Digital equalization method and device of loudspeaker, vehicle and medium

Abstract

The invention relates to a digital equalization method and device of a loudspeaker, a vehicle and a medium. The method comprises the following steps: acquiring a pulse response signal output by the loudspeaker at a signal acquisition point; determining a time domain equalization parameter of the filter according to the obtained pulse response signal and a preset target pulse response signal; performing time domain filtering on a test signal input into the loudspeaker according to the time domain equalization parameter of the filter; converting the signal after time domain filtering to a frequency domain to obtain an amplitude-frequency response signal; performing frequency domain equalization on the amplitude-frequency response signal obtained by conversion to obtain a frequency domain equalization parameter of the filter; and combining the frequency domain equalization parameter of the filter and the time domain equalization parameter of the filter to obtain a target parameter of the filter. According to the scheme, the filter is solved in two steps including the time domain and the frequency domain, balance of the loudspeaker is completed, and the problem that in the related technology, the sound quality of the loudspeaker is poor due to the fact that the filter is designed only for the frequency domain is solved.

Description

Technical Field

[0001] The present disclosure relates to the technical field of speaker equalization, and particularly to a digital equalization method, device, vehicle and medium for a speaker. Background Art

[0002] A speaker is a basic unit of an acoustic system, and the quality of the speaker directly affects the sound reproduction effect. Due to the deficiencies in the manufacturing process during the production of traditional speakers, various distortions of the speaker unit are likely to occur, resulting in the inability of the speaker system to form an ideal sound field space. An intelligent speaker system using digital signal processing technology, that is, a speaker digital equalization method, can make up for the deficiencies and defects in the production process of traditional speakers, making its frequency response and time-domain response tend to be ideal, correcting and compensating for the sound to obtain a more realistic sound reproduction effect. Usually, a filter is designed to achieve the digital equalization of the speaker. Summary of the Invention

[0003] To overcome the problems existing in the related art, the present disclosure provides a digital equalization method, device, vehicle and medium for a speaker.

[0004] According to a first aspect of an embodiment of the present disclosure, a digital equalization method for a speaker is provided, and the method includes:

[0005] Obtain the impulse response signal output by the speaker at the signal acquisition point;

[0006] Determine the time-domain equalization parameters of the filter according to the obtained impulse response signal and a preset target impulse response signal;

[0007] Perform time-domain filtering on the test signal input to the speaker according to the time-domain equalization parameters of the filter;

[0008] Convert the time-domain filtered signal to the frequency domain to obtain an amplitude-frequency response signal;

[0009] Perform frequency-domain equalization on the converted amplitude-frequency response signal to obtain the frequency-domain equalization parameters of the filter;

[0010] Combine the frequency-domain equalization parameters of the filter and the time-domain equalization parameters of the filter to obtain the target parameters of the filter.

[0011] Optionally, the determining the time-domain equalization parameters of the filter according to the obtained impulse response signal and a preset target impulse response signal includes:

[0012] Determine the time-domain equalization parameters of the filter by minimizing the loss function according to the obtained impulse response signal and a preset target impulse response signal.

[0013] Optionally, performing frequency-domain equalization on the converted amplitude-frequency response signal to obtain the frequency-domain equalization parameter of the filter, including:

[0014] Determining a compensated amplitude-frequency response signal according to the converted amplitude-frequency response signal and the target amplitude-frequency response signal;

[0015] Converting the linear frequency scale of the converted amplitude-frequency response signal and the compensated amplitude-frequency response signal into the Bark scale, and solving the frequency-domain equalization parameter of the filter.

[0016] Optionally, determining the time-domain equalization parameter of the filter according to the obtained impulse response signal and the preset target impulse response signal, including:

[0017] Converting the obtained impulse response signal from a discrete form into a circulant determinant;

[0018] Determining the time-domain equalization parameter of the filter according to the circulant determinant of the obtained impulse response signal and the preset target impulse response signal.

[0019] Optionally, determining the time-domain equalization parameter of the filter according to the circulant determinant of the obtained impulse response signal and the preset target impulse response signal, including:

[0020] Determining the time-domain equalization parameter of the filter according to the following formula:

[0021] w2 = G H F H Δ H (ΔΔ H +G) -1 D + C H F H ΔΔ H (ΔΔ H +G) -1 [I - BΔ H (ΔΔ H +G) -1 Δ] -1 BΔ H (ΔΔ H +G) -1 D

[0022]

[0023] where w2 is the time-domain equalization parameter of the filter, G is a conversion matrix composed of an identity matrix and a zero matrix, F is a Fourier transform matrix, Δ is the diagonal matrix of , is the circulant determinant of the obtained impulse response signal, C is a regularization matrix, I is an identity matrix, B is the desired time-domain curve, (·) His the conjugate transpose, (·) -1 is the inverse operation.

[0024] Optionally, the filter includes a cascaded FIR filter and an IIR filter; determining the time-domain equalization parameters of the filter according to the acquired impulse response signal and a preset target impulse response signal includes: determining the time-domain equalization parameters of the FIR filter according to the acquired impulse response signal and the preset target impulse response signal;

[0025] Performing frequency-domain equalization on the converted amplitude-frequency response signal to obtain the frequency-domain equalization parameters of the filter includes: performing frequency-domain equalization on the amplitude-frequency response signal filtered according to the time-domain equalization parameters of the FIR filter to obtain the frequency-domain equalization parameters of the FIR filter and the frequency-domain equalization parameters of the IIR filter.

[0026] Optionally, performing frequency-domain equalization on the amplitude-frequency response signal filtered according to the time-domain equalization parameters of the FIR filter to obtain the frequency-domain equalization parameters of the FIR filter and the frequency-domain equalization parameters of the IIR filter includes:

[0027] Taking the mean square error as the criterion, performing iterative frequency-domain equalization on the amplitude-frequency response signal filtered according to the time-domain equalization parameters of the FIR filter until the amplitude in the first frequency band in the filtered amplitude-frequency response signal is less than a predetermined amplitude, and then stopping the iteration to obtain the frequency-domain equalization parameters of the FIR filter;

[0028] Taking the mean square error as the criterion, performing iterative frequency-domain equalization on the amplitude-frequency response signal filtered according to the frequency-domain equalization parameters of the FIR filter until the amplitude in the second frequency band in the filtered amplitude-frequency response signal is less than the predetermined amplitude, and then stopping the iteration to obtain the frequency-domain equalization parameters of the IIR filter;

[0029] Wherein, the frequency in the first frequency band is greater than a predetermined frequency threshold, and the frequency in the second frequency band is less than the frequency threshold.

[0030] According to a second aspect of the embodiments of the present disclosure, there is provided a digital equalization device for a loudspeaker, the device including:

[0031] An acquisition module, configured to acquire the impulse response signal output by the loudspeaker at a signal acquisition point;

[0032] A determination module, configured to determine the time-domain equalization parameters of the filter according to the acquired impulse response signal and a preset target impulse response signal;

[0033] A filtering module, configured to perform time-domain filtering on a test signal input to the loudspeaker according to the time-domain equalization parameters of the filter;

[0034] A conversion module, configured to convert the time-domain filtered signal into the frequency domain to obtain an amplitude-frequency response signal;

[0035] An equalization module, configured to perform frequency-domain equalization on the obtained amplitude-frequency response signal to obtain the frequency-domain equalization parameter of the filter;

[0036] A combination module, configured to combine the frequency-domain equalization parameter of the filter and the time-domain equalization parameter of the filter to obtain the target parameter of the filter.

[0037] According to a third aspect of the embodiments of the present disclosure, there is provided a digital equalization device for a speaker, including:

[0038] A processor;

[0039] A memory for storing processor-executable instructions;

[0040] Wherein, the processor is configured to execute the steps of the above method provided by the present disclosure.

[0041] According to a fourth aspect of the embodiments of the present disclosure, there is provided a vehicle, including a cockpit, a speaker, and a controller, where the controller is configured to execute the steps of the above method provided by the present disclosure to perform digital equalization on the speaker for a signal acquisition point in the cockpit.

[0042] According to a fifth aspect of the embodiments of the present disclosure, there is provided a computer-readable storage medium, on which computer program instructions are stored, and when the program instructions are executed by a processor, the steps of the above method provided by the present disclosure are implemented.

[0043] The technical solutions provided by the embodiments of the present disclosure may include the following beneficial effects:

[0044] Perform time-domain equalization on the speaker to obtain the time-domain equalization parameter of the filter. After filtering the test signal according to the time-domain equalization parameter of the filter, the filtered signal is then converted into the frequency domain, and equalization is performed in the frequency domain to obtain the frequency-domain equalization parameter of the filter. The frequency-domain equalization parameter and the time-domain equalization parameter are combined to obtain the target parameter of the filter. This solution solves the problem of poor sound quality of the speaker caused by only designing the filter for the frequency domain by solving the filter in two steps in the time domain and the frequency domain to complete the equalization of the speaker.

[0045] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The accompanying drawings herein are incorporated into the specification and constitute a part of the specification, showing embodiments consistent with the present disclosure, and are used together with the specification to explain the principles of the present disclosure.

[0047] Figure 1 It is a schematic diagram of the framework for speaker equalization provided by an exemplary embodiment.

[0048] Figure 2 It is a flowchart of a digital equalization method for a speaker shown according to an exemplary embodiment.

[0049] Figure 3 It is a schematic diagram of a compensated amplitude-frequency response signal curve provided by an exemplary embodiment.

[0050] Figure 4 It is an amplitude-frequency curve before and after filtering provided by an exemplary embodiment.

[0051] Figure 5 It is a time-domain curve before and after filtering provided by an exemplary embodiment.

[0052] Figure 6 It is a time-domain curve before and after filtering provided by an exemplary embodiment.

[0053] Figure 7 It is a schematic diagram of filter cascading provided by an exemplary embodiment.

[0054] Figure 8 It is a block diagram of a digital equalization device for a speaker shown according to an exemplary embodiment.

[0055] Figure 9 It is a block diagram of a vehicle shown according to an exemplary embodiment. Detailed implementation manners

[0056] Here, the exemplary embodiments will be described in detail, and the examples are shown in the drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementation manners described in the following exemplary embodiments do not represent all implementation manners consistent with the present disclosure. On the contrary, they are merely examples of devices and methods consistent with some aspects of the present disclosure as detailed in the appended claims.

[0057] It should be noted that all actions of obtaining signals, information, or data in this application are carried out on the premise of complying with the corresponding data protection regulations and policies of the country where it is located and obtaining authorization from the owner of the corresponding device.

[0058] In the cockpit of a vehicle, a room, or an architectural acoustic system, there can be multiple speakers. Figure 1It is a schematic diagram of the framework for speaker equalization provided by an exemplary embodiment. There are L speakers, and the control area is the cockpit, room, or building of a vehicle, and the control area contains M control points (control point 1 to control point M). The control point is the position point for signal acquisition, and a microphone can be placed there.

[0059] The input signal x(t) can be white noise or a swept frequency. Taking the FIR filter as an example, the coefficient w(t) is the part that needs to be designed / solved. The acoustic transfer function from the speaker to the control point is a known condition and can be collected by the microphone set at the control point. In this framework, the input signal of each speaker is filtered through the filter and then re-sampled at the control point. If the obtained result is consistent with the expected curve, the designed filter algorithm is considered effective. Figure 1 In it, w i (t) represents the coefficient of the i-th filter. Among them, g ij (t) represents the transfer function from the i-th speaker to the j-th control point.

[0060] Taking the cockpit of a vehicle as an example, the design of the filter can be carried out through the following steps:

[0061] 1. Measure the transfer function from each speaker to each control point (through the microphone).

[0062] 2. For the current speaker, select the "sweet spot" among all M control points. The "sweet spot" is the point in space that can best represent the acoustic characteristics. For example, if the currently equalized speaker is the first speaker and control point 1 is selected for calculation, the transfer function involved in the calculation is g 11 (t).

[0063] 3. Design the target impulse response at the control point. Here, the delta function can be used as the target impulse response.

[0064] 4. Design the filter to optimize the time-domain impulse response and amplitude-frequency response of the signal to obtain the coefficients of, for example, the FIR filter.

[0065] 5. For the next speaker, loop through steps 2 to 4 until the equalization of all speakers is completed.

[0066] In the related art, usually only the optimization of the frequency-domain curve is considered, while ignoring the influence of the time-domain transient response on the listening experience. In this solution, as shown in step 4 above, both the time-domain impulse response and the amplitude-frequency response are optimized to design the filter. The following takes the vehicle cockpit as an example for specific description.

[0067] Figure 2 It is a flowchart of a digital equalization method for a speaker shown according to an exemplary embodiment. As Figure 2 shown, the method includes the following steps.

[0068] In step S11, a pulse response signal output by a speaker is acquired at a signal acquisition point.

[0069] The signal acquisition point is the above-mentioned control point. The signal acquisition point can be selected as the point closest to the headrest position or the point at the center of the cockpit. The points close to the headrest position are close to the human ear. Using these points as signal acquisition points for equalization can achieve better equalization effects. Selecting the point at the center of the cockpit takes into account the passengers in each seat.

[0070] In step S12, according to the acquired pulse response signal and a preset target pulse response signal, time-domain equalization parameters of a filter are determined.

[0071] The target pulse response signal can adopt a signal of a delta function. When performing time-domain equalization, the target pulse response signal can be used as an optimization target. For example, optimization criteria such as Least Squares (LS) estimation or Minimum Mean Squared Error (MMSE) are adopted to determine the time-domain equalization parameters of the filter. Among them, the equalization parameters of a Finite Impulse Response (FIR) filter can include filter coefficients, and the equalization parameters of an Infinite Impulse Response (IIR) filter can include center frequency, filter gain, Q value, etc.

[0072] In step S13, a test signal input to the speaker is filtered in the time domain according to the time-domain equalization parameters of the filter.

[0073] In step S14, the signal filtered in the time domain is converted to the frequency domain to obtain an amplitude-frequency response signal.

[0074] The Fourier transform can be used to convert the signal filtered in the time domain to the frequency domain.

[0075] In step S15, frequency-domain equalization is performed on the obtained amplitude-frequency response signal to obtain frequency-domain equalization parameters of the filter.

[0076] When performing frequency-domain equalization, the target amplitude-frequency response signal can be used as an optimization target, and LS or MMSE is adopted for optimization to obtain the frequency-domain equalization parameters of the filter.

[0077] In step S16, the frequency-domain equalization parameters of the filter and the time-domain equalization parameters of the filter are combined to obtain target parameters of the filter.

[0078] This solution solves the problem of poor sound quality of speakers caused by only designing filters in the frequency domain in the related art by solving the filter in two steps in the time domain and the frequency domain to complete the equalization of the speaker.

[0079] In another embodiment, according to the obtained impulse response signal and the preset target impulse response signal, the time-domain equalization parameters of the filter are determined, including:

[0080] According to the obtained impulse response signal and the preset target impulse response signal, the time-domain equalization parameters of the filter are determined by minimizing the loss function.

[0081] The loss function in the related art can be used to optimize with the target impulse response signal as the target. Taking the FIR filter as an example, the following loss function can be used for solution:

[0082]

[0083] where g is the transfer function from the selected speaker to the signal acquisition point, w is the time-domain equalization coefficient of the FIR filter to be solved, d is the target impulse response signal, |·| represents taking the absolute value, represents the sum of squares of the two-norm, represents w corresponding to the minimum value.

[0084] In this embodiment, the time-domain equalization parameters of the filter are solved by minimizing the loss function, and the method is simple and the optimization effect is good.

[0085] In another embodiment, frequency-domain equalization is performed on the converted amplitude-frequency response signal to obtain the frequency-domain equalization parameters of the filter, including:

[0086] According to the converted amplitude-frequency response signal and the target amplitude-frequency response signal, the compensated amplitude-frequency response signal is determined;

[0087] The linear frequency scale of the converted amplitude-frequency response signal and the compensated amplitude-frequency response signal is converted to the Bark scale to solve the frequency-domain equalization parameters of the filter.

[0088] Among them, the target amplitude-frequency response signal can be a signal with a flat amplitude response curve, and its characteristic is that the amplitude is stable and unchanged at the expected gain level in the frequency domain. In the graph of the amplitude spectrum, the superposition of the curve of the converted amplitude-frequency response signal and the curve of the compensated amplitude-frequency response signal can be a flat curve of the target amplitude-frequency response signal. That is, when solving the frequency-domain equalization parameters of the filter subsequently, the compensated amplitude-frequency response signal can be used as the target. In this way, the equalization effect of the filter in both the time domain and the frequency domain is optimized with the target amplitude-frequency response signal as the target.

[0089] Figure 3It is a schematic diagram of a compensated amplitude-frequency response signal curve provided by an exemplary embodiment. As Figure 3 shown, the thick curve represents the converted amplitude-frequency response signal curve, the thin curve represents the compensated amplitude-frequency response signal curve, and the flat line with zero amplitude represents the target amplitude-frequency response signal curve. The target amplitude-frequency response signal curve is the superposition of the converted amplitude-frequency response signal curve and the compensated amplitude-frequency response signal curve.

[0090] For example, a linear-phase FIR design method can be adopted, and the solution is obtained by minimizing the weighted integrated square error between the piecewise linear function and the amplitude-frequency response of the filter in the desired frequency band. Specifically, this method focuses on achieving the optimal frequency response in a specific frequency band.

[0091] Among them, the "piecewise linear function" can represent the ideal amplitude-frequency response target, which can be determined by the application scenario. For example, in a vehicle system, a very flat and rapidly decaying frequency characteristic. "Piecewise" means segmenting the frequency, and the piecewise linear function is a function of the frequency varying with time in different intervals. The core of the design is to be as close as possible to this ideal response.

[0092] The "amplitude-frequency response of the filter in the desired frequency band" is the amplitude-frequency response of the designed filter, which reflects the amplification or attenuation ability of the FIR filter at each frequency point.

[0093] The "weighted integrated square error" is a method for measuring the difference between the actual amplitude-frequency response and the ideal amplitude-frequency response. Specifically, weights can be assigned to different frequency bands. For the actual amplitude-frequency response and the ideal amplitude-frequency response respectively, the amplitudes of each frequency band are weighted and integrated, and the square of the difference between the weighted integrated result of the actual amplitude-frequency response and the weighted integrated result of the ideal amplitude-frequency response is calculated. Through weighted integration, different weights can be given according to the importance of different frequency regions. For example, in terms of the sense of hearing, low-frequency deviation may be more important than high-frequency deviation.

[0094] Finally, for the part of "obtaining the solution", an optimization algorithm can be used to find the best filter coefficients. It can be implemented by numerical methods (such as the gradient descent method), and the purpose is to find a set of filter coefficients that minimize the weighted integrated square error.

[0095] This FIR filter design method uses mathematical optimization techniques to determine a set of filter coefficients to be as close as possible to the ideal frequency response in a specific frequency band, while also considering actual design limitations and application requirements.

[0096] The human ear's perception of sound (such as frequency, pitch) is non-linear. The Bark scale maps the physical frequency to multiple critical bands in psychoacoustics in units of Hz.

[0097] Specifically, converting the linear frequency scale to the Bark scale can be achieved by, according to the Bark frequency band table, finding the center frequencies corresponding to the optimal working frequency bands of the speaker in the obtained amplitude-frequency response signal and the compensated amplitude-frequency response signal, and using the found center frequencies to replace the corresponding frequency intervals for solution.

[0098] Speakers can be divided into woofers, midrange speakers, and tweeters. The optimal working frequency band of a woofer is 40 - 200 Hz, the optimal working frequency band of a midrange speaker is 300 - 2000 Hz, and the optimal working frequency band of a tweeter is above 3000 Hz.

[0099] For example, in the Bark frequency band table, the center frequency corresponding to the frequency band interval of 3150 - 3700 Hz is 3700 Hz, and the center frequency corresponding to the frequency band interval of 3700 - 4400 Hz is 4000 Hz. If the optimal working frequency band of the speaker is above 3000 Hz, then 3700 Hz and 4000 Hz can be used to replace the frequency band intervals of 3150 - 3700 Hz and 3700 - 4400 Hz respectively, and so on.

[0100] In this embodiment, in the frequency domain design, mapping the physical frequency to the psychoacoustic frequency scale for equalization and designing at key frequencies according to the auditory characteristics of the human ear not only optimize the amplitude-frequency curve but also improve the sound quality performance.

[0101] Figure 4 is the amplitude-frequency curve graph before and after filtering provided by an exemplary embodiment. As Figure 4 shown, the solid line is the amplitude-frequency curve before filtering, and the dashed line is the amplitude-frequency curve after filtering. It can be seen that the speaker remains flat in the desired frequency band (above 3000 Hz), and the equalization effect is good.

[0102] Figure 5 is the time-domain curve graph before and after filtering provided by an exemplary embodiment. As Figure 5 shown, the lower curve is the time-domain curve before filtering, and the upper curve is the time-domain curve after filtering. It can be seen that the tail after the main peak (the peaks and valleys of the lower curve at points 50 - 70) is significantly suppressed, and according to the feedback of the listening testers, this equalization has achieved obvious effects (the human voice is clearer and the instrument separation is high).

[0103] In another embodiment, according to the obtained impulse response signal and the preset target impulse response signal, determining the time-domain equalization parameters of the filter, including:

[0104] Converting the obtained impulse response signal from the discrete form to a circulant determinant;

[0105] Determine the time-domain equalization parameters of the filter according to the circulant determinant of the obtained impulse response signal and the preset target impulse response signal.

[0106] Generally, the complexity of matrix inversion is cubic. For example, the complexity of inverting a 100-dimensional matrix is 10 to the power of 6, which is a very large amount of computation. In this embodiment, the impulse response signal is transformed into a circulant determinant, that is, transformed into a combination form of periodic solutions and non-periodic solutions, which avoids the matrix inversion operation while accelerating the operation speed.

[0107] Specifically, the impulse response signal is transformed from a discrete form into a circulant determinant. Its length is La. Denote the length of the solved filter as Lb, then the total length L = La + Lb. The general form of this solution is:

[0108]

[0109] where w2 is the time-domain equalization parameter of the filter. is the circulant determinant of the obtained impulse response signal, C is the regularization matrix, B is the target impulse response signal curve, for example, it can be a delta function, (·) H is the conjugate transpose, (·) -1 is the inversion.

[0110] The Fourier transform form of is F is the Fourier transform matrix, Δ is the diagonal matrix of. In order to obtain two solutions of periodic solutions and non-periodic solutions, can be transformed into the form of, D is the periodic part, and E is the non-periodic part. Since D is a Toeplitz matrix, then G is the transformation matrix composed of the identity matrix and the zero matrix. Further, the form of the solution can be written as:

[0111]

[0112] where d is the target impulse response, which can be a delta function.

[0113] Furthermore, according to the Woodbury matrix identity and the properties of the matrix (which are expressed as L - Lb residual points caused by periodic extension), the solution can be written as:

[0114] w2 = G H F H Δ H (ΔΔ H + G) -1 D + CH F H ΔΔ H (ΔΔ H +G) -1 [I - BΔ H (ΔΔ H +G) -1 Δ] -1 BΔ H (ΔΔ H +G) -1 D

[0115]

[0116] Among them, I is the identity matrix.

[0117] Therefore, according to the cyclic determinant of the obtained impulse response signal and the target impulse response signal, the time-domain equalization parameters of the filter are determined, including: determining the time-domain equalization parameters of the filter according to formula (3).

[0118] Compared with the solution using the least squares method (involving a large number of matrix inversions) in the related art, the method of this embodiment consumes less resources, has less operation time, is solved more quickly, avoids the computing power requirements brought by matrix inversion, and optimizes the time-domain transient characteristics of the speaker, improving the sound quality performance.

[0119] Figure 6 is the time-domain curve graph before and after filtering provided by an exemplary embodiment. As Figure 6 shown, the following curve is the time-domain curve before filtering, and the upper curve is the time-domain curve after filtering. It can be seen that the reflection part is greatly suppressed, achieving the expected goal.

[0120] The phase influence of the FIR filtering scheme is very small, but a relatively high order (usually 512 orders, 1024 orders) is often required to complete the speaker equalization. If FIR filters are used for all speakers, the computing power requirement for the chip will be very high. Due to the relatively large phase influence of the IIR filter itself, the time-domain performance of the speaker is poor, resulting in poor sound quality.

[0121] In view of this, in another embodiment, the filter includes a cascaded FIR filter and an IIR filter. For example, a 100-order FIR filter is cascaded with 15 IIR filters.

[0122] Figure 7 is the schematic diagram of the filter cascade provided by an exemplary embodiment. Figure 7Among them, the optimal control point is the "sweet spot", and the time-domain desired curve is the target impulse response signal curve. Among the designed speaker equalization schemes, the input signal is filtered by a 100th-order FIR filter and 15 IIR filters (IIR_1 to IIR_15) in sequence and then input to the speaker.

[0123] In this embodiment, according to the obtained impulse response signal and the preset target impulse response signal, the time-domain equalization parameters of the filter are determined, including: determining the time-domain equalization parameters of the FIR filter according to the obtained impulse response signal and the preset target impulse response signal.

[0124] Frequency-domain equalization is performed on the converted amplitude-frequency response signal to obtain the frequency-domain equalization parameters of the filter, including: performing frequency-domain equalization on the amplitude-frequency response signal filtered according to the time-domain equalization parameters of the FIR filter to obtain the frequency-domain equalization parameters of the FIR filter and the frequency-domain equalization parameters of the IIR filter.

[0125] That is, first design the time-domain equalization parameters of the FIR filter to approximate the target impulse response signal (the specific method for solving the time-domain equalization parameters can refer to the foregoing method), and then perform frequency-domain equalization on the FIR filter and the IIR filter respectively to design the frequency-domain equalization parameters of the FIR filter and the frequency-domain equalization parameters of the IIR filter. When designing the frequency-domain equalization parameters, the amplitude-frequency response signal filtered according to the time-domain equalization parameters of the FIR filter is used.

[0126] In this embodiment, compared with simply using the FIR filter, the computing power is reduced (a smaller order can be used), and compared with simply using the IIR filter, the sound quality performance is improved, balancing the characteristics of both.

[0127] In another embodiment, frequency-domain equalization is performed on the amplitude-frequency response signal filtered according to the time-domain equalization parameters of the FIR filter to obtain the frequency-domain equalization parameters of the FIR filter and the frequency-domain equalization parameters of the IIR filter, including:

[0128] Taking the mean square error as the criterion, perform iterative frequency-domain equalization on the amplitude-frequency response signal filtered according to the time-domain equalization parameters of the FIR filter until the amplitude in the first frequency band in the filtered amplitude-frequency response signal is less than the predetermined amplitude, and then stop the iteration to obtain the frequency-domain equalization parameters of the FIR filter;

[0129] Taking the mean square error as the criterion, perform iterative frequency-domain equalization on the amplitude-frequency response signal filtered according to the frequency-domain equalization parameters of the FIR filter until the amplitude in the second frequency band in the filtered amplitude-frequency response signal is less than the predetermined amplitude, and then stop the iteration to obtain the frequency-domain equalization parameters of the IIR filter;

[0130] Among them, the frequencies in the first frequency band are greater than a predetermined frequency threshold, and the frequencies in the second frequency band are less than the frequency threshold. The predetermined frequency threshold can be 3000 Hz. The first frequency band is the mid-high frequency band, and the second frequency band is the low frequency band.

[0131] The IIR filter iteration can be specifically as follows: Each IIR filter has a randomly initialized value, and then randomly updates the cut-off frequency, passband gain, stopband attenuation, and Q value to recalculate the amplitude-frequency response signal. Compare the recalculated amplitude-frequency response signal with the target amplitude-frequency response. If the gap is smaller than the previous update, continue to update the cut-off frequency, passband gain, stopband attenuation, and Q value based on the current IIR parameters. And so on, perform multiple iterations until the difference between the amplitude-frequency response signal and the target amplitude-frequency response signal is stable (no longer getting smaller) or the difference is less than a predetermined value, that is, the difference is zero.

[0132] In this embodiment, the FIR filter form with less influence on phase is adopted in the mid-high frequency band to ensure that the high-frequency performance of the speaker is "clear and clean" enough, and the IIR filter is used to complete the equalization in the low-frequency band where the human ear is less sensitive to phase. This can effectively save computing power and is very suitable for the in-vehicle system with tight computing power.

[0133] Based on the same inventive concept, the present disclosure also provides a digital equalization device for a speaker. Figure 8 is a block diagram of a digital equalization device for a speaker shown according to an exemplary embodiment. Refer to Figure 8 As shown in, the digital equalization device 800 for the speaker includes an acquisition module 801, a determination module 802, a filtering module 803, a conversion module 804, an equalization module 805, and a combination module 806.

[0134] The acquisition module 801 is configured to acquire the impulse response signal output by the speaker at the signal acquisition point.

[0135] The determination module 802 is configured to determine the time-domain equalization parameters of the filter according to the acquired impulse response signal and a preset target impulse response signal.

[0136] The filtering module 803 is configured to perform time-domain filtering on the test signal input to the speaker according to the time-domain equalization parameters of the filter.

[0137] The conversion module 804 is configured to convert the time-domain filtered signal to the frequency domain to obtain the amplitude-frequency response signal.

[0138] The equalization module 805 is configured to perform frequency-domain equalization on the converted amplitude-frequency response signal to obtain the frequency-domain equalization parameters of the filter.

[0139] The combination module 806 is configured to combine the frequency-domain equalization parameters of the filter and the time-domain equalization parameters of the filter to obtain the target parameters of the filter.

[0140] Optionally, the determination module is configured to: determine the time-domain equalization parameters of the filter by minimizing the loss function according to the acquired impulse response signal and the preset target impulse response signal.

[0141] Optionally, the equalization module is configured to: determine the compensated amplitude-frequency response signal according to the converted amplitude-frequency response signal and the target amplitude-frequency response signal; convert the linear frequency scale of the converted amplitude-frequency response signal and the compensated amplitude-frequency response signal into the Bark scale, and solve the frequency-domain equalization parameters of the filter.

[0142] Optionally, the determination module is configured to:

[0143] Convert the acquired impulse response signal from a discrete form into a circulant determinant;

[0144] Determine the time-domain equalization parameters of the filter according to the circulant determinant of the acquired impulse response signal and the preset target impulse response signal.

[0145] Optionally, the determination module is configured to: determine the time-domain equalization parameters of the filter according to the following formula:

[0146] w2 = G H F H Δ H (ΔΔ H + G) -1 D + C H F H ΔΔ H (ΔΔ H + G) -1 [I - BΔ H (ΔΔ H + G) -1 Δ] -1 BΔ H (ΔΔ H + G) -1 D

[0147]

[0148] where w2 is the time-domain equalization parameter of the filter, G is a transformation matrix composed of an identity matrix and a zero matrix, F is a Fourier transform matrix, Δ is the diagonal matrix of , is the circulant determinant of the acquired impulse response signal, C is a regularization matrix, I is an identity matrix, B is the desired time-domain curve, (·) H is the conjugate transpose, and (·) -1 is the inverse.

[0149] Optionally, the filter includes a cascaded FIR filter and an IIR filter.

[0150] The determination module is configured to: determine the time-domain equalization parameters of the FIR filter according to the acquired impulse response signal and a preset target impulse response signal;

[0151] The equalization module is configured to: perform frequency-domain equalization on the amplitude-frequency response signal filtered according to the time-domain equalization parameters of the FIR filter to obtain the frequency-domain equalization parameters of the FIR filter and the frequency-domain equalization parameters of the IIR filter.

[0152] Optionally, the equalization module is configured to:

[0153] Taking the mean square error as a criterion, perform iterations of frequency-domain equalization on the amplitude-frequency response signal filtered according to the time-domain equalization parameters of the FIR filter until the amplitude in the first frequency band in the filtered amplitude-frequency response signal is less than a predetermined amplitude, and then stop the iteration to obtain the frequency-domain equalization parameters of the FIR filter;

[0154] Taking the mean square error as a criterion, perform iterations of frequency-domain equalization on the amplitude-frequency response signal filtered according to the frequency-domain equalization parameters of the FIR filter until the amplitude in the second frequency band in the filtered amplitude-frequency response signal is less than a predetermined amplitude, and then stop the iteration to obtain the frequency-domain equalization parameters of the IIR filter;

[0155] Wherein, the frequency in the first frequency band is greater than a predetermined frequency threshold, and the frequency in the second frequency band is less than the frequency threshold.

[0156] Regarding the device in the above embodiments, the specific manners in which each module performs operations have been described in detail in the embodiments related to the method, and will not be elaborated here.

[0157] This solution performs time-domain equalization on the speaker to obtain the time-domain equalization parameters of the filter. After filtering the test signal according to the time-domain equalization parameters of the filter, the filtered signal is then converted to the frequency domain, and equalization is performed in the frequency domain to obtain the frequency-domain equalization parameters of the filter. The frequency-domain equalization parameters and the time-domain equalization parameters are combined to obtain the target parameters of the filter. This solution solves the problem of poor sound quality of the speaker caused by only designing the filter for the frequency domain in the related art by solving the filter in two steps in the time domain and the frequency domain to complete the equalization of the speaker.

[0158] Figure 9 It is a block diagram of a vehicle 600 shown according to an exemplary embodiment. For example, the vehicle 600 can be a hybrid vehicle, or a non-hybrid vehicle, an electric vehicle, a fuel cell vehicle, or other types of vehicles. The vehicle 600 can be an autonomous vehicle, a semi-autonomous vehicle, or a non-autonomous vehicle.

[0159] Reference Figure 9 Vehicle 600 may include various subsystems. For example, an infotainment system 610, a perception system 620, a decision control system 630, a drive system 640, and a computing platform 650. Among them, vehicle 600 may also include more or fewer subsystems, and each subsystem may include multiple components. In addition, each subsystem and each component of vehicle 600 may be interconnected by wired or wireless means.

[0160] In some embodiments, the infotainment system 610 may include a communication system, an entertainment system, a navigation system, etc.

[0161] The perception system 620 may include several sensors for sensing information about the environment around vehicle 600. For example, the perception system 620 may include a global positioning system (the global positioning system may be a GPS system, a Beidou system, or other positioning systems), an inertial measurement unit (IMU), lidar, millimeter-wave radar, ultrasonic radar, and a camera device.

[0162] The decision control system 630 may include a computing system, a vehicle controller, a steering system, an accelerator, and a braking system.

[0163] The drive system 640 may include components that provide powered movement for vehicle 600. In one embodiment, the drive system 640 may include an engine, an energy source, a powertrain, and wheels. The engine may be one or a combination of an internal combustion engine, an electric motor, and an air compression engine. The engine can convert the energy provided by the energy source into mechanical energy.

[0164] Some or all functions of vehicle 600 are controlled by the computing platform 650. The computing platform 650 may include at least one processor 651 and a memory 652. The processor 651 may execute instructions 653 stored in the memory 652.

[0165] The processor 651 may be any conventional processor, such as a commercially available CPU. The processor may also include, for example, a Graphic Process Unit (GPU), a Field Programmable Gate Array (FPGA), a System on Chip (SOC), an Application Specific Integrated Circuit (ASIC), or a combination thereof.

[0166] The memory 652 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0167] In addition to the instructions 653, the memory 652 can also store data, such as road maps, route information, data on the position, direction, speed, etc. of the vehicle. The data stored in the memory 652 can be used by the computing platform 650.

[0168] In an embodiment of the present disclosure, the processor 651 can execute the instructions 653 to complete all or part of the steps of the above-mentioned digital equalization method for the speaker.

[0169] The present disclosure also provides a digital equalization device for a speaker, including: a processor and a memory for storing processor-executable instructions. Among them, the processor is configured to execute the steps of the above method provided by the present disclosure.

[0170] The present disclosure also provides a vehicle, including a cockpit, a speaker, and a controller. The controller is used to execute the steps of the above method provided by the present disclosure to perform digital equalization on the speaker for a signal acquisition point in the cockpit.

[0171] The present disclosure also provides a computer-readable storage medium, on which computer program instructions are stored. The program instructions are executed by the processor to perform the steps of the above method provided by the present disclosure.

[0172] The present disclosure also provides a chip, including a processor and an interface; the processor is used to read instructions to execute the above method provided by the present disclosure.

[0173] In another exemplary embodiment, a computer program product is also provided. The computer program product includes a computer program that can be executed by a programmable device. The computer program has a code portion for executing the above-mentioned digital equalization method for the speaker when executed by the programmable device.

[0174] Those skilled in the art will readily think of other implementation manners of the present disclosure after considering the specification and practicing the present disclosure. This application is intended to cover any variations, uses, or adaptations of the present disclosure, which follow the general principles of the present disclosure and include well-known common knowledge or conventional technical means in the technical field not disclosed in the present disclosure. The specification and embodiments are only regarded as exemplary, and the true scope and spirit of the present disclosure are pointed out by the following claims.

[0175] It should be understood that the present disclosure is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present disclosure is only limited by the appended claims.

Claims

1. A digital equalization method for a speaker, characterized in that, The method includes: Obtaining the impulse response signal output by the loudspeaker at the signal acquisition point; Determining the time-domain equalization parameters of the filter according to the obtained impulse response signal and a preset target impulse response signal; Performing time-domain filtering on the test signal input to the loudspeaker according to the time-domain equalization parameters of the filter; Converting the time-domain filtered signal to the frequency domain to obtain an amplitude-frequency response signal; Performing frequency-domain equalization on the obtained amplitude-frequency response signal to obtain the frequency-domain equalization parameters of the filter; Combining the frequency-domain equalization parameters of the filter and the time-domain equalization parameters of the filter to obtain the target parameters of the filter.

2. The method according to claim 1, wherein The determining the time-domain equalization parameters of the filter according to the obtained impulse response signal and a preset target impulse response signal includes: Determining the time-domain equalization parameters of the filter by minimizing the loss function according to the obtained impulse response signal and a preset target impulse response signal.

3. The method according to claim 1, characterized in that The performing frequency-domain equalization on the obtained amplitude-frequency response signal to obtain the frequency-domain equalization parameters of the filter includes: Determining a compensated amplitude-frequency response signal according to the obtained amplitude-frequency response signal and a target amplitude-frequency response signal; Converting the linear frequency scales of the obtained amplitude-frequency response signal and the compensated amplitude-frequency response signal to the Bark scale, and solving the frequency-domain equalization parameters of the filter.

4. The method according to claim 1, characterized in that, The determining the time-domain equalization parameters of the filter according to the obtained impulse response signal and a preset target impulse response signal includes: Converting the obtained impulse response signal from a discrete form to a circulant determinant; Determining the time-domain equalization parameters of the filter according to the circulant determinant of the obtained impulse response signal and a preset target impulse response signal.

5. The method according to claim 4, wherein The determining the time-domain equalization parameters of the filter according to the circulant determinant of the obtained impulse response signal and a preset target impulse response signal includes: Determining the time-domain equalization parameters of the filter according to the following formula: w2 = G H F H Δ H (ΔΔ H + G) -1 D + C H F H ΔΔ H (ΔΔ H + G) -1 [I - BΔ H (ΔΔ H + G) -1 Δ] -1 BΔ H (ΔΔ H + G) -1 D wherein, w2 is the time-domain equalization parameter of the filter, G is a conversion matrix composed of an identity matrix and a zero matrix, F is a Fourier transform matrix, Δ is the diagonal matrix of , is the circulant determinant of the obtained impulse response signal, C is a regularization matrix, I is an identity matrix, B is the desired time-domain curve, (·) H is the conjugate transpose, (·) -1 is the inverse operation.

6. The method according to claim 1, characterized in that, The filter includes a cascaded FIR filter and an IIR filter; The determining the time-domain equalization parameters of the filter according to the obtained impulse response signal and a preset target impulse response signal includes: determining the time-domain equalization parameters of the FIR filter according to the obtained impulse response signal and a preset target impulse response signal; The performing frequency-domain equalization on the obtained amplitude-frequency response signal to obtain the frequency-domain equalization parameters of the filter includes: performing frequency-domain equalization on the amplitude-frequency response signal filtered according to the time-domain equalization parameters of the FIR filter to obtain the frequency-domain equalization parameters of the FIR filter and the frequency-domain equalization parameters of the IIR filter.

7. The method according to claim 6, wherein The performing frequency-domain equalization on the amplitude-frequency response signal filtered according to the time-domain equalization parameters of the FIR filter to obtain the frequency-domain equalization parameters of the FIR filter and the frequency-domain equalization parameters of the IIR filter includes: Taking the mean square error as a criterion, performing iteration of frequency-domain equalization on the amplitude-frequency response signal filtered according to the time-domain equalization parameters of the FIR filter until the amplitude in the first frequency band in the filtered amplitude-frequency response signal is less than a predetermined amplitude, and then stopping the iteration to obtain the frequency-domain equalization parameters of the FIR filter; Taking the mean square error as the criterion, perform iteration of frequency-domain equalization on the amplitude-frequency response signal filtered according to the frequency-domain equalization parameters of the FIR filter until the amplitude in the second frequency band in the filtered amplitude-frequency response signal is less than the predetermined amplitude, and then stop the iteration to obtain the frequency-domain equalization parameters of the IIR filter; Wherein, the frequency in the first frequency band is greater than a predetermined frequency threshold, and the frequency in the second frequency band is less than the frequency threshold.

8. A digital equalization device for a loudspeaker, characterized in that, The device includes: An acquisition module, configured to acquire the impulse response signal output by the speaker at the signal acquisition point; A determination module, configured to determine the time-domain equalization parameters of the filter according to the acquired impulse response signal and a preset target impulse response signal; A filtering module, configured to perform time-domain filtering on the test signal input to the speaker according to the time-domain equalization parameters of the filter; A conversion module, configured to convert the time-domain filtered signal to the frequency domain to obtain an amplitude-frequency response signal; An equalization module, configured to perform frequency-domain equalization on the obtained amplitude-frequency response signal to obtain the frequency-domain equalization parameters of the filter; A combination module, configured to combine the frequency-domain equalization parameters of the filter and the time-domain equalization parameters of the filter to obtain the target parameters of the filter.

9. A digital equalization device for a loudspeaker, characterized in that, Includes: A processor; A memory for storing instructions executable by the processor; Wherein, the processor is configured to execute the steps of the method according to any one of claims 1 to 7.

10. A vehicle, characterized in that, Includes a cockpit, a speaker, and a controller, and the controller is configured to execute the steps of the method according to any one of claims 1 to 7 to perform digital equalization on the speaker for the signal acquisition point in the cockpit.

11. A computer-readable storage medium having computer program instructions stored thereon, characterized in that, When the program instruction is executed by the processor, it implements the steps of the method according to any one of claims 1 to 7.

Images