WO2019189481A1

WO2019189481A1 - Acoustic analysis system

Info

Publication number: WO2019189481A1
Application number: PCT/JP2019/013390
Authority: WO
Inventors: 恵野崎; 直穂子豊嶋
Original assignee: 日本電産株式会社
Priority date: 2018-03-28
Filing date: 2019-03-27
Publication date: 2019-10-03
Also published as: CN112005087A

Abstract

An acoustic analysis system 1000 is provided with a plurality of microphone array modules each of which has a plurality of MEMS microphones, and which perform control relating to recording by the plurality of MEMS microphones, and a control unit which controls the plurality of microphone array modules. Each microphone array module has a receiving unit for receiving signals from the plurality of MEMS microphones, and a transmitting unit for transmitting the signals of the plurality of MEMS microphones to the control unit. The control unit receives the plurality of MEMS microphone signals from each of the plurality of microphone array modules, and processes the same as signals used for acoustic analysis.

Description

Acoustic analysis system

The present invention relates to an acoustic analysis system.

In recent years, due to the increasing demand for noise reduction of products, it is required to measure and analyze the spatial distribution of the sound field.
Patent Document 1 discloses a sound pressure distribution analysis system using a microphone array in which a plurality of microphones are arranged in a lattice pattern and sounds are detected at a plurality of positions. This sound pressure distribution analysis system includes an amplifier capable of amplifying a multi-channel signal, and the amplifier amplifies each sound signal of the microphone and outputs it to the analysis terminal. The analysis terminal A / D converts the sound signal input from the amplifier and records it as a time waveform.

Japanese Laid-Open Patent Publication No. 2005-91272

By the way, in acoustic analysis, there is a desire to adjust the number of microphones according to the size of the object to be measured, from the desire to measure all objects.
However, if a condenser microphone or dynamic microphone is used as in the conventional sound pressure distribution analysis system, the number of microphones is increased when the number of microphones is increased in response to a large object to be measured. Only an amplifier and an A / D converter are required separately, and complicated changes in system design are required.
Therefore, an object of the present invention is to provide an acoustic analysis system in which the number of microphones can be easily changed according to the size of an object to be measured.

In order to solve the above problem, an acoustic analysis system according to one aspect of the present invention includes a plurality of microphone array modules and a control unit that controls the plurality of microphone array modules. The plurality of microphone array modules have a plurality of MEMS microphones, respectively, and perform control related to recording of the plurality of MEMS microphones. The microphone array module includes a reception unit that receives signals from the plurality of MEMS microphones, and a transmission unit that transmits signals from the plurality of MEMS microphones to the control unit. The signals of the plurality of MEMS microphones are respectively received from the microphone array module and processed as signals used for acoustic analysis.

According to one aspect of the present invention, the number of microphones can be easily increased by adding a microphone array module having MEMS microphones each capable of incorporating an amplifier, an A / D converter, and the like. Therefore, the number of microphones can be easily changed according to the size of the object to be measured, and an appropriate sound field distribution according to the object to be measured can be measured.

FIG. 1 is a diagram illustrating an example of an acoustic analysis system. FIG. 2 is a diagram illustrating a configuration example of a control system of the microphone array. FIG. 3 is a diagram illustrating a configuration example of the microphone array module. FIG. 4 is a diagram illustrating a configuration example of the control unit. FIG. 5 is a diagram illustrating an example of an analog signal. FIG. 6 is a diagram illustrating a method for calculating the delay time. FIG. 7 is a diagram illustrating another example of the microphone array module.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
The scope of the present invention is not limited to the following embodiment, and can be arbitrarily changed within the scope of the technical idea of the present invention.

FIG. 1 is a configuration example of an acoustic analysis system 1000 including a microphone array 1 according to this embodiment.
The acoustic analysis system 1000 according to this embodiment is a system that analyzes a sound to be measured from the object to be measured (sound source) 2 using a near-field acoustic holography method. In the near-field acoustic holography method, it is necessary to measure a sound pressure distribution on a measurement surface that is close to and parallel to the sound source surface 2a, and a microphone array 1 in which a plurality of microphones mc are arranged in a lattice shape is used.
The microphone array 1 in the present embodiment includes M × N microphones mc arranged in a lattice pattern. The microphone mc may be a MEMS (Micro-Electrical-Mechanical Systems) microphone, for example. The acoustic analysis system 1000 analyzes a signal input from each of the M × N microphones mc, and detects a physical quantity representing a sound characteristic.

The acoustic analysis system 1000 includes an acoustic analysis device 100 and a display device 200.
The acoustic analysis device 100 includes a signal processing unit 101, an analysis processing unit 102, and a storage unit 103. The signal processing unit 101 performs predetermined signal processing on the signal from each microphone mc of the microphone array 1 to obtain a signal used for acoustic analysis. In the present embodiment, the signal processing unit 101 performs processing for synchronizing the signals of M × N microphones mc included in the microphone array 1. A specific configuration of the signal processing unit 101 will be described later.

The analysis processing unit 102 analyzes the signal subjected to the signal processing by the signal processing unit 101 and detects a physical quantity representing the feature of the sound. In other words, the analysis processing unit analyzes the signal processed by the control unit, and detects a physical quantity representing the feature of the sound. Here, the physical quantity representing the characteristics of the sound includes a sound pressure distribution, a particle velocity distribution, and the like. Then, the analysis processing unit 102 generates an image corresponding to the physical quantity representing the feature of the sound, and performs display control for displaying the image on the display device 200.
The storage unit 103 stores the analysis result by the analysis processing unit 102 and the like.
The display device 200 includes a monitor such as a liquid crystal display, and displays the image that is the analysis result of the acoustic analysis device 100.

FIG. 2 is a diagram illustrating a configuration example of a control system of the microphone array 1.
The control system of the microphone array 1 includes M microphone array modules C (1) to C (M) and a control unit B that controls these M microphone array modules C (1) to C (M). Prepare. The microphone array modules C (1) to C (M) have the same configuration.
The microphone array module C (m) controls the N microphones mc (m, 1) to mc (m, N) and the recording of these N microphones mc (m, 1) to mc (m, N). A microphone array control unit A (m). The N microphones mc (m, 1) to mc (m, N) are arranged in one horizontal row or one vertical row among the M × N microphones mc arranged in a lattice pattern included in the microphone array 1. Microphone mc. The microphone array module C (m) can be a smart speaker (AI speaker), for example.
M microphone array controllers A (1) to A (M) receive signals (microphone input signals) from N microphones, respectively. It is assumed that the N microphones controlled by one microphone array control unit A (m) are electrically synchronized.

The control unit B can be included in the signal processing unit 101 of FIG. The control unit B transmits a recording command and an analog signal S (t) to the microphone array modules C (1) to C (M), respectively.
Each microphone array module C (m) receives a recording command transmitted from the control unit B, and in response to the recording command, microphone input signals of N microphones mc and an analog signal from the control unit B A function of simultaneously receiving S (t) in the same sampling period dt. Further, each microphone array module C (m) performs A / D conversion on the N microphone input signals and the analog signal S (t) acquired at this time, and N microphone input data which are signals for acoustic analysis. D (m, 1, dt) to D (m, n, dt) and analog input data Ds (m, dt) are transmitted to the control unit B, respectively.

FIG. 3 is a diagram illustrating a configuration example of the microphone array module C (m).
The microphone array control unit A (m) includes an input / output control unit 111, N input signal reception units 112, and one analog signal reception unit 113.
The input / output control unit 111 controls input / output of data in the microphone array module C (m). Specifically, when the input / output control unit 111 receives the recording command transmitted from the control unit B, in response to the recording command, the N input signal receiving units 112 and the analog signal receiving unit 113 are the same. Control is performed so that signals are simultaneously received at the sampling period dt. That is, the receiving unit receives the signals of the plurality of MEMS microphones simultaneously with the analog signal in response to the recording command.

Under the control of the input / output control unit 111, the N input signal receiving units 112 receive microphone input signals Mic (m, 1, t) to Mic (m) from the microphones mc (m, 1) to mc (m, N), respectively. , N, t). The microphone mc incorporates an acoustic transducer (MEMS chip) using an MEMS technology and an amplifier, and is surface-mounted on a substrate. The microphone mc converts sound (sound pressure) into an electrical signal using an acoustic transducer, amplifies the converted electrical signal using an amplifier, and outputs the amplified signal.
Each input signal receiving unit 112 performs A / D conversion on the received microphone input signals Mic (m, 1, t) to Mic (m, N, t), respectively, and microphone input data D (m, 1, dt). ) To D (m, N, dt) are output to the input / output control unit 111.

Further, the analog signal receiving unit 113 receives the analog signal S (t) from the control unit B under the control of the input / output control unit 111. The analog signal S (t) transmitted from the control unit B is directly input to the analog signal receiving unit 113 without using a delay circuit. The analog signal receiving unit 113 A / D converts the received analog signal S (t) and outputs analog input data Ds (m, dt) to the input / output control unit 111.

The input / output control unit 111 includes N microphone input data D (m, 1, dt) to D (m, n, dt) output from the N input signal receiving units 112, and an analog signal receiving unit 113 outputs Analog input data Ds (m, dt) to be received. The input / output control unit 111 transmits the received microphone input data D (m, 1, dt) to D (m, n, dt) to the control unit B together with the analog input data Ds (m, dt). That is, the transmission unit transmits the signals of the plurality of MEMS microphones received by the reception unit to the control unit together with the analog signals.

FIG. 4 is a diagram illustrating a configuration example of the control unit B.
The control unit B has a function of transmitting a recording command to the microphone array control units A (1) to A (M) of the microphone array modules C (1) to C (M), and the microphone array control unit A (1). To transmit an analog signal S (t) to A (M). That is, the control unit includes a recording command unit that transmits a recording command to the plurality of microphone array modules, and an analog signal transmission unit that transmits an analog signal to the plurality of microphone array modules. The control unit B has a function of receiving microphone input data D (m, n, dt) and analog input data Ds (m, dt) from the microphone array control units A (1) to A (M). . Further, the control unit B has a function of aligning the phases of the microphone input data D (m, n) received from the microphone array control units A (1) to A (M). That is, the control unit performs a process of aligning the phases of the signals of the plurality of MEMS microphones received from the plurality of microphone array modules.

The control unit B includes a recording command unit 114, an analog signal transmission unit 115, a delay calculation unit 116, and a delay correction unit 117.
The recording command unit 114 generates a recording command and transmits the recording command to the M microphone array control units A (1) to A (M).
The analog signal transmission unit 115 includes a signal generation unit 115a and a D / A conversion unit 115b, generates an analog signal S (t), and M microphone array control units A (1) to A (M ).

FIG. 5 is a diagram illustrating an example of the analog signal S (t).
The analog signal S (t) is a signal that oscillates at a constant period and whose amplitude fluctuates. For example, the analog signal S (t) oscillates at a constant period T1 that is about 10 times the sampling period dt, and its amplitude A ₀ is twice the period T1. It can be set as the signal which fluctuates with the above period T2. In other words, the analog signal transmission unit transmits a signal that oscillates at a constant cycle and varies in amplitude as an analog signal.
That is, the analog signal S (t) can be expressed by the following equation.
S (t) = A ₀ (t) · sin (2πt / T1) (1)
Here, the amplitude A ₀ (t) can be expressed by the following equation.
A ₀ (t) = sin (2πt / T2) (2)

Returning to FIG. 4, the delay calculation unit 116 calculates the delay time td (m) of the microphone signal acquired from each microphone array module C (m). In the present embodiment, the delay calculation unit 116 calculates the delay time td (m) based on the analog input data Ds (m, dt) transmitted by each microphone array control unit A (m). That is, the delay calculation unit calculates the delay times of the signals of the plurality of MEMS microphones received from the plurality of microphone array modules based on the analog signal transmitted by the transmission unit.
Specifically, the delay calculation unit 116 transmits the analog input data Ds (m0, dt) transmitted from the arbitrary microphone array control unit A (m0) as a reference and transmits from the other microphone array control unit A (m). Pattern matching is performed on the analog input data Ds (m, dt). Then, based on the result of the pattern matching, the delay calculation unit 116 calculates the difference between the phase of the analog input data Ds (m0) and the phase of the analog input data Ds (m), and based on the difference, The delay time td (m) of data received from the microphone array module C (m) is calculated. That is, the delay calculation unit includes the phase of the analog signal transmitted from the transmission unit included in the first microphone array module, which is one of the plurality of microphone array modules, and the other one of the plurality of microphone array modules. The delay time is calculated based on the difference from the phase of the analog signal transmitted from the transmitter included in the second microphone array module.

FIG. 6 is a diagram illustrating a method for calculating the delay time td (m).
In FIG. 6, the triangle mark (▲) is a plot of analog input data Ds (m0) as a reference, and the cross mark (×) is a microphone array module C (target for calculating the delay time td (m). It is the figure which plotted the analog input data Ds (m) output from m).

As described above, the delay calculation unit 116 applies the data acquired from the microphone array module C (m0) based on the difference between the phase of the analog input data Ds (m0) and the phase of the analog input data Ds (m). A delay time td (m) of data acquired from the microphone array module C (m) is calculated.
The delay calculation unit 116 calculates delay times td (1) to td (M) for the microphone array modules C (1) to C (M), respectively. Then, the delay calculation unit 116 outputs the calculated delay times td (1) to td (M) to the delay correction unit 117.

The delay correction unit 117 calculates the phase of the microphone input data D (m, n, dt) acquired from each microphone array module C (m) based on the delay time td (m) calculated by the delay calculation unit 116. Alignment processing is performed, and corrected microphone input correction data D ′ (m, n, dt) is calculated. That is, the delay correction unit performs processing for aligning the phases of the signals of the plurality of MEMS microphones received from the plurality of microphone array modules based on the delay time.
For example, the delay correcting unit 117 rounds the delay time td (m) to an integral multiple k of the sampling period dt, and shifts the microphone input data D (m, n, dt) by the delay time (k points) in the time direction. The delay time is corrected by the shift method. That is, the delay correction unit shifts the signals of the plurality of MEMS microphones received from the microphone array module by the delay time.
D ′ (m, n, dt) = D (m, n, dt + k) (3)
k = int (td (m) / dt) (4)
In the above equation (4), int () is a function that truncates the value in parentheses after the decimal point.

As another correction method, for example, the delay correction unit 117 may use a method in which the microphone input data D (m, n, dt) is frequency-resolved and the phase is advanced for each frequency. In this case, using the transfer function H (k) = exp (−dω (k)), the calculation represented by the following equation is performed.
F ⁻¹ (F (D (m, n, dt))） H (k)) (5)
In the above equation (5), ◎ represents a convolution integral.

As described above, the delay correction unit 117 generates frequency domain data F (D (m, n, dt)) by performing Fourier transform on the microphone input data D (m, n, dt), and the frequency domain data is converted into the frequency domain data. The transfer function H (k) corresponding to the integer k derived from the delay time td (m) is convolved and inverse Fourier transformed. In this way, the delay time may be corrected in the frequency space. That is, the delay correction unit generates frequency domain data by performing Fourier transform on the signals of the plurality of MEMS microphones received from the microphone array module, and convolves the transfer function corresponding to the delay time with the frequency domain data to perform inverse Fourier transform. By doing so, the signals of several MEMS microphones are corrected in the frequency space.
The method for correcting the delay time is not limited to the above, and any method can be applied. Further, the function of the control unit B described above may be realized by an arbitrary microphone array control unit A (m0). That is, one of the plurality of microphone array modules may include a control unit.

By the way, a microphone array for acoustic analysis generally requires microphones of 32ch or more, and it is necessary to accurately synchronize signals from each microphone. For this reason, it is common that one control unit performs control related to recording of all microphones constituting the microphone array.
However, for example, when a large-scale microphone array exceeding 1000 channels is used, in order for one control unit to control all the microphones, there is a problem that a dedicated machine needs to be developed and the system becomes expensive.

Thus, in this embodiment, an acoustic analysis system including an M × N microphone array is configured by combining a plurality (M) of microphone array modules C (m) that perform control related to recording of a plurality (N) of microphones.
Specifically, the acoustic analysis system includes M microphone array modules C (m) that respectively control N microphones, and a control unit B that controls these M microphone array modules C (m). Prepare. The microphone array module C (m) includes a receiving unit that receives N microphone signals and a transmitting unit that transmits N microphone signals to the control unit B. The control unit B includes M number of signals. The microphone array module C (m) receives a microphone signal and processes it as a signal used for acoustic analysis.

With such a configuration, for example, a smart speaker (AI speaker) can be used for acoustic analysis that requires more than 32 channels by using multiple microphone array modules C (m) of about 8 channels that are widely used for consumer use. A large-scale acoustic analysis system can be realized at low cost.
Further, in acoustic analysis, there is a demand for adjusting the size of the microphone array (the number of microphones) according to the size of the object to be measured because of the desire to measure all objects. In the present embodiment, the microphone mc included in the microphone array module C (m) is a MEMS microphone including an amplifier, and the microphone array module C (m) includes a microphone, an amplifier, an A / D converter, and the like. Therefore, by adjusting the number of microphone array modules C (m) connected to the control unit B, the number of microphones constituting the microphone array can be easily adjusted without requiring complicated system design changes. Can do. As described above, the acoustic analysis system according to the present embodiment can be a system in which the number of microphones can be easily changed and an object to be measured of any size can be easily measured.

Here, as described above, the microphone array module C (m) can be a smart speaker (AI speaker). Thus, a plurality of AI speakers can be combined and used for acoustic analysis.
However, in the case of a system in which a plurality of AI speakers are combined, even if recording commands are simultaneously transmitted from the control unit to the plurality of AI speakers (microphone array module), each AI speaker (microphone array) may be delayed due to a delay in communication processing. Module) measures the sound at each timing. That is, the microphone signal can be easily synchronized within the AI speaker, but the microphone signal cannot be synchronized between different AI speakers.

As described above, in the acoustic analysis, it is necessary to accurately synchronize the signals of the microphones. Therefore, when the number of microphone array modules (AI speakers) is increased in order to increase the number of microphones constituting the microphone array, it is necessary to accurately synchronize microphone signals between the microphone array modules (AI speakers). Therefore, in the present embodiment, the control unit B inputs an analog signal S (t) as a synchronization signal to each microphone array module C (m). In each microphone array module C (m), the microphone input signals Mic (m, 1, t) to Mic (m, N, t) are received simultaneously with the analog signal S (t), and the microphone input data D (m , N, dt) is transmitted to the control unit B together with the analog input data Ds (m, dt).

As a result, even when the recording start timing is different in each microphone array module, the control unit B receives the analog input data Ds (m, dt) received together with the microphone input data D (m, n, dt). Can be used to align the phases of all microphone input data D (m, n). In other words, in a system in which different AI speakers are combined, it is possible to synchronize data between different AI speakers.
Here, the analog signal S (t) transmitted from the control unit B to each microphone array module C (m) is a signal that vibrates at a constant period. Thereby, the delay time of the signal of each microphone can be calculated and synchronization can be achieved relatively easily. Further, by varying the amplitude of the analog signal S (t), it is possible to appropriately synchronize even when the delay time is one period or more of the analog signal S (t).

As described above, in the present embodiment, the number of microphones can be easily changed according to the size of the object to be measured, and the sound that can be used to appropriately obtain synchronized sound data that can be used for acoustic analysis can be obtained. An analysis system can be realized.

(Modification)
In the above embodiment, the case where the microphone array control unit A (m) has a function of directly receiving the analog signal S (t) transmitted from the control unit B has been described. However, the microphone array control unit A (m) may receive sound based on the analog signal S (t) instead of directly receiving the analog signal S (t).
In this case, for example, as shown in FIG. 7, a microphone mc (m, N + 1) as a substitute for analog input is provided in the microphone array module C (m). A speaker SP (m) is provided at a certain close distance from the microphone mc (m, N + 1). The speaker SP (m) is a sound output unit that outputs sound in the ultrasonic band based on the analog signal S (t) from the control unit B under the control of the input / output control unit 111. The microphone mc (m, N + 1) picks up the sound output from the speaker SP (m) and outputs the microphone input signal Smic (m, t). The analog signal receiving unit 113 receives the microphone input signal Smic (m, t) as an analog signal S (t). That is, the microphone array module further includes an audio output unit that outputs audio based on the analog signal transmitted by the analog signal transmission unit, and the reception unit receives the audio as an analog signal.
Thus, the microphone array module C (m) may receive the analog signal S (t) by audio input. Thereby, even if the microphone array module C (m) does not have an analog input terminal, the same effect as that of the above-described embodiment can be obtained.

DESCRIPTION OF SYMBOLS 1 ... Microphone array, 2 ... Measured object (sound source), 2a ... Sound source surface, 100 ... Sound analysis apparatus, 101 ... Signal processing part, 102 ... Analysis processing part, 103 ... Memory | storage part, 200 ... Display apparatus, 1000 ... Sound Analysis system, A (1) to A (M) ... microphone array module, B ... control unit, mc ... microphone

Claims

A plurality of microphone array modules each having a plurality of MEMS microphones and performing control related to recording of the plurality of MEMS microphones;
A control unit for controlling the plurality of microphone array modules,
The microphone array module is
A receiver for receiving signals from the plurality of MEMS microphones;
A transmission unit that transmits signals of the plurality of MEMS microphones to the control unit;
The control unit receives signals of the plurality of MEMS microphones from the plurality of microphone array modules, respectively, and processes the signals as signals used for acoustic analysis.
The acoustic analysis system according to claim 1, wherein the control unit performs a process of aligning phases of signals of the plurality of MEMS microphones received from the plurality of microphone array modules.
The controller is
A recording command unit that transmits a recording command to the plurality of microphone array modules;
An analog signal transmitter that transmits an analog signal to the plurality of microphone array modules;
In response to the recording command, the receiving unit receives the signals of the plurality of MEMS microphones simultaneously with the analog signal,
The acoustic analysis system according to claim 2, wherein the transmission unit transmits the signals of the plurality of MEMS microphones received by the reception unit to the control unit together with the analog signal.
The microphone array module is
An audio output unit that outputs audio based on the analog signal transmitted by the analog signal transmission unit;
The acoustic analysis system according to claim 3, wherein the reception unit receives the voice as the analog signal.
The acoustic analysis system according to claim 3 or 4, wherein the analog signal transmission unit transmits a signal that vibrates at a constant period and varies in amplitude as the analog signal.
The controller is
A delay calculation unit that calculates delay times of signals of the plurality of MEMS microphones received from the plurality of microphone array modules based on the analog signals transmitted by the transmission unit;
The delay correction part which performs the process which aligns the phase of the signal of these MEMS microphones received from the said several microphone array module based on the said delay time further, It has any one of Claim 3 to 5 characterized by the above-mentioned. The acoustic analysis system according to claim 1.
The delay calculation unit
A phase of the analog signal transmitted from the transmitter included in the first microphone array module which is one of the plurality of microphone array modules;
Calculating the delay time based on a difference from the phase of the analog signal transmitted from the transmitter included in the second microphone array module which is another one of the plurality of microphone array modules. The acoustic analysis system according to claim 6.
The acoustic analysis system according to claim 6 or 7, wherein the delay correction unit shifts signals of the plurality of MEMS microphones received from the microphone array module by the delay time.
The delay correction unit includes:
Generate frequency domain data by Fourier transforming the signals of the plurality of MEMS microphones received from the microphone array module;
8. The signals of the plurality of MEMS microphones are corrected in a frequency space by convolving a transfer function corresponding to the delay time into the frequency domain data and performing inverse Fourier transform. Acoustic analysis system.
The acoustic analysis system according to any one of claims 1 to 9, wherein one of the plurality of microphone array modules includes the control unit.
The acoustic analysis system according to any one of claims 1 to 10, further comprising an analysis processing unit that analyzes the signal processed by the control unit and detects a physical quantity representing a characteristic of a sound.