Optimal delay acquisition method and device for sound field partition control, equipment and medium

Abstract

The invention discloses an optimal delay acquisition method and device for sound field partition control, equipment and a medium, and relates to the technical field of sound field control, and the method comprises the steps: building an audio signal library, and obtaining an expected signal corresponding to each audio signal in the audio signal library; obtaining each filtering signal obtained by each audio signal through a secondary path filter, obtaining an optimal control filter corresponding to each filtering signal according to each filtering signal and a corresponding expected signal, and further obtaining a theoretical noise reduction amount of each audio signal; and taking the average value of each theoretical noise reduction amount as a cost function, and finding the delay corresponding to the minimum cost function by using golden section search in a preset initial delay range as the optimal delay. According to the method, the optimal delay with high robustness can be effectively selected to carry out in-vehicle sound field partition control.

Description

Technical Field

[0001] The present application relates to the technical field of sound field control, and in particular to an optimal delay acquisition method, device, equipment and medium for sound field zoning control. Background Art

[0002] The in-car sound field zoning control is to design individual independent sound zones to meet the personalized sound field needs of the in-car occupants, and is an important part of the smart cockpit scene. At present, there are two main methods for sound field zoning control. The first method is to reconstruct the target sound field in a certain area by means of sound field reconstruction, while reducing the interference of this area to the sound fields in other areas, such as the acoustic energy contrast method (ACC) and the sound pressure matching method (PM). The second method is to use the active noise control (ANC) scheme to achieve in-car sound field zoning control. The sound field in the bright area is not controlled, but the filter is designed for the sound source in the bright area in the dark area, so that the bright area audio signal at the microphone in the dark area is offset. Compared with the first method, the sound field zoning control based on ANC will not reduce the sound quality of the bright area sound field, but due to the characteristics of ANC itself, it is often only effective for low-frequency signals.

[0003] In the related art, for the sound field partition control scheme based on ANC, a feedforward control mode is usually adopted, and the optimal filter of different areas is calculated by the reference signal and the error signal, and the output signal of each area is filtered and emitted by the speakers of each area. At this time, the reference signal is no longer the noise signal in the traditional ANC, but the media music, navigation prompt sound, conversation voice, etc. from other partitions. The "noise" related to the reference signal is filtered out by the ANC controller of this area, and the irrelevant signal is retained. For the feedforward ANC system, when the causality of the system is not satisfied, that is, the reference signal is processed by the filter and output by the secondary source, and arrives at the error microphone later than the expected signal, the noise reduction performance of the system will be greatly reduced. In the application of common ANC, causality is guaranteed by placing the reference sensor as close to the noise source as possible or reducing the delay of the secondary path. For the partition control of the sound field in the car, since the playback audio at different positions can be obtained in advance, the delay between the acquisition of the reference signal and the acquisition of the expected signal can be artificially adjusted, that is, the causality of the system can be fully guaranteed.

[0004] However, too long a delay is not conducive to the listening experience of passengers in the car, and there are many types of audio signals played in the car with obvious differences. The optimal delay for one audio signal may not be applicable to other audio signals, making it difficult to determine the optimal system delay. Summary of the invention

[0005] The present application provides a method, device, equipment and medium for obtaining the optimal delay of sound field partition control, which can solve the technical problem in the prior art that it is difficult to determine the optimal system delay.

[0006] In a first aspect, the present application provides a method for obtaining an optimal delay for sound field partition control, the method comprising:

[0007] Establishing an audio signal library, and obtaining an expected signal corresponding to each audio signal in the audio signal library;

[0008] Obtaining each filtered signal obtained by each audio signal passing through the secondary path filter, and obtaining the optimal control filter corresponding to each filtered signal according to each filtered signal and the corresponding expected signal, thereby obtaining the theoretical noise reduction amount of each audio signal;

[0009] The average value of each theoretical noise reduction amount is used as the cost function, and the golden section search is used to find the delay corresponding to the minimum cost function within the preset initial delay range as the optimal delay.

[0010] In combination with the first aspect, in one implementation, the above-mentioned using the golden section search within the preset initial delay range to find the delay corresponding to the minimum cost function specifically includes:

[0011] Obtain two golden section points within the above initial delay range, and obtain the average values ​​corresponding to the two golden section points; when the two average values ​​are different, the golden section point with the larger average value is taken as the first point, and the other golden section point is taken as the second point; when the two average values ​​are the same, any golden section point is taken as the first point, and the other golden section point is taken as the second point;

[0012] A new delay range is obtained by discarding the interval on the side of the first point away from the second point in the above initial delay range, taking the second point as a new golden section point of the new delay range, and calculating another new golden section point until the cost function is minimized and the corresponding delay is obtained.

[0013] In combination with the first aspect, in one implementation, when the delay value at the first point is less than the delay value at the second point, the first point is used as the lower limit of the new delay range, and the product of the difference between the upper limit and the lower limit of the new delay range and 0.618 is obtained, and the sum of the product and the lower limit of the new delay range is used as another new golden section point;

[0014] When the delay value of the first point is greater than the delay value of the second point, the first point is used as the upper limit of the new delay range, and the product of the difference between the upper limit and the lower limit of the new delay range and 0.382 is obtained, and the sum of the product and the lower limit of the new delay range is used as another new golden section point.

[0015] In combination with the first aspect, in one implementation, the two golden section points μ within the initial delay range are c and μ d They are:

[0016] μ c=μ a +0.382(μ b -μ a )

[0017] μ d =μ a +0.618(μ b -μ a )

[0018] Among them, μ a is the lower limit of the initial delay range, μ b The upper limit of the initial delay range.

[0019] In combination with the first aspect, in one implementation, obtaining the expected signal corresponding to each audio signal in the audio signal library specifically includes:

[0020] Determine the bright and dark areas based on the application scenario;

[0021] Each audio signal is used as a reference signal, and when the reference signal is played with a delay in the bright area, the collected microphone signal in the dark area is used as the expected signal corresponding to the reference signal.

[0022] In combination with the first aspect, in one implementation, before obtaining each filtered signal obtained by passing each audio signal through a secondary path filter, the method further includes:

[0023] The secondary sound system in the dark area plays white noise in turn and collects the microphone signal in the dark area for preprocessing. The normalized least mean square algorithm is used for system identification to obtain the secondary path filter in the dark area.

[0024] In combination with the first aspect, in one implementation, obtaining an optimal control filter corresponding to any filtered signal and a corresponding expected signal specifically includes:

[0025] Obtaining an autocorrelation matrix of the filtered signal and a cross-correlation vector between the filtered signal and the delayed desired signal;

[0026] According to the above autocorrelation matrix and cross-correlation vector, the optimal control filter corresponding to the filter signal is obtained.

[0027] In a second aspect, the present application provides an optimal delay acquisition device for sound field partition control, the device comprising:

[0028] A first acquisition module, which is used to establish an audio signal library and acquire an expected signal corresponding to each audio signal in the audio signal library;

[0029] A second acquisition module is used to obtain each filtered signal obtained by each audio signal through the secondary path filter, and obtain the optimal control filter corresponding to each filtered signal according to each filtered signal and the corresponding expected signal, so as to obtain the theoretical noise reduction amount of each audio signal;

[0030] The golden section search module is used to use the average value of each theoretical noise reduction amount as the cost function, and use the golden section search to find the delay corresponding to the minimum cost function within a preset initial delay range as the optimal delay.

[0031] In the third aspect, the present application provides an optimal delay acquisition device for sound field partition control, the above-mentioned optimal delay acquisition device for sound field partition control includes a processor, a memory, and an optimal delay acquisition program stored in the above-mentioned memory and executable by the above-mentioned processor, wherein when the above-mentioned optimal delay acquisition program is executed by the above-mentioned processor, the steps of the above-mentioned optimal delay acquisition method are implemented.

[0032] In a fourth aspect, the present application provides a computer-readable storage medium, on which an optimal delay acquisition program is stored, wherein when the optimal delay acquisition program is executed by a processor, the steps of the optimal delay acquisition method are implemented.

[0033] The beneficial effects of the technical solution provided by this application include:

[0034] By establishing an audio signal library, obtaining the expected signal corresponding to each audio signal in the above audio signal library, and obtaining each filtered signal obtained by each audio signal after passing through the secondary path filter, the optimal control filter corresponding to each filtered signal can be obtained according to each filtered signal and the corresponding expected signal, and then the theoretical noise reduction amount of each audio signal can be obtained; since there will be different optimal delays for different audio files, in order to obtain the optimal delay with high robustness and suitable for most audio files, the theoretical noise reduction amounts of all audio signals are averaged as the cost function of the final search algorithm, and the golden section search is used within the preset initial delay range to find the delay corresponding to the minimum cost function as the optimal delay. Therefore, the optimal delay can be effectively selected to perform in-vehicle sound field zoning control, and it is applicable to different sound source signals, which has important practical value and research significance for in-vehicle sound field zoning control based on active noise control algorithm. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 This is a flow chart of an embodiment of an optimal delay acquisition method of the present application;

[0036] Figure 2 A flowchart of another embodiment of the optimal delay acquisition method of the present application;

[0037] Figure 3 This is a schematic diagram of the control of the driver and co-passenger sound field partitions based on active noise control in an embodiment of the present application;

[0038] Figure 4 A schematic diagram of a golden section search algorithm (GSS) in an embodiment of the present application;

[0039] Figure 5 This is a comparison of the noise reduction amount without delay and with optimal delay in the dark area in the embodiment of this application;

[0040] Figure 6 This is a functional module diagram of an embodiment of an optimal delay acquisition device of the present application;

[0041] Figure 7 This is a schematic diagram of the hardware structure of the optimal delay acquisition device involved in the embodiment of the present application. DETAILED DESCRIPTION

[0042] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0043] In a first aspect, an embodiment of the present application provides an optimal delay acquisition method for sound field zoning control.

[0044] Reference Figure 1 , Figure 1 The following is a flow chart of an embodiment of an optimal delay acquisition method for sound field partition control of the present application. The above optimal delay acquisition method includes:

[0045] S1. Establish an audio signal library and obtain the desired signal corresponding to each audio signal in the audio signal library;

[0046] S2. Obtain each filtered signal obtained by each audio signal through the secondary path filter, and obtain the optimal control filter corresponding to each filtered signal according to each filtered signal and the corresponding expected signal, thereby obtaining the theoretical noise reduction amount of each audio signal;

[0047] S3. Taking the average value of each theoretical noise reduction amount as the cost function, using the golden section search within the preset initial delay range to find the delay corresponding to the minimum cost function as the optimal delay.

[0048] In this embodiment, by establishing an audio signal library, obtaining the expected signal corresponding to each audio signal in the above audio signal library, and obtaining each filtered signal obtained by each audio signal after passing through the secondary path filter, the optimal control filter corresponding to each filtered signal can be obtained according to each filtered signal and the corresponding expected signal, and then the theoretical noise reduction amount of each audio signal can be obtained; since there will be different optimal delays for different audio files, in order to obtain the optimal delay with high robustness and suitable for most audio files, the theoretical noise reduction amounts of all audio signals are averaged as the cost function of the final search algorithm, and the golden section search is used within the preset initial delay range to find the delay corresponding to the minimum cost function as the optimal delay. Therefore, the optimal delay can be effectively selected to perform in-vehicle sound field zoning control, and it is applicable to different sound source signals, which has important practical value and research significance for in-vehicle sound field zoning control based on active noise control algorithm.

[0049] Further, in one embodiment, in the above step S3, finding the delay corresponding to the minimum cost function by using the golden section search within the preset initial delay range specifically includes:

[0050] First, two golden section points within the above-mentioned initial delay range are obtained, and the average values ​​corresponding to the two golden section points are obtained; wherein, when the average values ​​corresponding to the two golden section points are different, the golden section point with the larger average value is taken as the first point, and the other golden section point is taken as the second point; when the average values ​​corresponding to the two golden section points are the same, any one of the golden section points is taken as the first point, and the other golden section point is taken as the second point;

[0051] Then, the new delay range is obtained by discarding the interval on the side of the first point away from the second point in the above initial delay range; the second point is used as a new golden section point of the new delay range, and another new golden section point is calculated to further narrow the new delay range until the cost function is minimized and the corresponding delay is obtained.

[0052] It can be understood that the golden section search algorithm is similar to the binary search algorithm. It is a heuristic iterative optimization algorithm that requires the determination of the search criteria, search range, and search stop conditions. The search criteria is the cost function that determines the search process and is used to determine whether the current search result is better.

[0053] In this embodiment, the cost function is selected as the theoretical noise reduction amount of the dark area. When the current parameters searched out increase the theoretical noise reduction amount of the dark area, the search boundary value is updated. When the newly searched parameters make the theoretical noise reduction amount stable and no longer change, the search is stopped. The parameters at this time are the optimal delay of the current audio signal.

[0054] Furthermore, in this embodiment, when calculating another new golden section point, the size of the first point and the second point is first determined.

[0055] When the delay value of the first point is less than the delay value of the second point, the first point is used as the lower limit of the new delay range, and the product of the difference between the upper limit and the lower limit of the new delay range and 0.618 is obtained, and the sum of the product and the lower limit of the new delay range is used as another new golden section point.

[0056] When the delay value of the first point is greater than the delay value of the second point, the first point is used as the upper limit of the new delay range, and the product of the difference between the upper limit and the lower limit of the new delay range and 0.382 is obtained, and the sum of the product and the lower limit of the new delay range is used as another new golden section point.

[0057] In this embodiment, when the in-vehicle sound field zoning based on active noise control is applied, the reference signal is usually acquired in advance. For the scene of the main driver's bright area and the co-driver's dark area, the reference signal of the co-driver's dark area is the audio signal input to the main driver's speaker, and the expected signal of the co-driver's dark area is the audio signal played by the main driver's bright area speaker received by the co-driver's microphone. When the acquisition of the reference signal can be ahead of the acquisition of the expected signal, it is often more beneficial to the causality of the active noise control algorithm, and better noise reduction performance can be obtained. Therefore, a certain delay can be added to the expected signal of the main driver, so that the reference signal is collected ahead of the expected signal to obtain better causality. However, this delay does not always improve the effect. On the contrary, when the increased delay is too large, it will cause the audio heard by the passengers to have a significant lag, which will bring adverse effects. Therefore, the algorithm based on the golden section search in this embodiment can realize the search for the optimal delay, thereby obtaining the optimal brightness and darkness contrast performance improvement.

[0058] Furthermore, in one embodiment, the two golden section points μ within the initial delay range are c and μ d They are:

[0059] μ c =μ a +0.382(μ b -μ a )

[0060] μ d =μ a +0.618(μ b -μ a )

[0061] Among them, μ a is the lower limit of the initial delay range, μ b The upper limit of the initial delay range.

[0062] Further, in one embodiment, in the above step S1, obtaining the expected signal corresponding to each audio signal in the above audio signal library specifically includes:

[0063] First, determine the bright and dark areas based on the application scenario;

[0064] Then, each audio signal is used as a reference signal, and when the reference signal is played with a delay in the bright area, the collected dark area microphone signal is used as the expected signal corresponding to the reference signal, that is, the expected signal corresponding to the audio signal.

[0065] Furthermore, in this embodiment, before obtaining each filtered signal obtained by each audio signal passing through the secondary path filter, the method further includes:

[0066] The secondary sound system in the dark area plays white noise in turn and collects the microphone signal in the dark area for preprocessing. The normalized least mean square algorithm is used for system identification to obtain the secondary path filter in the dark area.

[0067] In this embodiment, by playing white noise in sequence with the secondary sound of the dark area and collecting the microphone signal of the dark area for system identification, the secondary path filter of the dark area is obtained, and better estimation accuracy can be achieved.

[0068] Further, in one embodiment, in the above step S2, according to any filtered signal and the corresponding expected signal, obtaining the optimal control filter corresponding to the filtered signal specifically includes:

[0069] First, an autocorrelation matrix of the filtered signal and a cross-correlation vector between the filtered signal and the delayed desired signal are obtained;

[0070] Then, according to the above autocorrelation matrix and cross-correlation vector, the optimal control filter corresponding to the filter signal is obtained.

[0071] In this embodiment, the autocorrelation matrix of the filtered signal and the cross-correlation vector between the filtered signal and the expected signal can be calculated through offline secondary path identification, thereby calculating the Wiener filter result, and then the theoretical calculation formula is used to obtain the theoretical denoising result as the cost function of the golden section search.

[0072] The method of this embodiment collects a large number of audio signals to establish an audio signal database, uses the Wiener filter algorithm to calculate the optimal control filters for different sound sources, and simultaneously calculates the theoretical noise reduction amounts of different sound sources, averages them, and then uses the golden section search method to obtain the delay with the best noise reduction amount for the audio signal database as the optimal delay, thereby achieving the optimal delay in a scenario where multiple audios are used as sound sources.

[0073] like Figure 2 As shown, the above optimal delay acquisition method specifically includes:

[0074] A1. According to the application scenario, determine the bright area and dark area, and measure the secondary path transfer function, and then obtain the secondary path filter;

[0075] A2. Obtaining the expected signal corresponding to each audio signal, and then obtaining each filtered signal and the corresponding optimal control filter, and calculating the theoretical noise reduction amount of each audio signal;

[0076] A3. Set the initial delay range;

[0077] A4. Calculate the average value of the theoretical noise reduction amount at the golden section point and update the boundary of the delay range;

[0078] A5. Determine whether the cost function is minimized. If so, go to A6; otherwise, go to A4.

[0079] A6. Output the delay corresponding to the minimum cost function as the optimal delay.

[0080] In this embodiment, the process stops when the difference between the cost function value of the current iteration and the cost function value of the previous iteration is less than a preset minimum value, or when the search delay range is less than a preset minimum range. At this time, it is determined that the cost function is minimized.

[0081] In one embodiment, taking the selection of audio signal delay parameters when the active noise control solution is used to control the zoning of the sound field in the vehicle as an example, the specific implementation steps of the above method are as follows:

[0082] 1) Obtain the audio signal and the corresponding expected signal, and perform secondary path identification

[0083] like Figure 3 As shown, based on the actual application scenario, the corresponding bright area position and dark area position are determined (taking the main driver's bright area and the co-driver's dark area as examples), the main driver's bright area delays the playing of the audio signal in the audio signal library, and the dark area microphone signal is collected as the expected signal for the subsequent solution of the Wiener filter, and the undelayed audio signal is used as the reference signal; then, the secondary audio in the dark area is used to play white noise in sequence and collect the microphone signal in the dark area for preprocessing, and the normalized least mean square algorithm (NLMS) is used for system identification to obtain the secondary path filter in the dark area.

[0084] 2) Perform multi-channel Wiener filtering modeling on the input delayed audio signal to obtain the optimal control filter under the current delay. The derivation process is as follows:

[0085] The error signal of the co-pilot dark zone is expressed as

[0086]

[0087] Where j is the secondary source number, k is the error microphone number, m is the dummy variable of the secondary path convolved with the output signal, n is the current nth sampling point, μ is the delay added to the main audio signal, and d k (n-μ) represents the delayed desired signal caused by the delayed playback of the audio signal, s jk =[s jk (0),s jk (1),…,s jk (M-1)] T is a secondary path of length M, y j The output of the main audio signal after passing through an L-order FIR filter is expressed as:

[0088]

[0089] Where l represents the control filter coefficient w j (n) and the dummy variable of the vector convolution of the input signal, i.e., the reference signal x(n);

[0090] w j (n)=[w j,0 (n),w j,1 (n),…,w j,L-1 (n)] T

[0091] x(n)=[x(n),x(n-1),…,x(n-L+1)] T

[0092] The cost function of the filter design is defined as the mean square error:

[0093]

[0094] Where “E(·)” represents the time average of the independent variable. The gradient of the two filter coefficient vectors is calculated for the cost function, and recorded as The optimal control filter coefficient is calculated using the Wiener filtering algorithm:

[0095] w opt =-{E[R T (n)R(n)]}- 1 E[R T (n)d(n-μ)]

[0096] in,

[0097]

[0098] r jk (n) = [r jk (n),r jk (n-1),…,rjk (n-L+1)] T

[0099]

[0100] d(n-μ)=[d 1 (n-μ),d 2 (n-μ)] T

[0101] is the estimated secondary path from the jth secondary source to the kth error point.

[0102] 3) Calculate the theoretical noise reduction amount for different audio signals

[0103] The theoretical noise reduction amount of the current audio signal is calculated using the optimal control filter coefficients obtained above. The calculation formula is:

[0104] F i (μ)=E[d T (n-μ)d(n-μ)]-E[d T (n-μ)R(n)]{E[R T (n)R(n)]}- 1 E[R T (n)d(n-μ)]

[0105] Among them, i represents the i-th audio signal in the audio signal library. All audio signals in the audio signal library are traversed, their theoretical noise reduction amount is calculated, and finally the average is taken as the cost function of the golden section search method:

[0106]

[0107] 4) Use the golden section search method to search for the optimal delay

[0108] The golden section search method is an iterative optimization algorithm that finds the minimum of a unimodal continuous function by simply iteratively narrowing the search range without the need for derivative operations.

[0109] First, initialize the parameters to determine the optimal range of delay, that is, the initial delay range [μ a ,μ b ], where μ a Optional: 0, μ b It can be determined based on actual conditions or experience; then two golden section points within the range are selected as follows:

[0110] μ c =μ a +0.382(μ b -μ a )

[0111] μ d =μ a +0.618(μ b -μ a )

[0112] Then, calculate μ c and μ d The corresponding F(μ c ) and F(μ d ) and compare the sizes of the two.

[0113] like Figure 4 As shown in Iteration 1 in , when F(μ c )>F(μ d ), the new optimization range will be limited to the new delay range [μ c ,μ b ]. According to the new delay range, the new golden section point is:

[0114] μ a ←μ c

[0115] μ c ←μ d

[0116] μ d =μ a +0.618(μ b -μ a )

[0117] Get the new μ c and μ d After that, we only need to compare the new F(μ c ) and F(μ d ), the next optimization can be performed. From the above formula, it can be seen that μ of the current iteration c It is the μ of the previous iteration. d , indicating that the current iteration F(μ c ) and the previous iteration F(μ d ) are equal. In the current iteration, we only need to know F(μ d ) value, you can continue with subsequent iterations.

[0118] like Figure 4 As shown in the iteration 2 in the figure, when optimizing again, F(μ c ) is less than F(μ d ), therefore, the new optimization range will be locked in [μ a ,μ d ]. According to the new delay range, the new golden section point is:

[0119] μ b ←μ d

[0120] μ d ←μ c

[0121] μ c =μ a +0.382(μ b -μ a )

[0122] Accordingly, we only need to know F(μ c ) value to complete the current iteration and enter the next optimization.

[0123] like Figure 5 As shown in the figure, after obtaining the optimal delay, the noise reduction amount of the dark area without delay and the optimal delay is compared, and it is found that the noise reduction effect of the optimal delay is better.

[0124] In this embodiment, the optimization range is narrowed by continuous iteration until the optimal value of the function is found, that is, the optimal system delay suitable for the audio signal library is obtained.

[0125] This embodiment involves the sound field control technology of the intelligent cockpit. The optimal delay is calculated by using the golden section search algorithm. At the same time, the robustness of the delay under different audio signals is considered, and an effective optimal delay design scheme is proposed, which can be used to obtain the optimal delay for the partition control of the sound field in the car. When the active noise control algorithm is used for partition control in the car, the delay between the reference signal and the expected signal is adjusted to effectively improve the energy contrast of the sound field partition.

[0126] In a second aspect, the present application also provides an optimal delay acquisition device for sound field partition control. Figure 6 , Figure 6 This is a functional module diagram of an embodiment of an optimal delay acquisition device for sound field partition control of the present application. The optimal delay acquisition device comprises a first acquisition module, a second acquisition module and a golden section search module.

[0127] The first acquisition module is used to establish an audio signal library and acquire the expected signal corresponding to each audio signal in the audio signal library.

[0128] The second acquisition module is used to obtain each filtered signal obtained by each audio signal through the secondary path filter, and obtain the optimal control filter corresponding to each filtered signal according to each filtered signal and the corresponding expected signal, so as to obtain the theoretical noise reduction amount of each audio signal.

[0129] The golden section search module is used to use the average value of each theoretical noise reduction amount as the cost function, and use the golden section search to find the delay corresponding to the minimum cost function within the preset initial delay range as the optimal delay.

[0130] Furthermore, in one embodiment, the golden section search module is also used for:

[0131] Obtain two golden section points within the above initial delay range, and obtain the average values ​​corresponding to the two golden section points; when the two average values ​​are different, the golden section point with the larger average value is taken as the first point, and the other golden section point is taken as the second point; when the two average values ​​are the same, any golden section point is taken as the first point, and the other golden section point is taken as the second point;

[0132] A new delay range is obtained by discarding the interval on the side of the first point away from the second point in the above initial delay range, taking the second point as a new golden section point of the new delay range, and calculating another new golden section point until the cost function is minimized and the corresponding delay is obtained.

[0133] Furthermore, in one embodiment, the golden section search module is also used for:

[0134] When the delay value of the first point is less than the delay value of the second point, the first point is used as the lower limit of the new delay range, and the product of the difference between the upper limit and the lower limit of the new delay range and 0.618 is obtained, and the sum of the product and the lower limit of the new delay range is used as another new golden section point;

[0135] When the delay value of the first point is greater than the delay value of the second point, the first point is used as the upper limit of the new delay range, and the product of the difference between the upper limit and the lower limit of the new delay range and 0.382 is obtained, and the sum of the product and the lower limit of the new delay range is used as another new golden section point.

[0136] Furthermore, in one embodiment, the two golden section points μ within the initial delay range are c and μ d They are:

[0137] μ c =μ a +0.382(μ b -μ a )

[0138] μ d =μ a +0.618(μ b -μ a )

[0139] Among them, μ a is the lower limit of the initial delay range, μ bThe upper limit of the initial delay range.

[0140] Furthermore, in one embodiment, the first acquisition module is further used for:

[0141] Determine the bright and dark areas based on the application scenario;

[0142] Each audio signal is used as a reference signal, and when the reference signal is played with a delay in the bright area, the collected microphone signal in the dark area is used as the expected signal corresponding to the reference signal.

[0143] Furthermore, in one embodiment, the first acquisition module is further used for:

[0144] The secondary sound system in the dark area plays white noise in turn and collects the microphone signal in the dark area for preprocessing. The normalized least mean square algorithm is used for system identification to obtain the secondary path filter in the dark area.

[0145] Furthermore, in one embodiment, the second acquisition module is further used for:

[0146] Obtaining an autocorrelation matrix of the filtered signal and a cross-correlation vector between the filtered signal and the delayed desired signal;

[0147] According to the above autocorrelation matrix and cross-correlation vector, the optimal control filter corresponding to the filter signal is obtained.

[0148] Among them, the functional implementation of each module in the above-mentioned optimal delay acquisition device corresponds to the various steps in the above-mentioned optimal delay acquisition device embodiment, and its functions and implementation processes are no longer repeated here.

[0149] In a third aspect, an embodiment of the present application provides an optimal delay acquisition device for sound field zoning control, and the optimal delay acquisition device can be a personal computer (PC), a laptop computer, a server, or other device with data processing capabilities.

[0150] Reference Figure 7 , Figure 7 Schematic diagram of the hardware structure of the optimal delay acquisition device involved in the embodiment of the present application. In the embodiment of the present application, the optimal delay acquisition device may include a processor, a memory, a communication interface and a communication bus.

[0151] The communication bus may be of any type and is used to interconnect the processor, the memory, and the communication interface.

[0152] The communication interface includes an input / output (I / O) interface, a physical interface, and a logical interface, etc., which are used to interconnect the devices inside the optimal delay acquisition device, and an interface used to interconnect the optimal delay acquisition device with other devices (such as other computing devices or user devices). The physical interface can be an Ethernet interface, an optical fiber interface, an ATM interface, etc.; the user device can be a display, a keyboard, etc.

[0153] The memory can be various types of storage media, such as random access memory (RAM), read-only memory (ROM), non-volatile RAM (NVRAM), flash memory, optical storage, hard disk, programmable ROM (PROM), erasable PROM (EPROM), electrically erasable PROM (EEPROM), etc.

[0154] The processor may be a general-purpose processor, and the general-purpose processor may call the optimal delay acquisition program stored in the memory and execute the optimal delay acquisition method provided in the embodiment of the present application. For example, the general-purpose processor may be a central processing unit (CPU). The method executed when the optimal delay acquisition program is called may refer to the various embodiments of the optimal delay acquisition method of the present application, which will not be repeated here.

[0155] Those skilled in the art will understand that Figure 7 The hardware structure shown in the figure does not constitute a limitation on the present application, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0156] In a fourth aspect, an embodiment of the present application also provides a computer-readable storage medium.

[0157] The computer-readable storage medium of the present application stores an optimal delay acquisition program, wherein when the above-mentioned optimal delay acquisition program is executed by a processor, the steps of the above-mentioned optimal delay acquisition method are implemented.

[0158] Among them, the method implemented when the optimal delay acquisition program is executed can refer to the various embodiments of the optimal delay acquisition method of the present application, and will not be repeated here.

[0159] It should be noted that the serial numbers of the above-mentioned embodiments of the present application are only for description and do not represent the advantages or disadvantages of the embodiments.

[0160] The terms "including" and "having" and any variations thereof in the specification and claims of this application and the above-mentioned drawings are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices. The terms "first", "second" and "third" are used to distinguish different objects, etc., and do not represent a sequence, nor do they limit "first", "second" and "third" to different types.

[0161] In the description of the embodiments of the present application, "exemplary", "for example" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary", "for example" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary", "for example" or "for example" is intended to present related concepts in a specific way.

[0162] In the description of the embodiments of the present application, unless otherwise specified, “ / ” means or, for example, A / B can mean A or B; the “and / or” in the text is merely a description of the association relationship of associated objects, indicating that three relationships may exist, for example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, in the description of the embodiments of the present application, “multiple” refers to two or more than two.

[0163] In some processes described in the embodiments of the present application, multiple operations or steps that appear in a specific order are included, but it should be understood that these operations or steps may not be executed in the order in which they appear in the embodiments of the present application or in parallel, and the sequence number of the operation is only used to distinguish the different operations, and the sequence number itself does not represent any execution order. In addition, these processes may include more or fewer operations, and these operations or steps may be executed in sequence or in parallel, and these operations or steps may be combined.

[0164] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, disk, CD) as described above, and includes a number of instructions for a terminal device to execute the methods described in each embodiment of the present application.

[0165] The above are only preferred embodiments of the present application, and are not intended to limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A method for obtaining an optimal delay for sound field partition control, characterized in that: The method comprises: Establishing an audio signal library, and obtaining an expected signal corresponding to each audio signal in the audio signal library; Obtaining each filtered signal obtained by each audio signal passing through the secondary path filter, and obtaining the optimal control filter corresponding to each filtered signal according to each filtered signal and the corresponding expected signal, thereby obtaining the theoretical noise reduction amount of each audio signal; The average value of each theoretical noise reduction amount is used as the cost function, and the golden section search is used to find the delay corresponding to the minimum cost function within the preset initial delay range as the optimal delay.

2. The optimal delay acquisition method for sound field partition control according to claim 1, characterized in that: The method of finding the delay corresponding to the minimum cost function by using the golden section search within the preset initial delay range specifically includes: Obtain two golden section points within the initial delay range, and obtain the average values ​​corresponding to the two golden section points; when the two average values ​​are different, the golden section point with the larger average value is taken as the first point, and the other golden section point is taken as the second point; when the two average values ​​are the same, any one of the golden section points is taken as the first point, and the other golden section point is taken as the second point; A new delay range is obtained by discarding the interval of the first point away from the second point in the initial delay range, taking the second point as a new golden section point of the new delay range, and calculating another new golden section point until the cost function is minimized and the corresponding delay is obtained.

3. The optimal delay acquisition method for sound field partition control according to claim 2, characterized in that: When the delay value of the first point is less than the delay value of the second point, the first point is used as the lower limit of the new delay range, and the product of the difference between the upper limit and the lower limit of the new delay range and 0.618 is obtained, and the sum of the product and the lower limit of the new delay range is used as another new golden section point; When the delay value of the first point is greater than the delay value of the second point, the first point is used as the upper limit of the new delay range, and the product of the difference between the upper limit and the lower limit of the new delay range and 0.382 is obtained, and the sum of the product and the lower limit of the new delay range is used as another new golden section point.

4. The optimal delay acquisition method for sound field partition control according to claim 2, characterized in that: The two golden section points μ within the initial delay range c and μ d They are: m c =μ a +0.382(m b -m a ) m d =μ a +0.618(m b -m a ) Among them, μ a is the lower limit of the initial delay range, μ b The upper limit of the initial delay range.

5. The optimal delay acquisition method for sound field partition control according to claim 1, characterized in that: Acquiring the expected signal corresponding to each audio signal in the audio signal library specifically includes: Determine the bright and dark areas based on the application scenario; Each audio signal is used as a reference signal, and when the reference signal is played with a delay in the bright area, the collected microphone signal in the dark area is used as the expected signal corresponding to the reference signal.

6. The optimal delay acquisition method for sound field partition control according to claim 5, characterized in that: Before obtaining each filtered signal obtained by each audio signal passing through the secondary path filter, the method further includes: The secondary sound system in the dark area plays white noise in turn and collects the microphone signal in the dark area for preprocessing. The normalized least mean square algorithm is used for system identification to obtain the secondary path filter in the dark area.

7. The optimal delay acquisition method for sound field partition control according to claim 1, characterized in that: According to any filtered signal and the corresponding expected signal, an optimal control filter corresponding to the filtered signal is obtained, specifically including: Obtaining an autocorrelation matrix of the filtered signal and a cross-correlation vector between the filtered signal and the delayed desired signal; According to the autocorrelation matrix and the cross-correlation vector, the optimal control filter corresponding to the filter signal is obtained.

8. An optimal delay acquisition device for sound field partition control, characterized in that: The device comprises: A first acquisition module, which is used to establish an audio signal library and acquire an expected signal corresponding to each audio signal in the audio signal library; A second acquisition module is used to obtain each filtered signal obtained by each audio signal through the secondary path filter, and obtain the optimal control filter corresponding to each filtered signal according to each filtered signal and the corresponding expected signal, so as to obtain the theoretical noise reduction amount of each audio signal; The golden section search module is used to use the average value of each theoretical noise reduction amount as the cost function, and use the golden section search to find the delay corresponding to the minimum cost function within a preset initial delay range as the optimal delay.

9. An optimal delay acquisition device for sound field partition control, characterized in that: The optimal delay acquisition device for sound field partition control includes a processor, a memory, and an optimal delay acquisition program stored in the memory and executable by the processor, wherein when the optimal delay acquisition program is executed by the processor, the steps of the optimal delay acquisition method as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores an optimal delay acquisition program, wherein when the optimal delay acquisition program is executed by a processor, the steps of the optimal delay acquisition method according to any one of claims 1 to 7 are implemented.

Images