Breakup vibration suppression device, breakup vibration suppression method, and program

Abstract

The present invention suppresses breakup vibrations in a speaker by performing filter processing on an audio source signal. To this end, a breakup vibration suppression device according to the disclosed technology comprises a non-linear expansion unit, a filter processing unit, an error signal calculation unit, an adaptive processing unit, and a storage unit. Exponential coefficients that form exponential-trigonometric functions and filter coefficients for use in filter processing are stored in the storage unit. The non-linear expansion unit generates a non-linear expanded signal by decomposing an audio source signal into exponential-trigonometric functions using the exponential coefficients in the storage unit. The filter processing unit generates a speaker drive signal by performing the filter processing on the non-linear expanded signal using the filter coefficients in the storage unit. The error signal calculation unit calculates an error signal on the basis of the audio source signal and the speaker drive signal. The adaptive processing unit updates the exponential coefficients and the filter coefficients using an algorithm that minimizes the mean squared error of the error signal, and stores the updated coefficients in the storage unit.

Description

Division vibration suppression device, division vibration suppression method, and program

[0001] The disclosed technology relates to a signal processing technology for suppressing split vibrations of an electrodynamic speaker.

[0002] [Divided Vibration] The diaphragm of an electrodynamic speaker unit can reproduce accurate sound waves by moving back and forth as a piston without bending as a whole. However, it takes a certain amount of time for this movement to propagate to every corner of the diaphragm, so the vibration actually propagates in a wavy manner. This phenomenon occurs constantly while the speaker is operating, causing distortion in the reproduced sound and degrading sound quality. In particular, at pitches (frequencies) where the wavelength of the wavy vibration and the size of the diaphragm are in an integer ratio, a "resonance phenomenon" occurs in which the wavy vibration is greatly amplified, which is a major factor in degrading sound quality. Figure 1 shows a schematic diagram of concentric divided vibration. Figure 1(a) shows two-division vibration, and (b) shows three-division vibration. The "+" indicates the wavy motion toward the front of the speaker, and the "-" indicates the wavy motion toward the rear of the speaker.

[0003] Split vibration also affects the directional characteristics of the speaker. Figure 2 shows the directional characteristics of an enclosure-less speaker measured by changing the frequency and intensity of the reproduced sound. It can be seen that the speaker has ideal dipole characteristics in the low range (300 Hz), but at high frequencies (800 Hz), the directional characteristics become uneven.

[0004] [Conventional split vibration suppression technology] Conventional methods for suppressing split vibration include partially changing the thickness or shape of the diaphragm to disperse the resonance frequency, or applying a damping material (a viscous material) to convert the resonance energy into heat in order to weaken the above-mentioned resonance phenomenon. There are also cases where split vibration is suppressed by dividing the flat diaphragm and controlling its drive.

[0005] "Physics of Speakers III: Characteristics of Dynamic Speakers", [Retrieved January 25, 2024], Internet<https: / / www.ne.jp / asahi / shiga / home / MyRoom / 9722dynamicspeaker.pdf> .

[0006] Conventional split vibration suppression measures are symptomatic treatments that only address the unevenness in frequency characteristics caused by resonance, and do not provide a fundamental solution to suppress the wavy diaphragm phenomenon itself.Furthermore, planar diaphragm speakers require a larger installation area than general electrodynamic speakers, which can be inconvenient, and replacement costs for installed electrodynamic speakers can be high.

[0007] A divided vibration suppression device according to the disclosed technology suppresses divided vibrations of a speaker by filtering a sound source signal, and includes a nonlinear expansion unit, a filter processing unit, an error signal calculation unit, an adaptive processing unit, and a memory unit. The memory unit stores exponential coefficients constituting an exponential trigonometric function and filter coefficients used in the filter processing. The nonlinear expansion unit generates a nonlinear expanded signal by expanding the sound source signal with the exponential trigonometric function using the exponential coefficients in the memory unit. The filter processing unit generates a speaker drive signal by filtering the nonlinear expanded signal using the filter coefficients in the memory unit. The error signal calculation unit calculates an error signal based on the sound source signal and the speaker drive signal. The adaptive processing unit updates the exponential coefficient and the filter coefficient using an algorithm that minimizes the mean square error for the error signal, and stores them in the memory unit.

[0008] The disclosed technology suppresses split vibrations of the speaker, improving the quality of the sound emitted from the speaker, achieving targeted directional characteristics over a wide frequency range, and increasing the maximum volume that can be driven.

[0009] A diagram showing a schematic example of concentric divided vibration. A diagram showing the directional characteristics of an enclosure-less speaker measured by changing the frequency and intensity of the reproduced sound. A diagram explaining the main points of the disclosed technology. A functional block diagram of a divided vibration suppression device according to a first embodiment. A flowchart explaining the operation of the divided vibration suppression device. A diagram showing an example of the functional configuration of a computer.

[0010] Hereinafter, embodiments of the disclosed technology will be described in detail. Note that components having the same functions are assigned the same numbers, and duplicated descriptions will be omitted.

[0011] The key features of the disclosed technology are explained using Figure 3. In conventional speakers that directly drive a speaker with a sound source signal, split vibrations cause deflection in the diaphragm 30. In the disclosed technology, a signal acquired by a microphone 32 installed near an electrodynamic speaker 31 and a sound source signal 33 are input, and a filter 34 is designed to suppress the split vibrations of the speaker. Because split vibrations are thought to have nonlinear characteristics due to the structure of the phenomenon, a filter update algorithm that can take nonlinearity into account is employed. This invention employs a technique called Exponential Functional Link Network (hereinafter abbreviated as EFLN). EFLN can estimate the nonlinear components of an input signal by expanding the input signal with an exponential trigonometric function. A distinctive feature of EFLN is its ability to represent various nonlinearities. For example, it can represent nonlinearities related to power changes, nonlinearities related to frequency changes, and nonlinearities resulting from the product of both. Using this algorithm, the complex behavior of split vibrations can be decomposed into controllable parameters.

[0012] [First embodiment] Fig. 4 is a functional block diagram showing an example of the configuration of a split vibration suppression device 4 according to the first embodiment. The split vibration suppression device 4 includes a nonlinear expansion unit 41, a transfer characteristic superposition unit 42, a filter processing unit 43, a delay adding unit 44, an error signal calculation unit 45, an adaptive algorithm processing unit 46, and a storage unit 47. Fig. 5 is a flowchart explaining an example of the operation of the split vibration suppression device 4. Below, the split vibration suppression device 4 will be explained in detail using Figs. 4 and 5.

[0013] The divided vibration suppression device 4 receives a sound source signal x from a certain time n to M times before. → (n) is input (step S501).

[0014] [Nonlinear Expansion] The nonlinear expansion unit 41 → (n) is a nonlinearly expanded signal g → (n) is generated (step S502). where q(n) is the exponential coefficient, p=1,2,...,P is the expansion order, and g → iThe symbol between the exponential function and trigonometric function of (n) represents the Hadamard product. q(n) is stored in the memory unit 47. The power of the nonlinearly expanded signal is adjusted using the exponential term. A zero value for q(n) represents a constant value, a positive value represents exponential decay, and a negative value represents exponential increase. The trigonometric function term also extracts harmonic components. p specifies the maximum order of the harmonics to be tracked.

[0015] This is necessary for updating q(n) is also calculated in advance (step S502).

[0016] [Transfer characteristic superposition] The transfer characteristic from the speaker to the observation microphone, which has been measured in advance, is stored in the storage unit 47 as a transfer characteristic model s^. s^ is a vector quantity that does not depend on time. When the sound source signal after nonlinear expansion is input, the transfer characteristic superposition unit 42 calculates signals g^(n) and h^(n) by convolving the transfer characteristic s^ with the signal (step S504).

[0017] [Filtering] The memory unit 47 stores filter coefficients. When the sound source signal after nonlinear extension is input to the filter processing unit 43, the signal is multiplied by a filter coefficient to generate a signal y(n) for driving a speaker (step S505).

[0018] [Speaker Sound Emission and Microphone Observation] The speaker 31 is driven by the signal y(n) to emit sound (step S506), and the emitted sound after spatial transmission is observed by the microphone 32 (step S507). → Then, the microphone observation signal u → (n) is expressed as follows:

[0019] [Delay Addition Processing] The pure delay (delay time from time 0 to the arrival time of the first wavefront) is measured (delay is defined as τ) from the time waveform of the transfer characteristic from the speaker to the observation microphone, and the input sound source signal is delayed by this time to obtain a signal d →(n) is output (step S508). This is equivalent to adding the propagation delay from the speaker emission to the observation to the sound source signal. Expressed as an equation, it becomes as follows.

[0020] [Error signal calculation] Observation signal u → (n) and correct signal d → Error e of (n) → (n) is calculated (step S509) using the following equation.

[0021] [Adaptive algorithm processing] Calculate the exponential coefficient q(n+1) and filter coefficient w(n+1) for nonlinear expansion to be applied to the next sound source signal x(n+1). The observed signal can be expressed by the following equation using the speaker drive signal and the spatial transfer characteristics from the speaker to the microphone. Therefore, the error signal can be rewritten as: In the adaptive algorithm processing unit, the filter coefficients and exponent coefficients of Equation (12) are updated by adaptive processing based on the mean square error minimization criterion using the following equations (step S510). Equation (13) is an update equation that includes normalization processing, and μ w , μ g is the step size parameter, and,x,ξ,is the regularization parameter.

[0022] In the next cycle of the division vibration suppression process shown in FIG. → (n+1) is input (step S501), an extension signal is generated using q(n+1) (step S502), and w → (n+1) is used for filtering (step S505).

[0023] The above is the description of the first embodiment.

[0024] [Supplementary Information] In the above embodiment, the exponential coefficients, filter coefficients, and transfer characteristic model are stored in the storage unit 47. However, the exponential coefficients may be stored in the nonlinear expansion unit 41, the filter coefficients in the filter processing unit 43, and the transfer characteristic model in the transfer characteristic superposition unit 42. When the disclosed technology is applied to a PSZ speaker (a technology that uses out-of-phase sound waves emitted from the rear of a speaker to create an area where sound can be heard only near the speaker, or a technology that generates out-of-phase sound waves to create an area where ambient sound is blocked), the following additional effects can be obtained: (i) In spot playback applications, the sound containment effect can be broadened. In addition, the maximum playback volume can be increased. (ii) In active noise control applications, the cancellation effect and the manageable noise level are improved.

[0025] [Program, Recording Medium] The functions realized by the components described in this specification may be implemented in circuitry or processing circuitry, including general-purpose processors, application-specific processors, integrated circuits, ASICs (Application Specific Integrated Circuits), CPUs (Central Processing Units), conventional circuits, and / or combinations thereof, programmed to realize the described functions. A processor includes transistors and other circuits and is considered to be circuitry or processing circuitry. A processor may be a programmed processor that executes a program stored in a memory.

[0026] In this specification, a circuitry, unit, or means is hardware that is programmed to realize or performs the described functions, which may be any hardware disclosed herein or any hardware known to be programmed to realize or perform the described functions.

[0027] If the hardware is a processor considered to be a type of circuitry, the circuitry, means, or unit is a combination of the hardware and software used to configure the hardware and / or processor.

[0028] The various processes described above can be implemented by loading a program that executes each step of the above method into the recording unit 2020 of the computer 2000 shown in Figure 6, and operating the control unit 2010, input unit 2030, output unit 2040, display unit 2050, etc.

[0029] The program describing the processing contents can be recorded on a computer-readable recording medium, which may be, for example, a magnetic recording device, an optical disk, a magneto-optical recording medium, a semiconductor memory, or any other suitable recording medium.

[0030] The program may be distributed by, for example, selling, transferring, lending, etc. portable recording media such as DVDs and CD-ROMs on which the program is recorded. Furthermore, the program may be stored in a storage device of a server computer, and then transferred from the server computer to other computers via a network, thereby distributing the program.

[0031] A computer that executes such a program may first temporarily store the program recorded on a portable recording medium or transferred from a server computer in its own storage device. Then, when executing a process, the computer reads the program stored on its own recording medium and executes the process in accordance with the read program. Alternatively, the computer may read the program directly from a portable recording medium and execute the process in accordance with the program. Furthermore, the computer may execute the process in accordance with the program each time a program is transferred from a server computer to the computer. Alternatively, the server computer may not transfer the program to the computer, but may instead execute the process through a so-called ASP (Application Service Provider) service, which realizes the processing function by issuing an execution instruction and obtaining the results. Furthermore, the server computer may execute the process at the terminal using a so-called SaaS (Software as a Service) service, which allows users to use part of a server computer along with the program. In this embodiment, the program includes information used for processing by an electronic computer that is equivalent to a program (such as data that is not a direct instruction to a computer but has properties that dictate computer processing).

[0032] Furthermore, in this embodiment, the device is configured by executing a predetermined program on a computer, but at least a part of the processing contents may be realized by hardware.

[0033] 31 Speaker 32 Microphone 4 Divided vibration suppression device 41 Nonlinear expansion section 42 Transfer characteristic superposition section 43 Filter processing section 44 Delay adding section 45 Error signal detection section 46 Adaptive processing section 47 Storage section 2000 Computer 2010 Control section 2020 Recording section 2030 Input section 2040 Output section 2050 Display section

Claims

1. A method for suppressing split vibration of a speaker by filtering a sound source signal, comprising: acquiring an exponential coefficient and a filter coefficient; a nonlinear extension unit using the exponential coefficient to generate a nonlinear extended signal by expanding the sound source signal with an exponential trigonometric function; a filter processing unit using the filter coefficient to perform filter processing on the nonlinear extended signal to generate a speaker drive signal; an error signal calculation unit calculating an error signal based on the sound source signal and the speaker drive signal; and an adaptive processing unit updating the exponential coefficient and the filter coefficient using an algorithm that minimizes the mean square error of the error signal.

2. A method for suppressing split vibration of a speaker by filtering a sound source signal, comprising: acquiring an exponential coefficient and a filter coefficient; a nonlinear expansion unit using the exponential coefficient to generate a nonlinear expanded signal by expanding the sound source signal with an exponential trigonometric function; a filter processing unit using the filter coefficient to perform filter processing on the nonlinear expanded signal to generate a speaker drive signal; a speaker generating a radiated sound from the speaker drive signal; a microphone observing the radiated sound to obtain a microphone signal; a delay adding unit generating a delayed sound source signal based on the sound source signal; an error signal calculation unit calculating an error signal based on the microphone signal and the delayed sound source signal; and an adaptive processing unit updating the exponential coefficient and the filter coefficient using an algorithm that minimizes the mean square error for the error signal.

3. A divisional vibration suppression method according to claim 2, wherein a transfer characteristic superposition unit acquires the transfer characteristic from the speaker to the microphone and calculates a second nonlinear extension signal by adding the transfer characteristic to the nonlinear extension signal, and the adaptive processing unit updates the exponential coefficient and the filter coefficient using the second nonlinear extension signal.

4. A device for filtering a sound source signal to suppress split vibrations of a speaker, comprising: a memory unit that stores exponential coefficients and filter coefficients; a nonlinear expansion unit that uses the exponential coefficients to generate a nonlinear expansion signal by expanding the sound source signal with an exponential trigonometric function; a filter processing unit that uses the filter coefficients to filter the nonlinear expansion signal to generate a speaker drive signal; an error signal calculation unit that calculates an error signal based on the sound source signal and the speaker drive signal; and an adaptive processing unit that updates the exponential coefficients and filter coefficients of the memory unit using an algorithm that minimizes the mean square error for the error signal.

5. A device for filtering a sound source signal to suppress split vibrations of a speaker, comprising: a memory unit that stores exponential coefficients and filter coefficients; a nonlinear expansion unit that uses the exponential coefficients to generate a nonlinear expansion signal by expanding the sound source signal with an exponential trigonometric function; a filter processing unit that uses the filter coefficients to perform filter processing on the nonlinear expansion signal to generate a speaker drive signal; a speaker that generates radiated sound from the speaker drive signal; a microphone that observes the radiated sound and obtains a microphone signal; a delay adding unit that generates a delayed sound source signal based on the sound source signal; an error signal calculation unit that calculates an error signal based on the microphone signal and the delayed sound source signal; and an adaptive processing unit that updates the exponential coefficients and filter coefficients of the memory unit using an algorithm that minimizes the mean square error for the error signal.

6. A divided vibration suppression device according to claim 5, comprising a transfer characteristic superposition unit that acquires transfer characteristics from the speaker to the microphone and calculates a second nonlinear extension signal by adding the transfer characteristics to the nonlinear extension signal, and the adaptive processing unit updates the exponential coefficient and the filter coefficient using the second nonlinear extension signal.

7. A program for causing a computer to function as the split vibration suppression device according to any one of claims 4 to 6.

Images