Stereo balance adjustment method and device

Abstract

The embodiments of the present disclosure relate to a stereo balance adjustment method and a stereo balance adjustment device. The method includes: obtaining a stereo frequency response curve as a frequency response curve to be calibrated; performing differential processing on the frequency response curve to be calibrated and a preset target frequency response curve to obtain a frequency response difference curve; performing threshold discrimination on the frequency response difference curve in multiple frequency regions according to a preset threshold, and obtaining a frequency region where the frequency response difference curve exceeds the threshold the most as an error region; using a prescribed algorithm to perform calibration fitting on the frequency response difference curve in the error region, and outputting a calibrated frequency response difference curve; and merging the calibrated frequency response difference curve and the frequency response curve to be calibrated to obtain a new frequency response difference curve. According to the embodiments of the present disclosure, the algorithm can be simplified and the success rate of algorithm optimization can be improved.

Description

Technical Field

[0001] The present invention relates to a stereo balance adjustment method, and in particular to a stereo balance adjustment method and a stereo balance adjustment device for smart glasses or AR glasses. Background Art

[0002] In smart glasses or AR glasses, the sound-generating components are usually located behind or close to the ears, achieving a stereo effect by emitting sound from the left and right sides. Summary of the Invention

[0003] In view of the above problems, the embodiments of the present disclosure aim to provide a stereo balance adjustment method and a stereo balance adjustment device capable of optimizing the stereo balance of smart glasses.

[0004] In one aspect of the present disclosure, a method for adjusting stereo equalization is provided, comprising:

[0005] a curve acquisition step of acquiring a stereo frequency response curve as the frequency response curve to be calibrated;

[0006] a differential processing step of performing differential processing on the frequency response curve to be calibrated and a preset target frequency response curve to obtain a frequency response differential curve;

[0007] a threshold determination step of performing threshold determination on the frequency response difference curve in a plurality of frequency regions according to a preset threshold, and obtaining a frequency region in which the frequency response difference curve exceeds the threshold the most as an error region;

[0008] a calibration fitting step, performing calibration fitting on the frequency response difference curve located in the error region using a prescribed algorithm, and outputting a calibrated frequency response difference curve; and

[0009] The curve merging step is to merge the calibrated frequency response difference curve and the frequency response curve to be calibrated to obtain a new frequency response difference curve.

[0010] Optionally, the threshold determination step includes:

[0011] Pre-set sensitivity threshold;

[0012] Divide into multiple frequency regions according to frequency size;

[0013] performing judgment on the frequency response difference curve according to the threshold; and

[0014] Among the multiple frequency regions, the region where the frequency response difference curve exceeds the threshold the most is used as the error region.

[0015] Optionally, the division into multiple frequency regions according to frequency magnitude is performed in a manner such that the frequency response difference curve in each frequency region is located above the 0 axis or below the 0 axis.

[0016] Optionally, the stereo balance adjustment method further includes:

[0017] The new frequency response difference curve obtained in the curve merging step is re-input into the threshold determination step and the threshold determination step, the calibration step and the curve merging step are repeated until the new frequency response difference curve obtained in the curve merging step does not exceed the threshold.

[0018] Optionally, a plurality of filters are preset, and each time the calibration fitting step is performed, one filter is used to perform calibration fitting on the frequency response difference curve located in the error region.

[0019] Optionally, in the calibration fitting step, any one of the following algorithms is used for the frequency response difference curve located in the error region:

[0020] Genetic algorithm, simulated annealing algorithm.

[0021] Optionally, in the calibration fitting step, an EQ filter is used to perform calibration fitting on the frequency response difference curve located in the error region.

[0022] Optionally, the calibration fitting step includes:

[0023] For the frequency response difference curve located in the error region, defining the maximum distance that the frequency response difference curve deviates from the threshold by using a cost function; and

[0024] When the maximum distance is greater than 0, the EQ filter is used to perform calibration fitting on the frequency response difference curve of the error region.

[0025] Optionally, in the curve merging step, merging the calibrated frequency response difference curve and the frequency response curve to be calibrated is performed by adding the calibrated frequency response difference curve and the frequency response curve to be calibrated according to frequency points.

[0026] Optionally, in the curve merging step, filter parameters of the EQ filter are outputted simultaneously with the new frequency response difference curve.

[0027] In another aspect of the present disclosure, a balance adjustment device is provided, comprising:

[0028] A curve acquisition module, used to acquire a stereo frequency response curve as a frequency response curve to be calibrated;

[0029] A differential processing module is used to perform differential processing on the frequency response curve to be calibrated and a preset target frequency response curve to obtain a frequency response differential curve;

[0030] a threshold determination module, configured to perform threshold determination on the frequency response difference curve in a plurality of frequency regions according to a preset threshold, and obtain a frequency region in which the frequency response difference curve exceeds the threshold the most as an error region;

[0031] a calibration fitting module, which performs calibration fitting on the frequency response difference curve located in the error region using a prescribed algorithm and outputs a calibrated frequency response difference curve; and

[0032] The curve merging module is used to merge the calibrated frequency response difference curve and the frequency response curve to be calibrated to obtain a new frequency response difference curve.

[0033] Optionally, a sensitivity threshold is pre-set in the threshold judgment module, and the frequency region is divided into multiple frequency regions according to the frequency magnitude. The frequency response difference curve is subjected to threshold judgment according to the threshold. Among the multiple frequency regions, the region in which the frequency response difference curve exceeds the threshold the most is used as the error region.

[0034] Optionally, the new frequency response difference curve obtained in the curve merging module is input into the threshold determination module as the frequency response difference curve, and the threshold determination module, the calibration module, and the curve merging module repeatedly perform actions in sequence until the new frequency response difference curve obtained by the curve merging module does not exceed the threshold.

[0035] Optionally, in the calibration fitting module, a plurality of filters are pre-set, and one filter is used each time the frequency response difference curve is calibrated and fitted.

[0036] Optionally, in the calibration fitting module, an EQ filter is used to perform calibration fitting on the frequency response difference curve located in the error region.

[0037] In one aspect of the present disclosure, a computer-readable medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the stereo equalization adjustment method described in any one of the aforementioned embodiments is implemented.

[0038] In one aspect of the present disclosure, a computer device is provided, including a storage module, a processor, and a computer program stored on the storage module and executable on the processor. When the processor executes the computer program, the stereo equalization adjustment method described in any of the aforementioned embodiments is implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1It is a flowchart of a stereo equalization adjustment method according to an embodiment of the present disclosure.

[0040] Figure 2 It is a flowchart of a stereo equalization adjustment method according to another embodiment of the present disclosure.

[0041] Figure 3 FIG. 1 is a schematic diagram of an embodiment of determining a maximum error region.

[0042] Figure 4 The flowchart of one embodiment of performing calibration fitting on a frequency response difference curve is shown in FIG.

[0043] Figure 5 4 is a structural block diagram of a stereo equalization adjustment device according to an embodiment of the present disclosure.

[0044] Figure 6 It is a structural block diagram of a computer device according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0045] The following describes some of the various embodiments of the present invention, which are intended to provide a basic understanding of the present invention, but are not intended to identify the key or decisive elements of the present invention or to limit the scope of protection.

[0046] For the purpose of brevity and illustration, the principles of the present invention are described herein primarily with reference to exemplary embodiments thereof. However, those skilled in the art will readily appreciate that the same principles are equally applicable to and may be implemented in all types of stereo equalization methods and devices, and that any such variations do not depart from the true spirit and scope of the present patent application.

[0047] Moreover, in the following description, reference is made to the accompanying drawings, which illustrate specific exemplary embodiments. Electrical, mechanical, logical, and structural changes may be made to these embodiments without departing from the spirit and scope of the present invention. In addition, although a feature of the present invention is disclosed in conjunction with only one of several embodiments, it may be desirable and / or advantageous to combine this feature with one or more other features of other embodiments as may be desired and / or advantageous for any given or identifiable function. Therefore, the following description should not be regarded in a limiting sense, and the scope of the present invention is defined by the appended claims and their equivalents.

[0048] Terms such as “having” and “including” indicate that in addition to the units (modules) and steps directly and clearly stated in the specification and claims, the technical solution of the present invention does not exclude the situation where it has other units (modules) and steps that are not directly or clearly stated.

[0049] For smart glasses, stereo balance is generally achieved by placing speakers on the left and right temples. During the production process, due to individual differences in speakers and differences caused by the assembly process, the stereo sound transmitted by smart glasses to the wearer's left and right ears will be different, causing the wearer to perceive an imbalance in the ears, which in turn leads to a weakening of the sound quality. In current mainstream smart glasses or AR glasses, the acoustic structure design is generally on the entire housing, which further exacerbates the imbalance between the ears.

[0050] First, some technical terms that will appear in the embodiments of the present disclosure are briefly explained.

[0051] (1) EQ (Equalize): Equalization effector

[0052] (2) GA (Genetic Algorithm): Genetic Algorithm

[0053] (3) SA (Simulated Annealing): Simulated annealing algorithm

[0054] (4) Fitness Function: Cost function, used as the convergence target of the intelligent optimization algorithm.

[0055] (5) Smart glasses: glasses that can interact with users through their own data processing capabilities, or through data communication with mobile phones, tablets, computers, etc., including but not limited to AR glasses, VR glasses, MR glasses, Bluetooth glasses, etc.

[0056] The smart glasses referred to in the embodiments of the present disclosure may have display capabilities, such as AR glasses, VR glasses, and MR glasses; or may not have display capabilities, such as Bluetooth glasses.

[0057] Figure 1 It is a flowchart of a stereo equalization adjustment method according to an embodiment of the present disclosure.

[0058] like Figure 1 As shown, the stereo balance adjustment method of the embodiment of the present disclosure can be applied to the stereo balance adjustment of smart glasses, including:

[0059] Step S100: obtaining a frequency response curve to be calibrated, for example, obtaining a frequency response curve of a stereo to be calibrated of smart glasses as the frequency response curve to be calibrated;

[0060] Step S200: performing differential processing on the frequency response curve to be calibrated and the preset target frequency response curve to obtain a frequency response differential curve;

[0061] Step S300: performing threshold discrimination on the frequency response difference curve in multiple frequency regions according to a preset threshold, and obtaining a frequency region where the frequency response difference curve exceeds the threshold the most or the largest as an error region;

[0062] Step S400: performing calibration fitting on the frequency response difference curve located in the error region using a prescribed algorithm, and outputting a calibrated frequency response difference curve; and

[0063] Step S500: Merging the calibrated frequency response difference curve and the frequency response curve to be calibrated to obtain a new frequency response difference curve.

[0064] The new frequency response difference curve obtained in step S500 is input into step S300 again and steps S300, S400 and S400 are repeated until the new frequency response difference curve obtained in step S500 does not exceed the threshold.

[0065] In some embodiments, multiple filters are preset, and each time step S400 is performed, one filter is used to perform calibration fitting on the frequency response difference curve located in the error region.

[0066] In some embodiments, step S300 may include:

[0067] Presetting an upper limit and a lower limit of sensitivity as the threshold;

[0068] Divide into multiple frequency regions according to frequency size;

[0069] Performing threshold determination on the frequency response difference curve according to the upper limit and the lower limit; and

[0070] Among the multiple frequency regions, a region where the frequency response difference curve exceeds the upper limit and the lower limit the most or the largest is used as the error region.

[0071] In some embodiments, the frequency regions are divided according to frequency magnitude in such a manner that the frequency response difference curves in each frequency region are located above the 0 axis or below the 0 axis.

[0072] In step S400, any one of the following algorithms is used for the frequency response difference curve located in the error region:

[0073] Genetic algorithm, simulated annealing algorithm.

[0074] For example, in step S400 , an EQ filter is used to perform calibration fitting on the frequency response difference curve located in the error region.

[0075] In this case, step S400 includes:

[0076] For the frequency response difference curve located in the error region, a cost function is used to obtain a maximum distance that deviates from the threshold; and

[0077] When the maximum distance is greater than 0, an EQ filter is used to perform calibration fitting on the frequency response difference curve of the error region.

[0078] In some embodiments, in step S500 , merging the calibrated frequency response difference curve and the frequency response curve to be calibrated is adding the calibrated frequency response difference curve and the frequency response curve to be calibrated according to frequency points.

[0079] In some embodiments, in step S500 , the filter parameters of the EQ filter are output simultaneously with the output of the new frequency response difference curve.

[0080] Figure 2 It is a flowchart of a stereo equalization adjustment method according to another embodiment of the present disclosure.

[0081] Next, refer to Figure 2 A stereo balance adjustment method according to another embodiment of the present disclosure will be described.

[0082] exist Figure 2 It is divided into three areas: S01 input area, S02 calculation area, and S03 output area.

[0083] Enter the following in the S01 input area:

[0084] (1) Configuration file: includes the number of available filters, number of retries, and other interface information configurations for external interaction.

[0085] (2) Target frequency response curve: It can also be called the golden machine curve, that is, the frequency response curve of the target product. It is the target of the automatic calibration algorithm. The curve can be obtained by acoustic testing of the target product and using audio analysis software (such as Soundcheck). I will not elaborate on this here.

[0086] (3) Frequency response curve to be calibrated: that is, the frequency response curve of the product to be calibrated. The method of obtaining it is the same as the target frequency response curve, except that the product is different.

[0087] (4) Threshold: used to determine whether the frequency response difference curve meets the requirements.

[0088] Next, perform the following calculations in the S02 calculation area:

[0089] Step S1: performing curve difference processing on the target frequency response curve and the frequency response curve to be calibrated to obtain a frequency response difference curve;

[0090] Step S2: The frequency response difference curve is judged in combination with the threshold. If the threshold judgment is yes (i.e., it does not exceed the threshold), the calculation area is jumped out and the process goes to step S9 in the S03 output area. If the threshold judgment is no, the process continues to step S3.

[0091] Step S3: Determine the remaining number of available filters for subsequent loop iterations. If the determination is yes, proceed to step S4; if not, proceed to step S7.

[0092] Step S4: Divide the frequency region into multiple frequency regions, perform threshold discrimination on the frequency response difference curve according to the threshold, identify the maximum error region where the frequency response difference curve exceeds the threshold the most (or the maximum), and return the maximum error region as the "error region", which will be referred to later. Figure 3 An embodiment of how to determine the maximum error area is described;

[0093] Step S5: calibrating and fitting the frequency response difference curve in the error region using a specified optimization algorithm, and outputting the calibrated frequency response difference curve. Common optimization algorithms include GA and SA, and GA is used as an example in this example.

[0094] Step S6: Merge the calibrated frequency response difference curve into the frequency response difference curve to be calibrated to obtain a new frequency response difference curve, and send the new frequency response difference curve to step S2. Meanwhile, the number of available filters is reduced by 1.

[0095] Step S7: Determine whether there are retry times, if yes, proceed to step S8, if not, jump to step S9;

[0096] Step S8: Reset the frequency response difference curve, and reduce the number of retries by 1;

[0097] Step S9: The calibration is completed and successful, and the filter coefficients of the filter are output;

[0098] Step S10: If the calibration is completed and fails, the reason for the calibration failure is output.

[0099] Reference here Figure 3 An example of how to determine the maximum error region will be described.

[0100] Figure 3 FIG. 1 is a schematic diagram of an embodiment of determining a maximum error region.

[0101] exist Figure 3 The horizontal axis is frequency and the vertical axis is sensitivity. Figure 3 As shown, the frequency response difference curve A is Figure 3The dividing lines D1 to D4 (for example, 100 Hz, 1200 Hz, 8000 Hz, and 10 kHz, respectively) are divided into three regions, namely, region 1, region 2, and region 3 from left to right. Figure 3 The dotted lines G1 and G2 in the upper and lower directions represent the upper and lower thresholds respectively. Figure 3 The middle discrimination region 3 is the region with the largest error that exceeds the threshold the most.

[0102] The purpose of identifying the maximum error region that exceeds the threshold the most is to prioritize returning to the region with the largest error, so as to reduce the impact on the edge region during the processing. And here for multiple frequency regions ( Figure 3 The goal of dividing the area into area 1, area 2 and area 3 is to make Figure 3 The frequency response difference curve in the image approaches the 0 axis and is separated by the zero point so that all the frequency response difference curves in the area are above or below the 0 axis, which makes it easier to process them uniformly.

[0103] Next, the specific process of calibrating and fitting the frequency response difference curve in the error region in step S5 is described.

[0104] Figure 4 The flowchart of one embodiment of performing calibration fitting on a frequency response difference curve is shown in FIG.

[0105] This embodiment uses an EQ filter, including a peak filter within the EQ filter. The peak filter is a biquad IIR filter defined by three parameters: Fc (center frequency), Q (quality factor), and Gain. These three coefficients can be mapped to IIR filter coefficients. The core of the calibration in this embodiment lies in finding the matching Fc, Q, and Gain coefficients.

[0106] like Figure 4 As shown in Figure 1, the process of calibrating and fitting the frequency response difference curve includes:

[0107] Step S11: Initialization, i represents the number of filters, and the initial value of i is 0;

[0108] Step S12: Use the cost function (Fitness Function) to define the maximum distance (Distance) that the frequency response difference curve deviates from the threshold. Distance is the output value of the cost function and is defined as the maximum distance the current frequency response difference curve deviates from the threshold. A positive number indicates deviation from the threshold, while a negative number indicates it is within the threshold. The advantage of defining this here is that it prioritizes solving the most difficult problems, thereby avoiding the impact of subsequent optimization fitting of other error areas.

[0109] Step S13: Determine whether the maximum distance is greater than 0. If it is greater than 0, proceed to step S14; if it is not greater than 0, jump to step S17;

[0110] Step S14: Determine whether the number of filters is greater than 0. If so, proceed to step S15; if not, jump to step S18.

[0111] Step S15: the number of filters i=1, the number of remaining filters -1;

[0112] Step S16: using the GA algorithm to perform calibration fitting on the frequency response difference curve of the error region (i.e., region 3) using i filters, wherein as an example, 0.1 <= Q <= 5 -12 <= Gain <= 12 is preset;

[0113] Step S17: The region is calibrated and the filter coefficients are output. Here, the output filter coefficients refer to the filter coefficients of the peak filter. The filter is actually a second-order IIR filter with a total of 6 coefficients: b0, b1, b2, a0, a1, and a2, which can be obtained by Fc, Q, and Gain mapping.

[0114] Step S18: The calibration of the area fails, and the reason for the failure is output.

[0115] As described above, according to the stereo balance adjustment method of the embodiment of the present disclosure, by identifying the error area, it is possible to simplify the problem to be solved, reduce the complexity of the algorithm, shorten the calculation time and improve the optimization efficiency. Moreover, by adopting the prescribed algorithm for fitting calibration and adopting multi-band EQ filter collaborative processing for a single complex problem, the success rate of algorithm optimization can be improved. Moreover, when the stereo balance adjustment method is applied to smart glasses products, stereo balance adjustment can be performed during the production stage, thereby improving the yield rate and enhancing the effect and balance of the smart glasses products.

[0116] The above describes the stereo balance adjustment method according to the embodiment of the present disclosure. Next, the stereo balance adjustment device according to the embodiment of the present disclosure will be described.

[0117] Figure 5 4 is a structural block diagram of a stereo equalization adjustment device according to an embodiment of the present disclosure.

[0118] like Figure 5 As shown, the stereo equalization adjustment device 100 of the embodiment of the present disclosure includes:

[0119] The curve acquisition module 110 is used to obtain a stereo frequency response curve as the frequency response curve to be calibrated;

[0120] The differential processing module 120 is used to perform differential processing on the frequency response curve to be calibrated and a preset target frequency response curve to obtain a frequency response differential curve;

[0121] A threshold determination module 130 is configured to perform threshold determination on the frequency response difference curve in multiple frequency regions according to a preset threshold, and obtain a frequency region where the frequency response difference curve exceeds the threshold the most as an error region;

[0122] a calibration and fitting module 140 that performs calibration and fitting on the frequency response difference curve located in the error region using a prescribed algorithm and outputs a calibrated frequency response difference curve; and

[0123] The curve merging module 150 is configured to merge the calibrated frequency response difference curve and the frequency response curve to be calibrated to obtain a new frequency response difference curve.

[0124] The new frequency response difference curve obtained in the curve merging module 150 is re-input into the threshold determination module 130 as the frequency response difference curve, and the operations of the threshold determination module 130, the calibration module 140, and the curve merging module 150 are repeatedly executed until the new frequency response difference curve obtained by the curve merging module 150 does not exceed the threshold.

[0125] The threshold determination module 130 pre-sets an upper limit and a lower limit of sensitivity as the threshold. Furthermore, the frequency response difference curve is divided into multiple frequency regions according to the frequency magnitude, and a threshold determination is performed on the frequency response difference curve according to the upper limit and the lower limit. Among the multiple frequency regions, the region in which the frequency response difference curve exceeds the upper limit and the lower limit the most is determined as the error region.

[0126] The division into multiple frequency regions according to frequency magnitude is performed in such a manner that the frequency response difference curve in each frequency region is located above the 0 axis or below the 0 axis.

[0127] In the calibration fitting module 140 , a plurality of filters are pre-set, and each time the calibration fitting step is performed, one filter is used to perform calibration fitting on the frequency response difference curve located in the error region.

[0128] In the calibration fitting module 140 , any one of the following algorithms is used for the frequency response difference curve located in the error region: an equalizer algorithm, a genetic algorithm, and a simulated annealing algorithm.

[0129] In the calibration and fitting module 140 , an EQ filter is used to perform calibration and fitting on the frequency response difference curve located in the error region.

[0130] The calibration and fitting module 140 uses a cost function to obtain a maximum distance from the threshold for the frequency response difference curve in the error region, and uses the EQ filter to perform calibration fitting on the frequency response difference curve in the error region when the maximum distance is greater than 0.

[0131] In the curve merging module 150 , the calibrated frequency response difference curve and the frequency response curve to be calibrated are merged by adding the calibrated frequency response difference curve and the frequency response curve to be calibrated according to the frequency points.

[0132] Furthermore, the curve incorporating module 150 outputs the parameters of the EQ filter while outputting the new frequency response difference curve.

[0133] Figure 6 It is a structural block diagram of a computer device according to an embodiment of the present disclosure.

[0134] like Figure 6 As shown, an embodiment of the present disclosure further provides a computer device 500, comprising a memory 510 and a processor 520, wherein the memory 510 is configured to store computing instructions, and the processor 520 is configured to execute the stereo equalization adjustment method provided in any one of the aforementioned embodiments when running the computer instructions.

[0135] The memory 510 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory, electrically erasable programmable read-only memory, erasable programmable read-only memory, programmable read-only memory, read-only memory, magnetic memory, flash memory, magnetic disk or optical disk, etc.

[0136] The processor 520 can be a logical operation device with data processing capabilities and / or program execution capabilities, such as a central processing unit, a field programmable logic array, a single-chip microcomputer, a digital signal processor, or an application-specific integrated circuit. One or more processors can be configured to simultaneously execute the above-mentioned stereo equalization adjustment method as a parallel computing processor group, or configured to have some processors execute some steps in the above-mentioned stereo equalization adjustment method, while some processors execute other steps in the above-mentioned stereo equalization adjustment method, etc. Computer instructions include one or more processor operations defined by the instruction set architecture corresponding to the processor, and these computer instructions can be logically contained and represented by one or more computer programs.

[0137] The computer device 500 can also be connected to various input devices (such as user interfaces, keyboards, etc.), various output devices (such as speakers, etc.), and display screens to enable interaction between the computer device and other products or users, which will not be repeated here.

[0138] The above examples primarily illustrate the stereo balance adjustment method and stereo balance adjustment device of the embodiments of the present disclosure. Although only some specific embodiments of the present invention have been described, those skilled in the art will appreciate that the present invention may be implemented in many other forms without departing from its spirit and scope. Therefore, the examples and implementations presented are to be considered illustrative rather than restrictive, and the present invention may encompass various modifications and substitutions without departing from the spirit and scope of the present invention as defined in the appended claims.

Claims

1. A stereo equalization adjustment method, It is characterized in that include: A curve acquisition step, acquiring a stereo frequency response curve as the frequency response curve to be calibrated; A differential processing step of performing differential processing on the frequency response curve to be calibrated and a preset target frequency response curve to obtain a frequency response differential curve; A threshold determination step, performing threshold determination on the frequency response difference curve in multiple frequency regions according to a preset threshold, and obtaining a frequency region where the frequency response difference curve exceeds the threshold the most as an error region; A calibration fitting step, wherein a prescribed algorithm is used to perform calibration fitting on the frequency response difference curve located in the error region, and a calibrated frequency response difference curve is output; as well as a curve merging step, merging the calibrated frequency response difference curve and the frequency response curve to be calibrated to obtain a new frequency response difference curve, Wherein, the calibration fitting step comprises: For the frequency response difference curve located in the error region, a cost function is used to define the maximum distance that the frequency response difference curve deviates from the threshold, wherein the maximum distance is the output value of the cost function, which is defined as the maximum distance that the current frequency response difference curve deviates from the threshold. If it is a positive number, it deviates from the threshold, and if it is a negative number, it is within the threshold, and In the case where the maximum distance is a positive number, a filter is used to perform calibration fitting on the frequency response difference curve of the error region, and filter coefficients of the filter are output when the calibration fitting is completed.

2. The stereo equalization adjustment method according to claim 1, It is characterized in that The threshold determination step comprises: Pre-set sensitivity threshold; Divide into multiple frequency regions according to frequency size; The frequency response difference curve is judged according to the threshold; and Among the multiple frequency regions, a region where the frequency response difference curve exceeds the threshold the most is used as the error region.

3. The stereo equalization adjustment method according to claim 2, It is characterized in that The division into multiple frequency regions according to frequency magnitude is performed in such a way that the frequency response difference curve in each frequency region is located above the 0 axis or below the 0 axis.

4. The stereo balance adjustment method according to claim 1, It is characterized in that Further including: The new frequency response difference curve obtained in the curve merging step is re-input into the threshold determination step and the threshold determination step, the calibration fitting step and the curve merging step are repeated until the new frequency response difference curve obtained in the curve merging step does not exceed the threshold.

5. The stereo balance adjustment method according to claim 4, It is characterized in that Pre-set multiple filters, Each time the calibration fitting step is performed, a filter is used to perform calibration fitting on the frequency response difference curve located in the error region.

6. The stereo balance adjustment method according to claim 1, It is characterized in that In the calibration fitting step, any one of the following algorithms is used for the frequency response difference curve located in the error region: Genetic algorithm, simulated annealing algorithm.

7. The stereo balance adjustment method according to claim 6, It is characterized in that In the calibration and fitting step, an EQ filter is used to perform calibration and fitting on the frequency response difference curve located in the error region.

8. The stereo balance adjustment method according to claim 1, It is characterized in that In the curve merging step, the calibrated frequency response difference curve and the frequency response curve to be calibrated are fused by adding the calibrated frequency response difference curve and the frequency response curve to be calibrated according to frequency points.

9. The stereo balance adjustment method according to claim 7, It is characterized in that In the curve merging step, the filter parameters of the EQ filter are outputted while outputting the new frequency response difference curve.

10. A stereo balance adjustment device, It is characterized in that A curve acquisition module, used for acquiring a stereo frequency response curve as a frequency response curve to be calibrated; A differential processing module, used for performing differential processing on the frequency response curve to be calibrated and a preset target frequency response curve to obtain a frequency response differential curve; A threshold determination module, configured to perform threshold determination on the frequency response difference curve in multiple frequency regions according to a preset threshold, and obtain a frequency region where the frequency response difference curve exceeds the threshold the most as an error region; A calibration fitting module, used for performing calibration fitting on the frequency response difference curve located in the error region using a prescribed algorithm, and outputting a calibrated frequency response difference curve; as well as A curve merging module is used to merge the calibrated frequency response difference curve and the frequency response curve to be calibrated to obtain a new frequency response difference curve, Wherein, the calibration fitting module is used to perform the following steps: For the frequency response difference curve located in the error region, a cost function is used to define the maximum distance that the frequency response difference curve deviates from the threshold, wherein the maximum distance is the output value of the cost function, which is defined as the maximum distance that the current frequency response difference curve deviates from the threshold. If it is a positive number, it deviates from the threshold, and if it is a negative number, it is within the threshold, and In the case where the maximum distance is a positive number, a filter is used to perform calibration fitting on the frequency response difference curve of the error region, and filter coefficients of the filter are output when the calibration fitting is completed.

11. The stereo balance adjustment device according to claim 10, It is characterized in that The threshold determination module pre-sets a sensitivity threshold and divides the frequency region into multiple frequency regions according to the frequency magnitude. The frequency response difference curve is subjected to threshold determination according to the threshold. Among the multiple frequency regions, a region where the frequency response difference curve exceeds the threshold the most is taken as the error region.

12. The stereo balance adjustment device according to claim 10, It is characterized in that The new frequency response difference curve obtained in the curve merging module is input into the threshold determination module as a frequency response difference curve, and the threshold determination module, the calibration fitting module and the curve merging module repeatedly perform actions in sequence until the new frequency response difference curve obtained by the curve merging module does not exceed the threshold.

13. The stereo balance adjustment device according to claim 12, It is characterized in that In the calibration fitting module, a plurality of filters are preset, and one filter is used each time the frequency response difference curve is calibrated and fitted.

14. The stereo balance adjustment device according to claim 13, It is characterized in that In the calibration fitting module, an EQ filter is used to perform calibration fitting on the frequency response difference curve located in the error region.

15. A computer readable medium having a computer program stored thereon, It is characterized in that When the computer program is executed by a processor, the stereo equalization adjustment method according to any one of claims 1 to 9 is implemented.

16. A computer device comprising a storage module, a processor, and a computer program stored in the storage module and executable on the processor, It is characterized in that When the processor executes the computer program, the stereo equalization adjustment method according to any one of claims 1 to 9 is implemented.

17. A computer program product having a computer program stored therein, It is characterized in that When the computer program is executed by a processor, the stereo equalization adjustment method according to any one of claims 1 to 9 is implemented.

Images