Divergence detection method and device for hearing aid feedback elimination system

By performing frame-by-frame processing and frame-by-frame judgment of the microphone input signal and feedback signal of the hearing aid, the feedback path estimation error caused by the divergence of the adaptive filter is solved, and the stability of the hearing aid feedback elimination system and the improvement of the hearing aid feedback elimination system are achieved.

CN118945582APending Publication Date: 2024-11-12SHENZHEN JINGXINHAO TECH
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411192066.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-28
Publication Date
2024-11-12

AI Technical Summary

Technical Problem

The existing hearing aid feedback elimination system lacks an effective detection mechanism when the adaptive filter diverges, resulting in errors in feedback path estimation signals and inability to recover in time, which in turn aggravates the howling phenomenon and affects the auditory experience.

Method used

By performing frame-by-frame processing of the hearing aid microphone input signal and feedback signal, calculate the offset schedule frame by frame, and judge whether the adaptive filter is in a divergent state based on the preset threshold. If it is in a divergent state, reset the filter coefficient to reconverge, ensuring that Timely detection and correction.

Benefits of technology

It effectively reduces howling, improves the stability of the feedback elimination system, improves the user's auditory experience, and ensures the feedback suppression effect.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118945582A_ABST
    Figure CN118945582A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of adaptive filtering, and discloses a divergence detection method and device for a hearing aid feedback elimination system, and the method comprises the steps: carrying out the framing processing of an actual input signal of a hearing aid microphone and a feedback signal pre-estimated by the feedback elimination system; calculating the detuning degree of a feedback elimination system frame by frame based on the input signal after framing processing and a feedback signal estimated in advance; judging whether an adaptive filter of the feedback elimination system is in a divergent state or not based on the detuning degree calculated frame by frame and a preset threshold value; when the adaptive filter is in a divergent state, the adaptive filter coefficient is reset to reconverge the adaptive filter. The divergence state is determined by analyzing the imbalance degree and the preset threshold value, and once the divergence state is monitored, convergence is performed again by resetting the adaptive filter coefficient in the divergence state, so that the squeal phenomenon is reduced, the stability of the feedback elimination system is improved, and the auditory experience of a user is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of adaptive filtering technology, and in particular to a method and device for divergence detection of a hearing aid feedback elimination system. Background Art

[0002] Digital hearing aids convert acoustic signals into digital signals, and can use their built-in processors to process the collected signals before playing them back to improve listening comfort and clarity. However, in the process of using hearing aids, due to the positive feedback phenomenon in the hearing aids, that is, the microphone picks up the sound output by the speaker and amplifies it again, forming a closed-loop amplification system, which limits the maximum stable gain of the hearing aid, resulting in serious distortion of useful voice signals or feedback howling, which seriously affects the wearer's hearing experience.

[0003] In related technologies, adaptive feedback cancellation (AFC) technology is generally used to reduce the adverse effects of acoustic feedback caused by severe distortion of speech signals or feedback howling. The core of the AFC algorithm lies in the real-time adjustment of the adaptive filter. By real-time monitoring and adjustment of the adaptive filter coefficients of the hearing aid, the feedback signal is dynamically eliminated, thereby effectively suppressing the howling sound.

[0004] However, the AFC system may encounter divergence problems in actual applications, and the adaptive filter coefficients cannot converge stably. It can be understood that the current AFC system lacks an effective divergence detection mechanism, which causes the adaptive filter to be unable to recover quickly or even unable to recover when it diverges. When the adaptive filter begins to diverge, it is not detected and corrected in time, and the estimated feedback path is far from the actual feedback path, resulting in the continued existence of the erroneous feedback path estimation signal, which ultimately leads to the failure of feedback suppression and even aggravates the howling phenomenon, thereby generating noise. Summary of the invention

[0005] In view of this, the present invention provides a method and apparatus for divergence detection in a hearing aid feedback cancellation system to solve the problem of how to prevent an adaptive filter from estimating an erroneous feedback path, thereby improving the stability of the feedback cancellation system.

[0006] In a first aspect, the present invention provides a method for divergence detection of a hearing aid feedback cancellation system, the method comprising:

[0007] The actual input signal of the hearing aid microphone and the feedback signal estimated in advance by the feedback elimination system are processed by framing respectively;

[0008] Based on the input signal after frame processing and the pre-estimated feedback signal, calculating the detuning of the feedback elimination system frame by frame;

[0009] Based on the out-of-control calculated frame by frame and a preset threshold, determining whether the adaptive filter of the feedback elimination system is in a divergent state;

[0010] When the adaptive filter is in a divergent state, the adaptive filter coefficients are reset to make the adaptive filter converge again.

[0011] This embodiment processes the input signal and the feedback signal in frames so that the signal of each frame can be analyzed independently, and judges the state of the adaptive filter frame by frame based on the out-of-control and preset threshold value, thereby ensuring the timeliness of the adaptive filter detecting the divergent state, and determines the divergent state by analyzing the out-of-control and preset threshold value. Once the divergent state is detected, the adaptive filter coefficients in the divergent state are reset and re-converged, thereby reducing the howling phenomenon. Compared with the related art that lacks a timely detection and correction mechanism, resulting in feedback path estimation signal errors, this embodiment effectively improves the feedback suppression effect of the hearing aid, improves the stability of the feedback elimination system, and thus improves the user's auditory experience.

[0012] In an optional implementation, judging whether the adaptive filter of the feedback elimination system is in a divergent state based on the out-of-control calculated frame by frame and a preset threshold includes:

[0013] Determine whether the out-of-order value of the current frame is less than a first threshold value, and obtain a comparison result; wherein the comparison result includes a first comparison result and a second comparison result, the first comparison result indicates that the out-of-order value is less than the first threshold value, and the second comparison result indicates that the out-of-order value is greater than or equal to the first threshold value;

[0014] After obtaining the comparison result of each frame, the first comparison result is counted by the first counter, or the second comparison result is counted by the second counter; wherein, if the comparison result of the previous frame of the current frame is inconsistent with the out-of-order comparison result of the current frame, the value of the counter corresponding to the previous frame is cleared;

[0015] It is determined whether the adaptive filter corresponding to the current frame is in a divergent state according to the value of the first counter or the value of the second counter.

[0016] In this embodiment, the mismatch is calculated and compared frame by frame, and the comparison result of each frame is recorded. A counter is used as one of the judgment conditions for the state of the adaptive filter. The state is judged only when the comparison results of several consecutive frames are consistent. The final counter result has only one case, which provides a judgment basis for the specific analysis of the subsequent divergent state, so that the filter can clearly and effectively distinguish between the convergent state and the non-convergent state.

[0017] In an optional implementation, the actual input signal includes a sound source signal; judging whether the adaptive filter corresponding to the current frame is in a divergent state according to the value of the first counter specifically includes:

[0018] If the value of the first counter is greater than or equal to the second threshold, it is confirmed that the adaptive filter corresponding to the current frame is in a convergence completion state;

[0019] When the adaptive filter corresponding to the current frame is in a convergence completion state, determining a first ratio or a second ratio based on the output signal of the speaker and the sound source signal; wherein the first ratio represents the ratio of the energy of the output signal of the speaker to the energy of the sound source signal; and the second ratio represents the ratio of the amplitude of the output signal of the speaker to the amplitude of the sound source signal;

[0020] Determining whether any one of the first ratio or the second ratio is greater than or equal to a fourth threshold;

[0021] When either the first ratio or the second ratio is greater than or equal to a fourth threshold, it is confirmed that the adaptive filter corresponding to the current frame is in a divergent state.

[0022] In the embodiment of the present invention, the value of the first counter is compared with the second threshold to determine whether the adaptive filter has entered the convergence completion state. The coefficient of the filter has stabilized in the minimum error state. At this time, it is impossible to confirm whether it has diverged. Therefore, in the filter convergence completion state, the ratio of the energy and amplitude of the speaker output signal to the energy and amplitude of the sound source signal is calculated, and compared with the fourth threshold to determine whether it has diverged. If any item in the ratio exceeds the set fourth threshold, it is directly and accurately identified whether it has diverged in an unknown state, so that corresponding processing can be made in time.

[0023] In an optional implementation, judging whether the adaptive filter corresponding to the current frame is in a divergent state according to the value of the second counter specifically includes:

[0024] If the value of the second counter is greater than or equal to the third threshold, it is confirmed that the adaptive filter corresponding to the current frame is in a divergent state.

[0025] In the embodiment of the present invention, the out-of-control value of the current frame is greater than or equal to the first threshold. The value of the second counter is compared with the third threshold to determine whether the adaptive filter is converging or diverging. If it exceeds the set third threshold, it is directly determined that the filter is in a divergent state so that corresponding processing can be made in time.

[0026] In an optional implementation, judging whether the adaptive filter corresponding to the current frame is in a divergent state according to the value of the first counter specifically includes:

[0027] If the value of the first counter is greater than or equal to the second threshold, it is confirmed that the adaptive filter corresponding to the current frame is in a convergence completion state;

[0028] When the adaptive filter corresponding to the current frame is in a convergence completion state, the broadband gain of the forward path is obtained; based on the broadband gain and the first ratio, a first difference is determined; or based on the broadband gain and the second ratio, a second difference is determined; wherein the first ratio represents the ratio of the energy of the output signal of the loudspeaker to the energy of the sound source signal; the second ratio represents the ratio of the amplitude of the output signal of the loudspeaker to the amplitude of the sound source signal; the first difference represents the difference between the first ratio and the broadband gain, and the second difference represents the difference between the second ratio and the broadband gain;

[0029] Determine whether any one of the first difference or the second difference is greater than or equal to a fourth threshold;

[0030] When either the first difference or the second difference is greater than or equal to a fourth threshold, it is confirmed that the adaptive filter corresponding to the current frame is in a divergent state.

[0031] In the embodiment of the present invention, the value of the first counter is compared with the second threshold to determine whether the adaptive filter has entered the convergence completion state, and the coefficient of the filter has stabilized in the minimum error state. At this time, it is impossible to confirm whether it has diverged. Therefore, in the filter convergence completion state, the ratio of the energy and amplitude of the speaker output signal to the energy and amplitude of the sound source signal and the difference in broadband gain are calculated, and compared with the fourth threshold to determine whether it has diverged. If any item in the ratio exceeds the set fourth threshold, it is determined that it exceeds the upper and lower acceptable range of the broadband gain, giving a possibility of divergence, thereby accurately identifying whether it has diverged in an unknown state, so as to facilitate timely corresponding processing.

[0032] In an optional embodiment, the method further includes:

[0033] When the adaptive filter is in a divergent state, the values ​​of the first counter and the second counter are reset.

[0034] The embodiment of the present invention avoids misjudgment due to cumulative errors by resetting the values ​​of the first counter and the second counter. After the filter starts to converge again, a new judgment cycle is required, thereby effectively ensuring the accuracy of the judgment result.

[0035] In a second aspect, the present invention provides a device for divergence detection of a hearing aid feedback cancellation system, the device comprising:

[0036] A frame processing module, used for performing frame processing on the actual input signal of the hearing aid microphone and the feedback signal estimated in advance by the feedback elimination system respectively;

[0037] A feedback cancellation system out-of-step calculation module, used to calculate the out-of-step of the feedback cancellation system frame by frame based on the input signal after frame processing and the pre-estimated feedback signal;

[0038] A filter divergence state judgment module, used to judge whether the adaptive filter of the feedback elimination system is in a divergence state based on the misalignment calculated frame by frame and a preset threshold;

[0039] The adaptive filter coefficient resetting module is used to reset the adaptive filter coefficients when the adaptive filter is in a divergent state, so as to make the adaptive filter converge again.

[0040] In a third aspect, the present invention provides a computer device, comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the method for divergence detection of a hearing aid feedback elimination system according to the first aspect or any corresponding embodiment thereof by executing the computer instructions.

[0041] In a fourth aspect, the present invention provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to cause a computer to execute the method for divergence detection of a hearing aid feedback cancellation system of the first aspect or any corresponding embodiment thereof.

[0042] In a fifth aspect, the present invention provides a computer program product, comprising computer instructions for causing a computer to execute the method for divergence detection in a hearing aid feedback cancellation system according to the first aspect or any corresponding embodiment thereof. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0044] Figure 1 It is a schematic diagram of an acoustic feedback elimination method of an adaptive filter in the related art;

[0045] Figure 2 is a flow chart of a method for divergence detection of a hearing aid feedback elimination system according to an embodiment of the present invention;

[0046] Figure 3 is a flow chart of a method for divergence detection of another hearing aid feedback elimination system according to an embodiment of the present invention;

[0047] Figure 4 is a flow chart of a method for divergence detection of another hearing aid feedback elimination system according to an embodiment of the present invention;

[0048] Figure 5 is a spectrum diagram of output signal simulation before improvement of the method for divergence detection of the hearing aid feedback elimination system according to an embodiment of the present invention;

[0049] Figure 6 is a spectrum diagram of output signal simulation after the improved method for divergence detection of the hearing aid feedback elimination system according to an embodiment of the present invention;

[0050] Figure 7 is a structural block diagram of a device for divergence detection of a hearing aid feedback elimination system according to an embodiment of the present invention;

[0051] Figure 8 It is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0052] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0053] In the field of hearing aid technology, digital hearing aids convert acoustic signals into digital signals, and use built-in processors to process the signals before playing them back to improve listening comfort and clarity. In general, adaptive feedback cancellation (AFC) technology is used to reduce the adverse effects of acoustic feedback. The core of the AFC algorithm lies in the real-time adjustment of the adaptive filter. The traditional implementation of acoustic feedback cancellation based on the adaptive filter is as follows: Figure 1 As shown in Figure 1, G(n) is the electroacoustic forward path, which is used to provide the gain of the hearing aid; U(n) is the acoustic feedback path, which means that the sound output by the speaker is collected again by the microphone; W(n) is the feedback path estimated by the adaptive filter.

[0054] The main workflow of AFC is as follows:

[0055] (1) Calculate the error signal: the difference between the microphone reference signal (actual input signal) and the estimated feedback signal:

[0056] e(n)=d(n)-y(n);

[0057] Wherein, d(n)=s(n)+f(n) represents the actual signal received by the microphone, f(n) is the real feedback signal, s(n) is the sound source signal, and y(n) is the estimated feedback signal.

[0058] (2) Filter coefficient update: Use an adaptive algorithm to update the filter coefficients to minimize the error signal. Current technology usually uses the normalized least mean square (NLMS) algorithm to update the filter coefficients, and its update formula is:

[0059]

[0060] Wherein, w(n) represents the coefficient vector of the adaptive filter at time n, x(n) is the loudspeaker signal obtained by delaying and amplifying the error signal using the forward path, that is, x(n) is also the input vector of the adaptive filter at time n, μ is the step size factor, δ is a correction factor greater than zero to prevent the denominator from being zero, and ||·|| represents the norm operator.

[0061] (3) Estimated feedback path modeling: Dynamically adjust the filter to accurately estimate the signal in the feedback path:

[0062] y(n)=w T (n)x(n);

[0063] By adjusting the adaptive filter, an estimated feedback signal y(n) is generated to offset the feedback signal f(n) of the feedback path, thereby suppressing the feedback.

[0064] In hearing aids, there are three main performance indicators of the AFC system: feedback residual, convergence time, and comfort. Feedback residual: also called steady-state error, refers to the presence of unprocessed feedback signals in the sound source signal after being processed by the AFC system. The smaller the feedback residual, the better. Convergence time: the time required to adjust the filter coefficient from the initial state until the output error or coefficient change reaches a certain stable state. This stable state is usually defined as the output error falling within a preset threshold range, and the fluctuation amplitude within this range is very small, or the rate of change of the filter coefficient is reduced to a negligible level. Comfort: refers to the comfort of the patient using the hearing aid. This performance indicator can only be evaluated from a subjective perspective.

[0065] The AFC system algorithm dynamically eliminates feedback signals by real-time monitoring and adjusting the filter coefficients of the hearing aid, thereby effectively suppressing howling and improving sound quality and user experience. However, in the process of using hearing aids, due to the positive feedback phenomenon in the hearing aids, that is, the microphone picks up the sound output by the speaker and amplifies it again, forming a closed-loop amplification system. In this case, the maximum stable gain of the hearing aid is limited, resulting in serious distortion of the useful voice signal or causing feedback howling, which not only seriously affects the wearer's auditory experience, but may also cause users to be dissatisfied with the hearing aids and refuse to use them.

[0066] This embodiment provides a method for divergence detection of a hearing aid feedback cancellation system, which is applied to various application scenarios where feedback howling occurs. By utilizing the unbalanced control and threshold, the state of the filter is analyzed frame by frame. Once the divergence state is detected, it is immediately re-converged to avoid the adaptive filter estimating the wrong feedback path, reduce the howling phenomenon, and thus improve the stability of the feedback cancellation system.

[0067] According to an embodiment of the present invention, a method embodiment of divergence detection of a hearing aid feedback elimination system is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0068] In this embodiment, a method for divergence detection of a hearing aid feedback cancellation system is provided, which can be used in the above-mentioned computer. Figure 2 is a flow chart of a method for divergence detection of a hearing aid feedback elimination system according to an embodiment of the present invention. Figure 2 As shown, the process includes the following steps:

[0069] Step S201 : performing frame processing on the actual input signal of the hearing aid microphone and the feedback signal estimated in advance by the feedback elimination system respectively.

[0070] It should be noted that frame processing refers to dividing a continuous signal into multiple smaller frames. By processing the signal frame by frame, it can be ensured that when the adaptive filter is in a divergent state, it can be detected and controlled in time. The input signal refers to the actual sound signal from the hearing aid microphone. The feedback signal refers to the feedback path signal pre-estimated by the feedback cancellation system. During the use of the hearing aid, the feedback cancellation (AFC) system will pre-amplify the input signal to provide the broadband gain required for hearing aid. And in a real environment, the input signal also includes a noise signal, which is not limited in the simulation process of the present invention.

[0071] It is understandable that the estimated feedback path signal is not the same as the actual feedback path signal of the hearing aid microphone. Only the feedback path signal pre-estimated by the feedback cancellation system can provide a data basis for the subsequent mismatch calculation.

[0072] Specifically, by using digital signal processing (DSP) technology, the input signal and the feedback signal are divided into frames, and each frame can be analyzed independently to ensure that when detecting the state of the adaptive filter, whether it is divergent can be confirmed in time.

[0073] Step S202, based on the input signal after frame processing and the pre-estimated feedback signal, the detuning of the feedback elimination system is calculated frame by frame.

[0074] It should be noted that the out-of-control refers to the degree to which the feedback cancellation system deviates from the optimal state. By comparing the out-of-control calculated frame by frame with the preset threshold, it is possible to determine whether the adaptive filter diverges, and the moment of divergence can be determined more accurately, thereby avoiding further aggravation of the divergence phenomenon.

[0075] Specifically, the imbalance of the adaptive filter is calculated by the feedback path signal in the input signal and the pre-estimated feedback signal. The smaller the imbalance, the smaller the estimation error, and the closer the feedback path estimated by the adaptive filter is to the real feedback path. At the same time, the faster the imbalance decreases, the faster the convergence speed. The definition of imbalance is expressed by the following formula:

[0076]

[0077] Where u(n) represents the actual feedback path in the input signal, w(n) represents the feedback path estimated by the adaptive filter, and ||·|| represents the norm operator.

[0078] Step S203: judging whether the adaptive filter of the feedback elimination system is in a divergent state based on the out-of-control calculated frame by frame and a preset threshold.

[0079] It is understandable that in the initial adjustment stage of the adaptive filter, the misalignment is usually high because the filter coefficients are far from the optimal value. At this time, the feedback cancellation (AFC) system has not yet converged and is in a convergence state, with large errors. As the adaptive filter continues to adjust its coefficients, the filter coefficients gradually approach the actual feedback path, the error gradually decreases, and the misalignment also gradually decreases. When the misalignment reaches a relatively low and stable value, the AFC system enters a stable state. At this time, the output of the adaptive filter is highly matched with the expected output, and the system error remains at a low level.

[0080] If the uncontrolled fluctuations are large during or after the convergence process, it may indicate that the AFC system is unstable or has diverged.

[0081] It is worth noting that the adaptive filter dynamically adjusts the filter coefficients during the entire hearing aid use process to adapt to the changing feedback path. However, since the relevant technology cannot effectively detect divergence, when the adaptive filter begins to diverge, it is not detected and corrected in time. The estimated feedback path is far different from the actual feedback path, resulting in the continued existence of the erroneous feedback path estimation signal, which ultimately leads to the failure of feedback suppression and even aggravates the howling phenomenon, thereby generating noise and affecting the effect of the hearing aid.

[0082] Therefore, in theory, this embodiment determines the stability of the adaptive filter by detuning, and then combines the preset threshold to analyze whether it is in a divergent state under different circumstances, so as to more accurately distinguish the divergence between the stable state and the unstable state.

[0083] Step S204: when the adaptive filter is in a divergent state, the adaptive filter coefficients are reset to make the adaptive filter converge again.

[0084] It should be noted that the divergence state refers to the inability of the adaptive filter to correctly track the feedback path, resulting in an increase in system error. Resetting the filter coefficients means restoring the coefficients of the adaptive filter to the initial state and re-converging, which is equivalent to interrupting the divergence process of the adaptive filter at the moment the divergence begins, thereby minimizing the howling phenomenon.

[0085] Specifically, resetting the filter aims to reset the coefficients of the filter, reset all coefficients of the adaptive filter to initial values ​​(set to zero), and reinitialize any related state variables or buffers.

[0086] This embodiment processes the input signal and the feedback signal in frames so that the signal of each frame can be analyzed independently, and judges the state of the adaptive filter frame by frame based on the out-of-control and preset threshold value, thereby ensuring the timeliness of the adaptive filter detecting the divergent state, and determines the divergent state by analyzing the out-of-control and preset threshold value. Once the divergent state is detected, the adaptive filter coefficients in the divergent state are reset and re-converged, thereby reducing the howling phenomenon. Compared with the related art that lacks a timely detection and correction mechanism, resulting in feedback path estimation signal errors, this embodiment effectively improves the feedback suppression effect of the hearing aid, improves the stability of the feedback elimination system, and thus improves the user's auditory experience.

[0087] In this embodiment, a method for divergence detection of a hearing aid feedback cancellation system is provided, which can be used in the above-mentioned computer. Figure 3 is a flow chart of a method for divergence detection of a hearing aid feedback elimination system according to an embodiment of the present invention. Figure 3 As shown, the process includes the following steps:

[0088] Step S301, the actual input signal of the hearing aid microphone and the feedback signal estimated in advance by the feedback elimination system are processed by frame division. Figure 1 Step S101 of the illustrated embodiment will not be described in detail here.

[0089] Step S302, based on the input signal after frame processing and the pre-estimated feedback signal, the detuning of the feedback elimination system is calculated frame by frame. Figure 1 Step S102 of the illustrated embodiment will not be described in detail here.

[0090] Step S303: judging whether the adaptive filter of the feedback elimination system is in a divergent state based on the out-of-control calculated frame by frame and a preset threshold.

[0091] Specifically, the above step S303 includes:

[0092] Step S3031, determine whether the out-of-order value of the current frame is less than a first threshold, and obtain a comparison result; wherein the comparison result includes a first comparison result and a second comparison result, the first comparison result indicates that the out-of-order value is less than the first threshold, and the second comparison result indicates that the out-of-order value is greater than or equal to the first threshold.

[0093] Step S3032, after obtaining the comparison result of each frame, count the first comparison result through the first counter, or count the second comparison result through the second counter; wherein, if the comparison result of the previous frame of the current frame is inconsistent with the out-of-scheduling comparison result of the current frame, clear the value of the counter corresponding to the previous frame.

[0094] Step S3033: determine whether the adaptive filter corresponding to the current frame is in a divergent state according to the value of the first counter or the value of the second counter.

[0095] It should be noted that the first threshold is a standard threshold set according to actual experiments, indicating that in this state, the hearing aid is in an intermediate state.

[0096] Specifically, when setting the first threshold, this embodiment provides two methods: a statistical analysis method and an experimental debugging method.

[0097] Method 1: Set the threshold by statistical analysis of historical data. The specific steps are as follows: Run an adaptive filter, collect the out-of-synchronization data under different conditions, including the out-of-synchronization changes in a noise-free environment, different signal powers, different noise levels, different ambient noises, etc. Calculate the mean (μ) and standard deviation (σ) of these data. For example, calculate the mean and standard deviation of the out-of-synchronization after convergence. Set the threshold based on statistical characteristics. Usually, the threshold can be set as the mean plus several times the standard deviation. A common choice is μ + 3σ. The specific choice can be adjusted according to application requirements.

[0098] Method 2: Set the threshold through experiments and debugging. The specific steps are as follows: Set an initial threshold based on experience or the parameters of a similar system. Run the system, observe the change in the out-of-synchronization of the filter, and record the out-of-synchronization of the system under different conditions. Adjust the threshold according to the experimental results to ensure that the system can accurately distinguish the convergence state and the non-convergence state during actual operation.

[0099] Exemplarily, for example, calculate the out-of-synchronization (Mis) frame by frame, and determine whether the adaptive filter is in a divergent state according to a preset first threshold (Th1). The first threshold (Th1) is 0.5, and the out-of-synchronization (Mis) of the first five frames is 0.3, 0.6, 0.4, 0.7, 0.8. The first comparison result indicates Mis < Th1, and the second comparison result indicates Mis >= Th1.

[0100] Among them, Frame 1: Mis = 0.3, and the comparison result is the first comparison result (Mis < Th1). Frame 2: Mis = 0.6, and the comparison result is the second comparison result (Mis >= Th1). Frame 3: Mis = 0.4, and the comparison result is the first comparison result (Mis < Th1). Frame 4: Mis = 0.7, and the comparison result is the second comparison result (Mis >= Th1). Frame 5: Mis = 0.8, and the comparison result is the second comparison result (Mis >= Th1).

[0101] That is to say, starting from Frame 4, the first counter corresponding to the first comparison result is cleared. At this time, the out-of-synchronization has exceeded the threshold, and the filter becomes unstable. The second counter corresponding to the second comparison result starts timing. Until the calculation of Frame 5 is completed, at this time, the value of the first counter is 0, and the value of the second counter is 2.

[0102] In this embodiment, the out-of-synchronization is calculated and compared frame by frame, and the comparison result of each frame is recorded. The counter is used as one of the judgment conditions for the state of the adaptive filter. Only when the comparison results of several consecutive frames are consistent, the state will be judged. The final counter result has only one situation, providing a judgment basis for the specific analysis of the subsequent divergent state, so that the filter can clearly and effectively distinguish the convergence state and the non-convergence state.

[0103] In an optional embodiment, the actual input signal includes a sound source signal, and step S3033 determines whether the adaptive filter corresponding to the current frame is in a divergent state according to the value of the first counter, specifically including:

[0104] A1: If the value of the first counter is greater than or equal to a second threshold, it is confirmed that the adaptive filter corresponding to the current frame is in a convergence completion state.

[0105] A2. When the adaptive filter corresponding to the current frame is in a convergence completion state, a first ratio or a second ratio is determined based on the output signal of the speaker and the sound source signal; wherein the first ratio represents the ratio of the energy of the output signal of the speaker to the energy of the sound source signal; and the second ratio represents the ratio of the amplitude of the output signal of the speaker to the amplitude of the sound source signal.

[0106] A3: Determine whether any one of the first ratio or the second ratio is greater than or equal to a fourth threshold.

[0107] A4: When any one of the first ratio or the second ratio is greater than or equal to a fourth threshold, it is confirmed that the adaptive filter corresponding to the current frame is in a divergent state.

[0108] It should be noted that the setting of the first threshold and the second threshold are both related to the filter length. For example, if the filter length is 128, the threshold is set to 128, so as to ensure that there are enough sampling points to evaluate the state of the filter. The threshold is set according to the filter length, and the judgment result is the most representative, thereby reducing the situation of misjudgment.

[0109] In addition, the setting of the first threshold and the second threshold is also related to the convergence speed of the adaptive filter. The setting of the threshold needs to be combined with the convergence speed of the filter under normal working conditions. If the threshold is too low, the filter will be judged as converged when it is not fully converged. If the threshold is too high, the correct judgment of the filter state will be delayed and timely analysis cannot be performed.

[0110] Taking the contents of steps S3031 to 3033 as an example, if the current frame is frame 3, the value of the first counter is 3, and the second threshold is set to 128, the result is that the second threshold is not reached, and the adaptive filter is in the convergence state. At this time, there must be no divergence. At this time, since the result of A1 has not reached the convergence completion state, steps A2 to A4 are skipped. However, if the value of the first counter is 129 in the 300th frame, when the previous frame number is continuous, it is greater than 128. At this time, the adaptive filter has converged, so further judgment of steps A2 to A4 is required.

[0111] Furthermore, the fourth threshold is required for judgment and analysis in steps A2 to A4. Under ideal conditions, the feedback path is estimated as an unbiased estimate, that is, assuming that the feedback is completely eliminated, then the energy ratio R is equal to the broadband gain K, and the ratio of the speaker output signal energy to the sound source signal energy should be equal to the amplification factor K. In actual situations, due to the presence of acoustic feedback and noise, the energy ratio R will be greater than the broadband gain K. Therefore, the fourth threshold should be set to a value slightly larger than the amplification factor K to reflect the actual acoustic feedback and noise effects. The specific setting value is adjusted according to the actual situation.

[0112] Specifically, the output of the adaptive filter is e(n), and the hearing aid needs to amplify e(n) to provide sufficient gain. The actual output signal of the speaker is obtained through the forward path G(n), x(n) = G(n)e(n). Assume that the broadband gain of the forward path G(n) is K, and s(n) is the sound source signal. The judgment process is:

[0113] (1) The short-time energy of the speaker's output signal is calculated frame by frame using the following formula:

[0114] E[x(n)]=||x(n)|| 2 ;

[0115] (2) The short-time energy of the sound source signal input to the microphone is calculated frame by frame using the following formula:

[0116] E[s(n)]=||s(n)|| 2 ;

[0117] (3) The ratio of the short-time energy of the loudspeaker output signal to the short-time energy of the sound source signal is calculated using the following formula:

[0118] R = E[x(n)] / E[s(n)];

[0119] (4) A fourth threshold Th4 is set. If the energy ratio R is greater than Th4, the adaptive filter is determined to be in a divergent state. Otherwise, the adaptive filter is determined to be in a stable state or a normal state. The subsequent process is processed according to the actual flow after convergence.

[0120] In the embodiment of the present invention, the value of the first counter is compared with the second threshold to determine whether the adaptive filter has entered the convergence completion state. The coefficient of the filter has stabilized in the minimum error state. At this time, it is impossible to confirm whether it has diverged. Therefore, in the filter convergence completion state, the ratio of the energy and amplitude of the speaker output signal to the energy and amplitude of the sound source signal is calculated, and compared with the fourth threshold to determine whether it has diverged. If any item in the ratio exceeds the set fourth threshold, it is directly and accurately identified whether it has diverged in an unknown state, so that corresponding processing can be made in time.

[0121] In an optional embodiment, the actual input signal includes a sound source signal, and step S3033 determines whether the adaptive filter corresponding to the current frame is in a divergent state according to the value of the first counter, specifically including:

[0122] B1: If the value of the first counter is greater than or equal to a second threshold, it is confirmed that the adaptive filter corresponding to the current frame is in a convergence completion state.

[0123] B2. When the adaptive filter corresponding to the current frame is in a convergence completion state, the broadband gain of the forward path is obtained; based on the broadband gain and the first ratio, a first difference is determined; or based on the broadband gain and the second ratio, a second difference is determined; wherein the first ratio represents the ratio of the energy of the output signal of the loudspeaker to the energy of the sound source signal; the second ratio represents the ratio of the amplitude of the output signal of the loudspeaker to the amplitude of the sound source signal; the first difference represents the difference between the first ratio and the broadband gain, and the second difference represents the difference between the second ratio and the broadband gain.

[0124] B3: Determine whether any one of the first difference or the second difference is greater than or equal to a fourth threshold.

[0125] B4: When any one of the first difference or the second difference is greater than or equal to a fourth threshold, it is confirmed that the adaptive filter corresponding to the current frame is in a divergent state.

[0126] It should be noted that the process of judging the divergent state in steps B1 to B4 can refer to steps A1 to A4 and will not be repeated here. However, in this embodiment, since the comparison and judgment are made by the first difference or the second difference and the information of the broadband gain is combined, the fourth threshold of B1 to B4 is different from the fourth threshold setting of steps A1 to A4 and should be pre-set to a standard deviation value. If this standard deviation value is exceeded, it indicates that it exceeds the upper and lower acceptable range of the broadband gain, and it is directly determined that the adaptive filter corresponding to the current frame is in a divergent state.

[0127] In the embodiment of the present invention, the value of the first counter is compared with the second threshold to determine whether the adaptive filter has entered the convergence completion state, and the coefficient of the filter has stabilized in the minimum error state. At this time, it is impossible to confirm whether it has diverged. Therefore, in the filter convergence completion state, the ratio of the energy and amplitude of the speaker output signal to the energy and amplitude of the sound source signal and the difference in broadband gain are calculated, and compared with the fourth threshold to determine whether it has diverged. If any item in the ratio exceeds the set fourth threshold, it is determined that it exceeds the upper and lower acceptable range of the broadband gain, giving a possibility of divergence, thereby accurately identifying whether it has diverged in an unknown state, so as to facilitate timely corresponding processing.

[0128] In an optional implementation, step S3033 determines whether the adaptive filter corresponding to the current frame is in a divergent state according to the value of the second counter, specifically including:

[0129] C1: If the value of the second counter is greater than or equal to a third threshold, it is confirmed that the adaptive filter corresponding to the current frame is in a divergent state.

[0130] It should be noted that if the out-of-control value of the current frame is greater than or equal to the first threshold, the value of the second counter exists at this time, and the value of the first counter is reset. The adaptive filter is in an unstable state, and there are only two states, one is converging and the other is diverging. Therefore, it is only necessary to compare the value of the second counter with the third threshold to determine whether it is in a divergent state. The setting of the third threshold is similar to the aforementioned second threshold, and will not be repeated here.

[0131] In the embodiment of the present invention, the out-of-control value of the current frame is greater than or equal to the first threshold. The value of the second counter is compared with the third threshold to determine whether the adaptive filter is converging or diverging. If it exceeds the set third threshold, it is directly determined that the filter is in a divergent state so that corresponding processing can be made in time.

[0132] Step S304: When the adaptive filter is in a divergent state, the adaptive filter coefficients are reset to make the adaptive filter converge again. Figure 1 Step S104 of the illustrated embodiment will not be described in detail here.

[0133] In an optional implementation, the method for divergence detection of the hearing aid feedback cancellation system provided in this embodiment further includes: when the adaptive filter is in a divergent state, resetting the values ​​of the first counter and the second counter.

[0134] It is understandable that when the adaptive filter is reset, the filter starts to converge again. Therefore, a new round of judgment is required, and the values ​​of the counters are reset to the initial values ​​accordingly, so as to ensure that judgment is made again in the subsequent new convergence process.

[0135] In the embodiment of the present invention, the values ​​of the first counter and the second counter are reset to avoid misjudgment due to cumulative errors. After the filter starts to converge again, a new judgment cycle is required, thereby effectively ensuring the accuracy of the judgment result.

[0136] This embodiment provides an overall flow chart of a method for divergence detection of a hearing aid feedback elimination system, which can be referred to as Figure 4 ,Firstly the two signals are framed and then the mismatched MIS of the signals are calculated frame by frame.

[0137] If MIS is less than TH1, COUNT1 is counted. If COUNT1 is less than COUNT2, the next stitch offset is recalculated. If COUNT1 is greater than or equal to TH2, the energy ratio is calculated. If the ratio is less than TH4, the filter is reset. If the ratio is greater than or equal to TH4, the normal process is carried out.

[0138] If MIS is greater than or equal to TH1, COUNT2 is counted. If COUNT2 is less than TH3, the next frame mismatch is recalculated. If COUNT2 is greater than or equal to TH3, the filter is reset.

[0139] When the divergence state detection is performed without the divergence detection method of the hearing aid feedback elimination system provided by this embodiment, the effect presented by the spectrum diagram of the output signal simulation is as follows: Figure 5 As shown in the figure, the horizontal axis represents time, the vertical axis represents frequency, and the color represents the strength of the signal. As time goes by, the energy at 1000Hz and 1200Hz in the blue box is high and there is acoustic feedback. This indicates that the filter diverges and the AFC system fails.

[0140] After the divergence state detection is performed by the divergence detection method of the hearing aid feedback elimination system provided in this embodiment, the effect presented by the spectrum diagram of the output signal simulation is as follows: Figure 6 As shown, the acoustic feedback is effectively eliminated, and the acoustic feedback at 1000Hz and 1200Hz in the blue box is effectively eliminated, effectively controlling the divergence of the filter. It is worth noting that the gain of the AFC system cannot be infinite, otherwise huge noise and acoustic feedback will be generated when the hearing aid is started.

[0141] In this embodiment, a device for divergence detection of a hearing aid feedback elimination system is also provided, and the device is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made are not repeated here. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0142] This embodiment provides a device for detecting divergence of a hearing aid feedback cancellation system. Figure 7 As shown, including:

[0143] A frame processing module 701 is used to perform frame processing on the actual input signal of the hearing aid microphone and the feedback signal pre-estimated by the feedback elimination system respectively;

[0144] A feedback cancellation system out-of-step calculation module 702, configured to calculate the out-of-step of the feedback cancellation system frame by frame based on the input signal after frame processing and the pre-estimated feedback signal;

[0145] A filter divergence state judgment module 703 is used to judge whether the adaptive filter of the feedback elimination system is in a divergence state based on the misalignment calculated frame by frame and a preset threshold;

[0146] The adaptive filter coefficient resetting module 704 is used to reset the adaptive filter coefficients when the adaptive filter is in a divergent state, so as to make the adaptive filter converge again.

[0147] In an optional implementation, the filter divergence state determination module 703 includes:

[0148] An out-of-order comparison unit is used to determine whether the out-of-order of the current frame is less than a first threshold value, and obtain a comparison result; wherein the comparison result includes a first comparison result and a second comparison result, the first comparison result indicates that the out-of-order is less than the first threshold value, and the second comparison result indicates that the out-of-order is greater than or equal to the first threshold value;

[0149] A comparison result counting unit, used for counting the first comparison result through the first counter after obtaining the comparison result of each frame, or counting the second comparison result through the second counter; wherein, if the comparison result of the previous frame of the current frame is inconsistent with the out-of-order comparison result of the current frame, the value of the counter corresponding to the previous frame is cleared;

[0150] The filter state judgment unit is used to judge whether the adaptive filter corresponding to the current frame is in a divergent state according to the value of the first counter or the value of the second counter.

[0151] In an optional implementation manner, the filter state determination unit specifically includes:

[0152] a counter value comparison subunit, configured to confirm that the adaptive filter corresponding to the current frame is in a convergence completion state if the value of the first counter is greater than or equal to a second threshold;

[0153] A ratio determination subunit, configured to determine a first ratio or a second ratio based on the output signal of the loudspeaker and the sound source signal when the adaptive filter corresponding to the current frame is in a convergence completion state; wherein the first ratio represents the ratio of the energy of the output signal of the loudspeaker to the energy of the sound source signal; and the second ratio represents the ratio of the amplitude of the output signal of the loudspeaker to the amplitude of the sound source signal;

[0154] a ratio determination subunit, configured to determine whether any one of the first ratio or the second ratio is greater than or equal to a fourth threshold;

[0155] The filter state confirmation subunit is used to confirm that the adaptive filter corresponding to the current frame is in a divergent state when any one of the first ratio or the second ratio is greater than or equal to a fourth threshold.

[0156] In an optional implementation manner, the filter state determination unit specifically includes:

[0157] The filter state confirmation subunit is used to confirm that the adaptive filter corresponding to the current frame is in a divergent state if the value of the second counter is greater than or equal to a third threshold.

[0158] In an optional implementation manner, the filter state determination unit specifically includes:

[0159] a counter value comparison subunit, configured to confirm that the adaptive filter corresponding to the current frame is in a convergence completion state if the value of the first counter is greater than or equal to a second threshold;

[0160] A difference determination subunit, used for obtaining the broadband gain of the forward path when the adaptive filter corresponding to the current frame is in a convergence completion state; determining a first difference based on the broadband gain and the first ratio; or determining a second difference based on the broadband gain and the second ratio; wherein the first ratio represents the ratio of the energy of the output signal of the loudspeaker to the energy of the sound source signal; the second ratio represents the ratio of the amplitude of the output signal of the loudspeaker to the amplitude of the sound source signal; the first difference represents the difference between the first ratio and the broadband gain, and the second difference represents the difference between the second ratio and the broadband gain;

[0161] a ratio determination subunit, configured to determine whether any one of the first difference or the second difference is greater than or equal to a fourth threshold;

[0162] The filter state confirmation subunit is used to confirm that the adaptive filter corresponding to the current frame is in a divergent state when any one of the first difference value or the second difference value is greater than or equal to a fourth threshold value.

[0163] In an optional embodiment, the device further comprises:

[0164] The counter resetting module is used to reset the values ​​of the first counter and the second counter when the adaptive filter is in a divergent state.

[0165] The further functional description of each of the above modules and units is the same as that of the above corresponding embodiments and will not be repeated here.

[0166] The device for divergence detection of the hearing aid feedback elimination system in this embodiment is presented in the form of a functional unit, where the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that executes one or more software or fixed programs, and / or other devices that can provide the above functions.

[0167] The embodiment of the present invention also provides a computer device having the above Figure 7 An apparatus for divergence detection in a hearing aid feedback cancellation system is shown.

[0168] See also Figure 8 , Figure 8 is a schematic diagram of the structure of a computer device provided by an optional embodiment of the present invention, such as Figure 8 As shown, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Various components are connected to each other using different buses for communication, and can be installed on a common mainboard or installed in other ways as needed. The processor can process the instructions executed in the computer device, including instructions stored in or on the memory to display the graphical information of the GUI on an external input / output device (such as, a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Figure 8 A processor 10 is taken as an example.

[0169] The processor 10 may be a central processing unit, a network processor or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.

[0170] The memory 20 stores instructions executable by at least one processor 10, so that at least one processor 10 executes the method shown in the above embodiment.

[0171] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the computer device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely arranged relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0172] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid state drive; the memory 20 may also include a combination of the above types of memory.

[0173] The computer device further comprises a communication interface 30 for the computer device to communicate with other devices or a communication network.

[0174] The embodiment of the present invention also provides a computer-readable storage medium. The method according to the embodiment of the present invention can be implemented in hardware, firmware, or can be implemented as a computer code that can be recorded in a storage medium, or can be implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and will be stored in a local storage medium through a network download, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state hard disk, etc.; further, the storage medium can also include a combination of the above types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor, or hardware, the method shown in the above embodiment is implemented.

[0175] A part of the present invention may be applied as a computer program product, such as a computer program instruction, which, when executed by a computer, can call or provide the method and / or technical solution according to the present invention through the operation of the computer. Those skilled in the art should understand that the existence of the computer program instruction in a computer-readable medium includes, but is not limited to, a source file, an executable file, an installation package file, etc., and accordingly, the way in which the computer program instruction is executed by the computer includes, but is not limited to: the computer directly executes the instruction, or the computer compiles the instruction and then executes the corresponding compiled program, or the computer reads and executes the instruction, or the computer reads and installs the instruction and then executes the corresponding installed program. Here, the computer-readable medium may be any available computer-readable storage medium or communication medium accessible to the computer.

[0176] Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations are all within the defined scope.

Claims

1. A method for detecting divergence of a hearing aid feedback cancellation system, characterized in that: The method comprises: The actual input signal of the hearing aid microphone and the feedback signal estimated in advance by the feedback elimination system are processed by framing respectively; Based on the input signal after frame processing and the pre-estimated feedback signal, calculating the detuning of the feedback elimination system frame by frame; Based on the out-of-control calculated frame by frame and a preset threshold, determining whether the adaptive filter of the feedback elimination system is in a divergent state; When the adaptive filter is in a divergent state, the adaptive filter coefficients are reset to make the adaptive filter converge again.

2. The method according to claim 1, characterized in that: The step of judging whether the adaptive filter of the feedback elimination system is in a divergent state based on the out-of-control calculated frame by frame and a preset threshold value includes: Determine whether the out-of-order value of the current frame is less than a first threshold value, and obtain a comparison result; wherein the comparison result includes a first comparison result and a second comparison result, the first comparison result indicates that the out-of-order value is less than the first threshold value, and the second comparison result indicates that the out-of-order value is greater than or equal to the first threshold value; After obtaining the comparison result of each frame, the first comparison result is counted by the first counter, or the second comparison result is counted by the second counter; wherein, if the comparison result of the previous frame of the current frame is inconsistent with the out-of-order comparison result of the current frame, the value of the counter corresponding to the previous frame is cleared; Whether the adaptive filter corresponding to the current frame is in a divergent state is determined according to the value of the first counter or the value of the second counter.

3. The method according to claim 2, characterized in that The actual input signal includes a sound source signal; judging whether the adaptive filter corresponding to the current frame is in a divergent state according to the value of the first counter, specifically includes: If the value of the first counter is greater than or equal to the second threshold, it is confirmed that the adaptive filter corresponding to the current frame is in a convergence completion state; When the adaptive filter corresponding to the current frame is in a convergence completion state, determining a first ratio or a second ratio based on the output signal of the speaker and the sound source signal; wherein the first ratio represents the ratio of the energy of the output signal of the speaker to the energy of the sound source signal; and the second ratio represents the ratio of the amplitude of the output signal of the speaker to the amplitude of the sound source signal; Determining whether any one of the first ratio or the second ratio is greater than or equal to a fourth threshold; When either the first ratio or the second ratio is greater than or equal to a fourth threshold, it is confirmed that the adaptive filter corresponding to the current frame is in a divergent state.

4. The method according to claim 2, characterized in that: Determining whether the adaptive filter corresponding to the current frame is in a divergent state according to the value of the second counter specifically includes: If the value of the second counter is greater than or equal to the third threshold, it is confirmed that the adaptive filter corresponding to the current frame is in a divergent state.

5. The method according to claim 2, characterized in that: Determining whether the adaptive filter corresponding to the current frame is in a divergent state according to the value of the first counter specifically includes: If the value of the first counter is greater than or equal to the second threshold, it is confirmed that the adaptive filter corresponding to the current frame is in a convergence completion state; When the adaptive filter corresponding to the current frame is in a convergence completion state, the broadband gain of the forward path is obtained; based on the broadband gain and the first ratio, a first difference is determined; or based on the broadband gain and the second ratio, a second difference is determined; wherein the first ratio represents the ratio of the energy of the output signal of the loudspeaker to the energy of the sound source signal; the second ratio represents the ratio of the amplitude of the output signal of the loudspeaker to the amplitude of the sound source signal; the first difference represents the difference between the first ratio and the broadband gain, and the second difference represents the difference between the second ratio and the broadband gain; Determining whether any one of the first difference or the second difference is greater than or equal to a fourth threshold; When either the first difference or the second difference is greater than or equal to a fourth threshold, it is confirmed that the adaptive filter corresponding to the current frame is in a divergent state.

6. The method according to any one of claims 2 to 5, characterized in that: The method further comprises: When the adaptive filter is in a divergent state, the values ​​of the first counter and the second counter are reset.

7. A device for detecting divergence of a hearing aid feedback cancellation system, characterized in that: The device comprises: A frame processing module, used for performing frame processing on the actual input signal of the hearing aid microphone and the feedback signal estimated in advance by the feedback elimination system respectively; A feedback cancellation system out-of-step calculation module, used to calculate the out-of-step of the feedback cancellation system frame by frame based on the input signal after frame processing and the pre-estimated feedback signal; A filter divergence state judgment module, used to judge whether the adaptive filter of the feedback elimination system is in a divergence state based on the misalignment calculated frame by frame and a preset threshold; The adaptive filter coefficient resetting module is used to reset the adaptive filter coefficients when the adaptive filter is in a divergent state, so as to make the adaptive filter converge again.

8. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the method for divergence detection of a hearing aid feedback elimination system according to any one of claims 1 to 6 by executing the computer instructions.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the method for divergence detection of a hearing aid feedback cancellation system according to any one of claims 1 to 6.

10. A computer program product, characterized in that The method comprises computer instructions for causing a computer to execute the method for divergence detection of a hearing aid feedback cancellation system according to any one of claims 1 to 6.