0015] A plurality of microphone elements 12 are arranged on the surface of the sphere 11. This has the advantage of providing uniform resolution in all directions and enabling precise sound field analysis compared to an arrangement on the surface of a polyhedron. In particular, when performing spherical harmonic analysis as described below, the details of the sound field can be analyzed with high precision.

[0016] The plurality of microphone elements 12 are elements that output sound pressure signals indicating sound pressure by collecting sound, and are, for example, MEMS (microelectromechanical systems) microphone elements.

[0017] A MEMS microphone element is an element that includes a MEMS chip, which serves as an acoustic transducer, and an IC (Integrated Circuit) chip, which serves as a signal processing circuit, such as an ASIC (Application Specific Integrated Circuit), mounted on a substrate and packaged as an assembly. The MEMS chip, for example, has a diaphragm and a backplate that function like a parallel-plate capacitor. It is obtained by forming a microstructure on a silicon wafer using semiconductor manufacturing techniques such as photolithography and etching. The IC chip includes, for example, a boost circuit and an amplifier circuit. The boost circuit supplies a bias voltage to the backplate. A voltage signal corresponding to the capacitance between the diaphragm and the backplate is output from the MEMS chip. The amplifier circuit amplifies the electrical signal and performs pulse density modulation. A sound pressure signal indicating the sound pressure is output as a digital signal. Although not shown, the digital signal output from all microphone elements 12 is converted into serial data and transmitted to cable 30 via wiring running from sphere 11 to support 13.

[0018] In this way, because each of the multiple microphone elements 12 is a MEMS microphone element, a sound pressure signal indicating the sound pressure measured by each microphone element 12 can be transmitted as a digital signal to the outside of the sphere 11. Therefore, even if the processing device 20 for processing the sound pressure signal is placed at a position far from the sphere 11, it is possible to reduce the mixing of noise into the sound pressure signal. Furthermore, by placing the processing device 20 at a position far from the sphere 11, it is also possible to reduce noise caused by sound reflected by the housing of the processing device 20, etc.

[0019] The microphone elements 12 are arranged on the surface of the sphere 11 according to the coordinate positions of a spherical t-design. This ensures the orthogonality of high-order spherical harmonics even when the number of microphone elements 12 is small. This allows for highly accurate measurement or analysis of a three-dimensional sound field. This point will be described in detail below.

[0020] The coefficient α of the μ-th order spherical harmonic function from the measured sound pressure ν μ Considering the case of obtaining (k), the sound pressure p≦N(r,k) at position r expressed by coordinate values ​​(r,k) is expressed by the following formula (1). Here, ≦N assumes that the sound pressure can be expressed by a series of Nth order or less, and this is stated explicitly.

number

[0021] When equation (1) is expanded into spherical harmonics, the coefficient α ν μ (k) is expressed by the following formula (2).

number

[0022] When observing with a finite (Q) number of microphones, it is not possible to directly apply equation (2), and the integral over the spherical surface must be discretized and expressed as in equation (3) below.

number

[0023] The condition for connecting equations (2) and (3) with an equal sign is the following equation (4).

number

[0024] Equation (4) can be regarded as the orthogonality of a finite number of spherical harmonic functions, which can be seen to be equivalent to the following equation (5) if only the angular function (spherical harmonic function) part of the sound pressure expressed by equation (1) is extracted. Therefore, the optimal arrangement of the multiple microphone elements 12 satisfies the following equation (5).

number

[0025] For any t-th degree polynomial f, if the following formula (6) is satisfied, then the set X ⊂ S n-1 is the sphere S n-1 This is called t-design in

number

[0026] In the above equation (4), p≦N(r,k) is a polynomial of degree N at most, and Y ν μ (θ,φ) * The 2Nth-order polynomial expressed as the product of these corresponds to f in equation (6). Therefore, it can be seen that t = 2N is necessary to satisfy the relationship in equation (4). Therefore, the arrangement using spherical 2N-design guarantees the relationship in equation (4) up to the Nth order.

[0027] Ensuring the orthogonality of high-order spherical harmonics in this way has extremely important significance in sound field analysis, such as ensuring mode independence, uniqueness of the expansion of spherical harmonics, improving calculation efficiency, removing noise, clarifying physical interpretation, etc. This allows for the acquisition of highly reliable, efficient, and accurate results in sound field analysis.

[0028] Furthermore, since the spherical t-design for arranging the plurality of microphone elements 12 is a spherical 2N-design, the orthogonality of spherical harmonics up to the Nth order can be guaranteed.

[0029] Furthermore, since t=10, i.e., the spherical t-design for arranging the plurality of microphone elements 12 is a spherical 10-design, orthogonality of spherical harmonic functions up to the fifth order can be guaranteed. Also, since the number of microphone elements 12 is 60, 62, or 64, sound pressure signals from the plurality of microphone elements 12 can be processed using a processing device 20 that can process sound pressure signals of up to 64 channels.

[0030] Fig. 2 is a diagram showing an example of the arrangement of multiple microphone elements 12 in an embodiment. Fig. 3 is a diagram showing the coordinate positions of the example arrangement shown in Fig. 2. In the example shown in Fig. 2, the number of microphone elements 12 arranged on the surface of a sphere 11 is 60. In other words, the arrangement of the multiple microphone elements 12 is an arrangement based on a spherical surface 10 design. Note that Fig. 3 shows the coordinate positions (X, Y, Z) of 60 points 1 to 60 when the diameter of the sphere 11 is 0.05 m.

[0031] As shown in FIG. 2, microphone element 12-0, which is one of the microphone elements 12 arranged on the surface of sphere 11, is located on an intersection line L between an imaginary plane that passes through the center of sphere 11 and is parallel to the horizontal plane and the surface of sphere 11. When using microphone array 10, microphone element 12-0 is installed so that microphone element 12-0 faces forward. This allows microphone element 12-0 to be used as a reference. By using microphone element 12-0 as a reference in this way, setting work such as calibration can be easily performed. In the example shown in FIG. 3, the coordinate position of microphone element 12-0 is the coordinate position of point 59.

[0032] 2, the number of microphone elements 12 arranged on the surface of the sphere 11 is 60. In an embodiment in which the number of microphone elements 12 arranged on the surface of the sphere 11 is more than 20, the number of microphone elements 12 can be increased compared to an embodiment in which microphone elements 12 are arranged at each vertex of a regular polyhedron. As a result, the measurement accuracy or analysis accuracy of a three-dimensional sound field can be improved.

[0033] Furthermore, if the number of microphone elements 12 is 60, sound pressure signals from the microphone elements 12 can be processed using 60 channels of a processing device capable of processing 64 channels of sound pressure signals, and the remaining channels can be effectively utilized for functional expansion, etc.

[0034] None of the multiple microphone elements 12 is arranged in a predetermined area R that includes the lower end of the sphere 11 in the vertical direction. This allows the predetermined area R of the sphere 11 to be used as a space for passing wires from each microphone element 12. The sphere 11 can also be supported by the support body 13 in the predetermined area R. Here, the area R is an area that is larger than the cross section of the shaft of the support body 13.

[0035] Here, if the number of microphone elements 12 is 60, it is possible to arrange one microphone element on the intersection line L of the sphere 11 without arranging the microphone elements 12 in the region R of the sphere 11. The intersection line L can also be said to be a great circle along the horizontal plane of the sphere 11.

[0036] FIG. 4 is a graph showing the evaluation results of the orthogonality between modes of spherical harmonics in a Fibonacci spiral arrangement in dB. FIG. 5 is a graph showing the evaluation results of the orthogonality between modes of spherical harmonics in a spherical t-design arrangement. In FIGS. 4 and 5, the vertical and horizontal axes represent different modes. In FIGS. 4 and 5, the dot products of spherical harmonics of different modes are shown as amplitudes, with the dot products closer to zero being displayed in darker colors and closer to 1 being displayed in brighter colors. When only the diagonal components of the dot products are 1 (0 dB) and the other components are 0 (-∞ dB), this indicates that the different modes are orthogonal to each other.

[0037] Orthogonality (Ξ) of spherical harmonics shown in Figures 4 and 5 n,m ν,μ is an evaluation result obtained by a simulation using the following formula (7). It is expressed as:

number

[0038] As shown in FIG. 4, in the Fibonacci spiral arrangement, there are parts where the inner products of the spherical harmonics between different modes are not zero, and therefore orthogonality between different modes is not guaranteed.

[0039] In contrast, as shown in FIG. 5, in the spherical t-design configuration, the dot products of the spherical harmonics between different modes are zero, thus ensuring the orthogonality of different modes.

[0040] In this way, in a spherical t-design arrangement, each mode of spherical harmonic function is independent due to orthogonality, so the spherical harmonic function of each mode can be analyzed without being affected by the spherical harmonic functions of other modes, making it possible to analyze complex three-dimensional sound fields.

[0041] Figure 6 shows the results of sound collection by beamforming for horizontal angle φ at elevation angle θ = 0 in a Fibonacci spiral arrangement. Figure 7 shows beamforming for horizontal angle φ at elevation angle θ = 0 in a spherical t-design arrangement. In Figures 6 and 7, the vertical axis represents horizontal angle φ, and the horizontal axis represents frequency. Figures 6 and 7 show the frequency characteristics of sound collected by beamforming using a microphone array, with brighter colors indicating higher amplitudes.

[0042] The beamforming ω(η, k) shown in Figures 6 and 7 is an evaluation result obtained using the following equation (8) based on the measurement results of the impulse response, where η is the angle of the beam direction and j is the imaginary unit. It is expressed as:

[0043]

number

[0044] In the spherical t-design arrangement shown in Figure 7, energy is concentrated in the direction where the horizontal angle φ is zero, compared to the Fibonacci spiral arrangement shown in Figure 6. This sharpening of the beam directionality makes it possible to suppress unwanted reflections or scattering from the left and right, and from the front and back, resulting in more precise measurements.

[0045] Figure 8 shows the results of sound collection by beamforming for elevation angle θ at horizontal angle φ = 0 in a Fibonacci spiral arrangement. Figure 9 shows the results of sound collection by beamforming for elevation angle θ at horizontal angle φ = 0 in a spherical t-design arrangement. In Figures 8 and 9, the vertical axis represents elevation angle θ, and the horizontal axis represents frequency. Figures 8 and 9 show the frequency characteristics of sound collected by beamforming using a microphone array, with brighter colors indicating higher amplitudes.

[0046] In the spherical t-design arrangement shown in Figure 9, energy is concentrated in the direction where the elevation angle θ is zero, compared to the Fibonacci spiral arrangement shown in Figure 8. This sharpening of the beam directionality makes it possible to suppress unnecessary reflections or scattering from above and below, resulting in more precise measurements.

[0047] 10 is a diagram showing the signal-to-noise ratios of microphone arrays in the Fibonacci spiral configuration and the spherical t-design configuration. As shown in FIG. 10, the spherical t-design configuration improves the signal-to-noise ratio compared to the Fibonacci spiral configuration.

[0048] As described above, since multiple microphone elements 12 are arranged on the surface of sphere 11 according to the coordinate positions of the spherical t-design, it is possible to ensure the orthogonality of high-order spherical harmonic functions even when the number of microphone elements 12 is small. This makes it possible to measure or analyze a three-dimensional sound field with high accuracy.

[0049] 2. Variations The present disclosure is not limited to the above-described embodiments, and various modifications are possible as described below. Furthermore, the embodiments and modifications may be combined as appropriate.

[0050] 2-1. Variation 1 In the above-described embodiment, an example is given in which the number of microphone elements 12 arranged on the surface of sphere 11 is 60, but this is not limitative and the number is arbitrary as long as it is the number that can be arranged on the surface of sphere 11. However, as described above, the number is preferably 20 or more.

[0051] 2-2. Variation 2 In the above-described embodiment, an example is given in which the microphone element 12 is a MEMS microphone element, but this is not limiting and, for example, the microphone element 12 may be a condenser microphone element. In this case, the sound pressure signal output from the microphone element 12 is an analog signal, and it is preferable to place a circuit that converts the analog signal into a digital signal inside the sphere 11. This makes it possible to transmit the sound pressure signal as a digital signal outside the sphere 11. Note that the sound pressure signal may also be transmitted outside the sphere 11 as an analog signal.

[0052] 3. Notes From the above-described exemplary embodiments and modifications, the following aspects can be understood, for example.

[0053] (Supplementary Note 1) A first aspect, which is a preferred example of a microphone array according to the present disclosure, comprises a sphere and a plurality of microphone elements arranged on the surface of the sphere according to the coordinate positions of a spherical T-design. In this aspect, since the plurality of microphone elements are arranged on the surface of the sphere according to the coordinate positions of a spherical T-design, orthogonality of high-order spherical harmonics can be guaranteed even when the number of microphone elements is small. This makes it possible to measure or analyze a three-dimensional sound field with high accuracy.

[0054] (Supplementary Note 2) In a second aspect, which is a preferred example of the first aspect, each of the plurality of microphone elements is a MEMS microphone element. In the above aspect, a sound pressure signal indicating the sound pressure measured by each microphone element can be transmitted to the outside of the sphere as a digital signal. Therefore, even if a processing device for processing the sound pressure signal is placed far from the sphere, it is possible to reduce the contamination of noise into the sound pressure signal. Furthermore, by placing the processing device far from the sphere, it is also possible to reduce noise caused by sound reflected by the housing of the processing device, etc.

[0055] (Supplementary Note 3) In a third aspect, which is a preferred example of the first or second aspect, one of the plurality of microphone elements is disposed on the intersection of an imaginary plane that passes through the center of the sphere and is parallel to the horizontal plane with the surface of the sphere. In the above aspect, by using the one microphone element as a reference, setting work such as calibration can be easily performed.

[0056] (Supplementary Note 4) In a fourth aspect, which is a preferred example of any of the first to third aspects, none of the plurality of microphone elements is arranged in a predetermined area that includes the lower end of the sphere in the vertical direction. In this aspect, the predetermined area of ​​the sphere can be used as a space for passing wiring from each microphone element. The sphere can also be supported in the predetermined area.

[0057] (Supplementary Note 5) In a fifth aspect, which is a preferred example of any of the first to fourth aspects, the number of the plurality of microphone elements is greater than 20. In this aspect, the number of microphone elements can be increased compared to an aspect in which microphone elements are arranged at each vertex of a regular polyhedron. As a result, the measurement accuracy or analysis accuracy of a three-dimensional sound field can be improved.

[0058] (Supplementary Note 6) In a sixth aspect which is a preferred example of any of the first to fifth aspects, the spherical t-design is a spherical 2N-design. In the above aspect, orthogonality of spherical harmonics up to the Nth order can be guaranteed.

[0059] (Supplementary Note 7) In the seventh aspect, which is a preferred example of the sixth aspect, the spherical t-design is a spherical 10-design. In the above aspect, orthogonality of spherical harmonics up to the fifth order can be guaranteed. In addition, since the number of microphone elements is 60, 62, or 64, sound pressure signals from multiple microphone elements can be processed using a processing device capable of processing 64-channel sound pressure signals.

[0060] (Supplementary Note 8) In an eighth aspect which is a preferred example of any of the first to seventh aspects, the number of the plurality of microphone elements is 60. In the above aspect, sound pressure signals from the plurality of microphone elements are processed using 60 channels of a processing device capable of processing 64 channels of sound pressure signals, and the remaining channels can be effectively utilized for functional expansion, etc. Furthermore, instead of arranging microphone elements in a predetermined range including the lower end of the sphere in the vertical direction, one microphone element can be arranged on the intersection line between an imaginary plane which passes through the center of the sphere and is parallel to the horizontal plane and the surface of the sphere. [Explanation of symbols]

[0061] 10...microphone array, 11...sphere, 12...microphone element, 12-0...microphone element, 13...support, 20...processing device, 30...cable, 100...system, L...intersection line, R...area.

Claims

1. A sphere and a plurality of microphone elements arranged on the surface of the sphere according to coordinate positions of a spherical t-design; Microphone array.

2. Each of the plurality of microphone elements is a MEMS microphone element. The microphone array of claim 1 .

3. one microphone element of the plurality of microphone elements is disposed on an intersection line between an imaginary plane that passes through the center of the sphere and is parallel to a horizontal plane and a surface of the sphere; The microphone array of claim 1 .

4. none of the plurality of microphone elements is disposed in an area of ​​a predetermined range including a lower end of the sphere in the vertical direction; The microphone array of claim 1 .

5. the number of microphone elements is greater than 20; The microphone array of claim 1 .

6. The spherical t-design is a spherical 2N-design. The microphone array of claim 5 .

7. The spherical t-design is a spherical 10-design. The microphone array of claim 6.

8. The number of the plurality of microphone elements is 60. The microphone array of claim 7.

Images