A method and device for pre-detecting whistling, a method and device for whistling control

Abstract

The present invention provides a pre-whistling detection method for an active noise reduction device, which includes: receiving a noise signal collected by an audio acquisition device and a secondary sound source driving signal output by a noise reduction device; calculating the signal power of the noise signal and the secondary sound source driving signal at the current moment; performing binarization processing on the noise signal and the secondary sound source driving signal to obtain a binarized noise signal and a binarized secondary sound source driving signal; calculating the phase difference between the binarized noise signal and the binarized secondary sound source driving signal, and aligning the phases of the binarized noise signal and the binarized secondary sound source driving signal according to the phase difference; performing a correlation operation on the phase-aligned binarized noise signal and binarized secondary sound source driving signal to obtain the audio signal correlation value at the current moment; if the audio signal correlation value at the current moment is greater than the audio signal correlation threshold, and the signal power of any one of the noise signal and the secondary sound source driving signal at the current moment is greater than the power threshold, it is determined that whistling will occur.

Description

Technical Field

[0001] The present invention relates to acoustic processing technologies, and particularly to a pre-whistle detection method and device, and a whistle control method and device. Background Art

[0002] Noise not only affects people's normal life, but is even harmful to hearing. As noise pollution becomes more and more serious, technologies for suppressing noise have received more attention from researchers. Active noise cancellation is based on the principle of waveform interference, and generates a waveform with the same amplitude and opposite phase as the noise through a secondary path (such as a speaker) for cancellation. When an active noise-canceling headphone performs noise cancellation and transparency processing, the signal output by the speaker leaks to the reference microphone, forming a feedback system. This phenomenon is called acoustic feedback. When certain conditions are met (gain, phase), it causes whistling, and whistling is a sharp and harsh noise that will bring noise interference to the user and the surrounding environment of the speaker, greatly affecting the user experience.

[0003] In traditional technical solutions, when whistling occurs, the power of a certain frequency that meets the whistling conditions increases rapidly until the maximum power output. Since environmental noise and voice signals are mostly broadband signals, the detection method of whistling can be simplified to the detection of a single-frequency signal.

[0004] For example, Figure 1 is a schematic diagram of a hybrid active noise-canceling headphone system in the prior art, including a reference microphone 101, an error microphone 103, an active noise-canceling processing module 105, and a speaker 107. The reference microphone 101 collects external signals, and after being processed by the active noise-canceling processing module 105, it is sent to the speaker 107 for playback to form a secondary sound source. The secondary sound source cancels out the external signal (noise) to form a noise-canceling effect. The secondary sound source and the external signal are complementary to form a transparency effect. The error microphone 103 monitors the noise-canceling error in real time, and further cancels the noise-canceling error through a feedback control loop. In addition to interfering with the external audio, a part of the sound output by the speaker 107 also propagates to the reference microphone 101, and this part is called acoustic feedback. When the open-loop gain F*C formed by the frequency response F of the acoustic feedback path and the frequency response C of the active noise-canceling processing meets certain conditions, whistling will be formed. There are also feedforward active noise cancellation (only including a reference microphone) and feedback active noise cancellation (only including an error microphone) in the prior art.

[0005] In order to prevent and control whistling, it is necessary to detect the prerequisite signal for whistling generation - a single-frequency signal, that is, pre-whistle detection. Figure 2 is the data of the speaker 107 and the reference microphone 101 when whistling occurs. The power of the single-frequency signal increases exponentially, so the pre-whistle detection can be simplified to the detection of a single-frequency signal.

[0006] In the existing detection technology, it is considered that detecting a high-power single-frequency signal is a howling, and then subsequent howling suppression is carried out to control the howling. For example, the time-domain signal is converted into a frequency-domain signal, and the frequency-domain values are monitored and detected. When the power of a certain frequency exceeds a preset value, a howling flag bit is triggered. This method can effectively detect howling, but the operation is complex and the howling detection delay is large, resulting in howling control only after howling is detected. Therefore, the best processing method is to predict howling, that is, preventing howling from forming is an urgent problem to be solved at the speaker end.

[0007] The traditional technical solutions have the following drawbacks:

[0008] 1. It is necessary to convert the collected time-domain signal to the frequency domain for calculation, with a large amount of calculation;

[0009] 2. The howling detection delay is large. Generally, howling can only be detected after it appears, and howling cannot be processed in time. Summary of the Invention

[0010] Based on the above situation, the main object of the present invention is to provide a howling pre-detection and control method and device with a small amount of calculation and fast detection.

[0011] To achieve the above object, the technical solution adopted by the present invention is as follows:

[0012] A howling pre-detection method for an active noise reduction device, the active noise reduction device includes an audio acquisition device and a noise reduction device. The audio acquisition device is used to acquire an environmental noise signal, and the noise reduction device is used to perform operations on the noise signal acquired by the audio acquisition device to obtain a secondary sound source drive signal. The secondary sound source drive signal is provided to the audio playback device for playback to cancel the environmental noise signal and thus achieve active noise reduction.

[0013] The howling pre-detection method includes the steps of:

[0014] Receiving the noise signal acquired by the audio acquisition device and the secondary sound source drive signal output by the noise reduction device;

[0015] Calculating the signal power of the noise signal and the secondary sound source drive signal at the current moment;

[0016] Performing binarization processing on the noise signal and the secondary sound source drive signal to obtain a binarized noise signal and a binarized secondary sound source drive signal;

[0017] Calculating the phase difference between the binarized noise signal and the binarized secondary sound source drive signal, and aligning the phases of the binarized noise signal and the binarized secondary sound source drive signal according to the phase difference;

[0018] Perform a correlation operation on the binarized noise signal and the binarized secondary sound source drive signal after phase alignment to obtain the correlation value of the audio signal at the current moment;

[0019] When the correlation value of the audio signal at the current moment is greater than a preset audio signal correlation threshold, and the signal power of any one of the noise signal and the secondary sound source drive signal at the current moment is greater than a preset power threshold, it is determined that there is a single-frequency signal at the current moment, and the power of the single-frequency signal increases at the current moment, which will generate a howl. Among them, the single-frequency signal is the single-frequency point signal that will generate a howl.

[0020] Preferably, the noise signal is the noise time-domain audio amplitude value collected by the audio acquisition device, and the secondary sound source drive signal is the time-domain audio amplitude value of the secondary sound source drive signal output by the noise reduction device.

[0021] Preferably, the binarization process of the noise signal and the secondary sound source drive signal includes:

[0022] When the noise time-domain audio amplitude value is greater than 0, the binarized noise signal is 1; otherwise, the binarized noise signal is -1;

[0023] When the time-domain audio amplitude value of the secondary sound source drive signal is greater than 0, the binarized secondary sound source drive signal is 1; otherwise, the binarized secondary sound source drive signal is -1.

[0024] Preferably, calculating the phase difference between the binarized noise signal and the binarized secondary sound source drive signal, and aligning the phases of the binarized noise signal and the binarized secondary sound source drive signal according to the phase difference includes:

[0025] Calculate the zero-crossing points of the binarized noise signal and the binarized secondary sound source drive signal, and obtain the phase difference between the binarized noise signal and the binarized secondary sound source drive signal according to the difference of the zero-crossing points;

[0026] Control the caching of the binarized noise signal and / or the binarized secondary sound source drive signal according to the phase difference to achieve the phase alignment of the binarized noise signal and the binarized secondary sound source drive signal.

[0027] Preferably, performing a correlation operation on the binarized noise signal and the binarized secondary sound source drive signal after phase alignment to obtain the correlation value of the audio signal at the current moment includes:

[0028] The relevant value at the current moment is the relevant value at the previous moment plus the product of the binary noise signal and the binary secondary sound source drive signal at the current moment, and then minus the product of the binary noise signal and the binary secondary sound source drive signal at the L-th moment before, where L is the length value of the relevant operation.

[0029] Preferably, the noise signal includes a feedforward noise signal and / or a feedback noise signal.

[0030] The feedforward noise signal is the noise signal outside the active noise reduction device; the feedback noise signal is the noise signal of the external noise signal entering the active noise reduction device.

[0031] When the noise signal is a feedforward noise signal and the feedback noise signal,

[0032] Calculate the signal power of the feedforward noise signal, the feedback noise signal, and the secondary sound source drive signal at the current moment.

[0033] Perform binary processing on the feedforward noise signal, the feedback noise signal, and the secondary sound source drive signal to obtain a binary feedforward noise signal, a binary feedback noise signal, and a binary secondary sound source drive signal.

[0034] Calculate the phase difference between the binary feedforward noise signal and the binary secondary sound source drive signal, and align the phases of the binary feedforward noise signal and the binary secondary sound source drive signal according to the phase difference; calculate the phase difference between the binary feedback noise signal and the binary secondary sound source drive signal, and align the phases of the binary feedback noise signal and the binary secondary sound source drive signal according to the phase difference.

[0035] Perform a correlation operation on the phase-aligned binary feedforward noise signal and the binary secondary sound source drive signal to obtain the relevant value of the feedforward audio signal at the current moment; perform a correlation operation on the phase-aligned binary feedback noise signal and the binary secondary sound source drive signal to obtain the relevant value of the feedback audio signal at the current moment.

[0036] When any one of the relevant values of the feedforward audio signal and the feedback audio signal at the current moment is greater than a preset audio signal correlation threshold, and the signal power of any one of the feedforward noise signal, the feedback noise signal, and the secondary sound source drive signal at the current moment is greater than a preset power threshold, it is determined that there is a single-frequency signal at the current moment, and the power of the single-frequency signal increases at the current moment, resulting in a howling sound.

[0037] The present invention also provides a howling control method for an active noise reduction device, which includes an audio acquisition device and a noise reduction device. The howling pre-detection method of the present invention is used to detect whether howling will occur. When it is detected that howling will occur at the current moment, the gain value of the secondary sound source is reduced to control howling.

[0038] The present invention also provides a howling pre-detection device, which includes an audio acquisition module, an active noise reduction module, a power detection module, a correlation operation module and a control module.

[0039] The audio acquisition module is used to acquire a noise signal.

[0040] The active noise reduction module is used to perform an operation on the noise signal acquired by the audio acquisition device to obtain a secondary sound source drive signal.

[0041] The power detection module is used to calculate the signal power of the noise signal and the secondary sound source drive signal at the current moment.

[0042] The correlation operation module includes a binarization calculation unit, a phase alignment unit and a correlation calculation unit.

[0043] The binarization calculation unit performs binarization processing on the noise signal and the secondary sound source drive signal to obtain a binarized noise signal and a binarized secondary sound source drive signal.

[0044] The phase alignment unit calculates the phase difference between the binarized noise signal and the binarized secondary sound source drive signal, and aligns the phases of the binarized noise signal and the binarized secondary sound source drive signal according to the phase difference.

[0045] The correlation calculation unit performs a correlation operation on the binarized noise signal and the binarized secondary sound source drive signal after phase alignment to obtain the audio signal correlation value at the current moment.

[0046] The control module is used to determine that there is a single-frequency signal at the current moment and the single-frequency signal increases and howling will occur when the audio signal correlation value at the current moment is greater than a preset audio signal correlation threshold and the signal power value of any one of the noise signal and the secondary sound source drive signal at the current moment is greater than a preset power threshold, where the single-frequency signal is the single-frequency point signal of the howling that will occur.

[0047] Preferably, the noise signal acquired by the audio acquisition module is a time-domain audio amplitude value, and the secondary sound source drive signal output by the active noise reduction module is a time-domain audio amplitude value.

[0048] Preferably, the binarization calculation unit includes a first judgment unit and a second judgment unit.

[0049] The first determination unit is used to perform binarization processing on the noise signal. When the time-domain audio amplitude value of the noise is greater than 0, the binarized noise signal is 1; otherwise, the binarized noise signal is -1.

[0050] The second determination unit is used to perform binarization processing on the secondary sound source drive signal. When the time-domain audio amplitude value of the secondary sound source drive signal is greater than 0, the binarized secondary sound source drive signal is 1; otherwise, the binarized secondary sound source drive signal is -1.

[0051] Preferably, the phase alignment unit includes a zero-crossing detection unit, a first buffer unit, and a second buffer unit.

[0052] The zero-crossing detection unit calculates the zero-crossing points of the binarized noise signal and the binarized secondary sound source drive signal, and obtains the phase difference between the binarized noise signal and the binarized secondary sound source drive signal according to the difference of the zero-crossing points.

[0053] The first buffer unit is used to buffer the binarized noise signal.

[0054] The second buffer unit is used to buffer the binarized secondary sound source drive signal.

[0055] The zero-crossing detection unit also controls the storage of the binarized noise signal in the first buffer unit and / or controls the storage of the binarized secondary sound source drive signal in the second buffer unit according to the phase difference, so as to align the phases of the binarized noise signal and the binarized secondary sound source drive signal.

[0056] Preferably, the correlation calculation unit calculates the correlation value at the current moment as

[0057] The correlation value at the previous moment plus the product of the binarized noise signal and the binarized secondary sound source drive signal at the current moment, and then subtracts the product of the binarized noise signal and the binarized secondary sound source drive signal at the L-th moment before, where the L is the length of the correlation calculation.

[0058] Preferably, the audio acquisition device includes a front-feed microphone and / or a rear-feed microphone. The front-feed microphone is arranged outside the howling pre-detection device and is used to collect the noise signal outside the howling pre-detection device. The rear-feed microphone is arranged inside the howling pre-detection device and is used to collect the noise signal of the external noise signal entering the inside of the howling pre-detection device.

[0059] The present invention also provides a howling control device, which includes the howling pre-detection device of the present invention, and further includes a gain value update module.

[0060] When the howling pre-detection device detects that howling will occur at the current moment, the control module configures a gain update value to the gain value update module to reduce the gain value of the secondary sound source signal.

[0061] Preferably, the gain value update unit is a digital gain value update unit or an analog gain value update unit.

[0062] The present invention also provides an active noise reduction chip that can execute the howling pre-detection method or the howling control method of the present invention.

[0063] The present invention also provides an active noise reduction earphone, including the howling pre-detection device of the present invention, or the howling control device of the present invention, or the active noise reduction chip of the present invention.

[0064] The present invention also provides a storage medium that stores a program, wherein the program is used to be executed to implement the howling pre-detection method or the howling control method of the present invention.

[0065] The howling pre-detection method of the present invention performs correlation processing on the data of the audio acquisition device and the noise reduction device, and at the same time combines power indicators to quickly determine whether a howling signal will be generated. Binaryize the noise signal and the secondary sound source drive signal, then align the phases of the binaryized noise signal and the secondary sound source drive signal, and then calculate the correlation value between the binaryized noise signal and the binaryized secondary sound source drive signal after phase alignment, which can reduce the data storage amount, reduce the operation amount of howling detection, and ensure the howling detection effect. The detection is fast and effective, and the howling pre-signal can be captured before the howling is formed, thereby suppressing the generation of howling.

[0066] The howling control method of the present invention adopts the howling pre-detection method of the present invention. If it is determined that howling will occur at the current moment, the gain value of the secondary sound source is reduced, so that the output signal of the audio playback device can be reduced, and thus howling can be controlled.

[0067] The howling pre-detection device of the present invention, by adding a power detection module and a correlation operation module, after collecting the noise signal and the secondary sound source drive signal, the power detection module calculates the power of the corresponding sound source signal, and at the same time the correlation operation module calculates the correlation value between the noise signal and the secondary sound source drive signal, which can quickly determine whether a howling signal will be generated. Binaryize the noise signal and the secondary sound source drive signal, then align the phases of the binaryized noise signal and the secondary sound source drive signal, and then calculate the correlation value between the binaryized noise signal and the binaryized secondary sound source drive signal after phase alignment, which can reduce the data storage amount, reduce the operation amount of howling detection, and ensure the howling detection effect. The detection is fast and effective, and the howling pre-signal can be captured before the howling is formed, thereby suppressing the generation of howling.

[0068] The howling control device of the present invention includes the howling pre-detection device of the present invention, and further includes a gain value update module. When it is detected that howling will occur at the current moment, a gain update value is configured to the gain value update module to reduce the gain value of the audio playback module, so as to reduce the secondary sound source signal, thereby achieving the effect of controlling the generation of howling.

[0069] The active noise reduction chip of the present invention executes the howling pre-detection method or the howling control method of the present invention, which can reduce the computational complexity of the howling detection of the chip and ensure the howling detection effect.

[0070] The active noise reduction earphone of the present invention adopts the howling pre-detection device or the howling control device of the present invention, realizes the prediction and early control of howling, prevents the formation of howling, and improves the user experience.

[0071] Other beneficial effects of the present invention will be described in the specific implementation manners through the introduction of specific technical features and technical solutions. Those skilled in the art should be able to understand the beneficial technical effects brought by the said technical features and technical solutions through the introduction of these technical features and technical solutions. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] The preferred embodiments of the howling pre-detection method and device, and the howling control method and device according to the present invention will be described below with reference to the accompanying drawings. In the figures:

[0073] Figure 1 is a schematic diagram of a hybrid active noise reduction system in the prior art;

[0074] Figure 2 is Figure 1 the output data curves of the speaker and the reference microphone when howling occurs in

[0075] Figure 3 is a flowchart of a howling pre-detection method according to a preferred embodiment of the present invention;

[0076] Figure 4 is the phase output after zero-crossing detection of the binary noise signal and the binary secondary sound source drive signal according to a preferred embodiment of the present invention;

[0077] Figure 5 is a schematic flowchart of processing the noise signal and the secondary sound source drive signal according to a preferred embodiment of the present invention;

[0078] Figure 6 is a simulation result diagram of the howling pre-detection method according to a preferred embodiment of the present invention;

[0079] Figure 7 is a flowchart of a howling control method according to a preferred embodiment of the present invention;

[0080] Figure 8 Block diagram of a howling pre-detection device according to a preferred embodiment of the present invention;

[0081] Figure 9 is Figure 8 Schematic diagram of the structure of the phase alignment unit in

[0082] Figure 10 Block diagram of a howling control device according to a preferred embodiment of the present invention;

[0083] Figure 11 Block diagram of a howling pre-detection circuit according to a preferred embodiment of the present invention;

[0084] Figure 12 Block diagram of a howling control circuit according to a preferred embodiment of the present invention;

[0085] Figure 13 Block diagram of a howling control circuit according to another preferred embodiment of the present invention. Specific embodiments

[0086] Figure 3 Flowchart 200 of a howling pre-detection method according to a preferred embodiment of the present invention. This howling pre-detection method is used for active noise reduction devices, such as active noise reduction headphones, active noise reduction microphones, active noise reduction speakers, etc. The microphones and speakers of these devices are usually in the same environment.

[0087] The above active noise reduction device generally includes an audio acquisition device and a noise reduction device. The audio acquisition device is used to acquire ambient noise signals, and the noise reduction device is used to calculate a secondary sound source drive signal based on the noise signals acquired by the audio acquisition device. The secondary sound source drive signal is provided to an audio playback device for playback to cancel the ambient noise signals, thereby achieving active noise reduction. The howling pre-detection method includes: Step 201, receiving the noise signals acquired by the audio acquisition device and the secondary sound source drive signal output by the noise reduction device; Step 203, calculating the signal power of the noise signals and the secondary sound source drive signal at the current moment; Step 205, performing binarization processing on the noise signals and the secondary sound source drive signal to obtain binarized noise signals and binarized secondary sound source drive signals; Step 207, calculating the phase difference between the binarized noise signals and the binarized secondary sound source drive signal, and aligning the phases of the binarized noise signals and the binarized secondary sound source signals according to the phase difference; Step 209, performing a correlation operation on the binarized noise signals and the binarized secondary sound source drive signal with aligned phases to obtain the audio signal correlation value at the current moment; Step 211, when the audio signal correlation value at the current moment is greater than a preset audio signal correlation threshold, and the signal power value of any one of the noise signals and the secondary sound source drive signal at the current moment is greater than a preset power threshold, execute Step 213, determine that there is a single-frequency signal at the current moment, and the power of the single-frequency signal at the current moment increases, and determine that howling will occur at the current moment, where the single-frequency signal is the single-frequency point signal of the howling that will occur.

[0088] The audio signal correlation threshold and the power threshold are usually given based on test experience. Among them, the power threshold may include a noise signal power threshold and / or a secondary sound source signal power threshold, and the two may be the same value or different values, both of which are given based on test experience.

[0089] According to the howling pre-detection method of the present invention, since the single-frequency signal has strong correlation and a very high signal-to-noise ratio when howling occurs, the data acquired by the audio acquisition device and the data output by the noise reduction device are subjected to correlation processing, and combined with the power index, it is possible to quickly determine whether a howling signal will be generated. Binarizing the noise signals and the secondary sound source drive signal, then aligning the phases of the binarized noise signals and the secondary sound source drive signal, and then calculating the correlation value of the noise signals and the secondary sound source drive signal with aligned phases can reduce the data storage amount, reduce the computational amount of howling detection, and ensure the howling detection effect. The detection is fast and effective, and the howling pre-signal can be captured before the howling is formed, thereby suppressing the generation of howling.

[0090] In one embodiment, the noise signal of the audio acquisition device may be the time-domain audio amplitude value of the noise acquired by the audio acquisition device, and the secondary sound source drive signal of the noise reduction device may be the time-domain audio amplitude value of the secondary sound source drive signal output by the noise reduction device. In the existing detection technologies, the time-domain signal is usually converted into a frequency-domain signal, and the frequency-domain values are monitored and detected. When the power of a certain frequency exceeds a preset value, a howling flag bit is triggered. This method requires collecting the time-domain signal and converting it to the frequency domain for calculation, and the operation is complex. In this embodiment, the time-domain audio amplitude value of the audio acquisition device and the time-domain audio amplitude value of the secondary sound source drive signal output by the noise reduction device are directly processed and calculated, reducing the operation complexity.

[0091] In one embodiment, the binarization process of the noise signal and the secondary sound source drive signal may be as follows: when the time-domain audio amplitude value of the audio acquisition device is greater than 0, the binarized noise signal is 1; otherwise, the binarized noise signal is -1; when the time-domain audio amplitude value of the noise reduction device is greater than 0, the binarized secondary sound source drive signal is 1; otherwise, the binarized secondary sound source drive signal is -1. Binarization usually means quantifying the time-domain audio amplitude value of the audio acquisition device and then quantizing it to 1 bit, and quantizing the time-domain audio amplitude value of the noise reduction device to 1 bit. Through this process, the storage amount and the amount of calculation can be further reduced, the operation speed can be increased, and thus the howling pre-detection sensitivity can be improved.

[0092] In a specific embodiment, the digital sign function (sign) may be used to binarize the noise signal and the secondary sound source drive signal.

[0093] In one embodiment, calculating the phase difference between the binarized noise signal and the binarized secondary sound source drive signal may be: calculating the zero-crossing points of the binarized noise signal and the binarized secondary sound source drive signal, and obtaining the phase difference between the binarized noise signal and the binarized secondary sound source drive signal according to the difference of the zero-crossing points. Figure 4 The phase output after the zero-crossing detection of the binarized noise signal and the binarized secondary sound source drive signal is shown. It can be seen that there is a certain phase difference between the two. Further, the caching of the binarized noise signal and / or the binarized secondary sound source drive signal may be controlled according to the phase difference to align the phases of the binarized noise signal and the binarized secondary sound source drive signal. Aligning the phases before the relevant operations ensures the maximum relevant value output, thus ensuring the sensitivity of the howling detection. The traditional method of phase alignment is a phase detector, but the operation process is complex. In the present invention, the zero-crossing detection method is used to align the phases of the binarized noise signal and the binarized secondary sound source drive signal, with a small amount of calculation, further improving the sensitivity of the howling pre-detection.

[0094] In one embodiment, the following formula may be used to calculate the correlation value between the binarized noise signal and the binarized secondary sound source drive signal.

[0095]

[0096] Among them, x is the noise signal, and y is the secondary sound source driving signal.

[0097] x = [x(k), x(k - 1), x(k - 2), ……, x(k - L + 1)],

[0098] y = [y(k), y(k - 1), y(k - 2), ……, y(k - L + 1)].

[0099] k is the sampling time point, and L is the length of the correlation operation, which is also the number of sampling points of the noise signal and the secondary sound source driving signal;

[0100] Since the values of both x and y are 1 and -1, the denominator in the above formula for calculating the correlation value of the binary noise signal and the binary secondary sound source driving signal becomes a constant. To further simplify the operation, for example, the divider can be omitted in the logic circuit, and the above formula is simplified to:

[0101]

[0102] Among them, x b is the binary noise signal, and y b is the binary secondary sound source driving signal.

[0103] In one embodiment, in order to further reduce the RAM operation and the amount of computation, the correlation operation can use a recursive form to perform the correlation value operation of the binary noise signal and the binary secondary sound source driving signal. Specifically, the correlation value at the current moment is the correlation value at the previous moment plus the product of the binary noise signal and the binary secondary sound source driving signal at the current moment, and then subtract the product of the binary noise signal and the binary secondary sound source driving signal at the previous moment.

[0104] When the audio acquisition device is a reference microphone, the correlation operation is to calculate the correlation coefficient between the noise signal collected by the reference microphone and the secondary sound source driving signal output by the noise reduction device . As Figure 5 shown, it is a schematic flowchart of the processing of the noise signal and the secondary sound source driving signal .

[0105] First, in order to reduce the complexity, the noise signal and the secondary sound source driving signal are binarized, as shown in the following formula:

[0106]

[0107]

[0108] After the above formula calculation, when > 0, the binary noise signal x b (k) is 1, otherwise, the binary noise signal x b (k) is -1. When > 0, the binary secondary sound source drive signal y b (k) is 1, otherwise, the binary secondary sound source drive signal y b (k) is -1. The noise signal and the secondary sound source drive signal are both binary-processed.

[0109] After the binary processing and before the relevant operations, it is necessary to align the phases to ensure the maximum relevant value output. The zero-crossing detection can be performed on the binary noise signal and the binary secondary sound source drive signal. By recording the zero-crossing positions of the binary noise signal and the binary secondary sound source drive signal and comparing the differences in the zero-crossing positions, the delay between the reference microphone data and the speaker data can be obtained. The delay information is used to control the buffer module to ensure the data is aligned in phase.

[0110] The relevant operation can use a recursive form to calculate the correlation value between the binary noise signal and the binary secondary sound source drive signal, as shown in the following formula.

[0111]

[0112] Among them, the correlation value at time n is the correlation value at time n-1 plus the product of the binary noise signal and the binary secondary sound source drive signal at the current time, and then subtracts the product of the binary noise signal and the binary secondary sound source drive signal at the Lth previous time. L represents the length of the correlation operation, that is, the number of sampling points in a single correlation operation. The initial correlation value can be set to 0.

[0113] Active noise cancellation usually includes feedforward active noise cancellation, feedback active noise cancellation, and hybrid active noise cancellation. Hybrid active noise cancellation includes both feedforward noise cancellation and feedback noise cancellation. The technical solution of the present invention can be applied to any of the above active noise cancellation methods.

[0114] In one embodiment, the audio acquisition device can be a feedforward microphone in an active noise cancellation system, or a feedback microphone, or a feedforward microphone and an error microphone. The feedforward microphone is usually arranged outside the active noise cancellation device and collects the noise signal outside the device. The feedback microphone is usually arranged inside the active noise cancellation device and collects the noise signal after the external environmental noise enters the active noise cancellation device.

[0115] When the audio acquisition device is a feedforward microphone, the acquired noise signal is a feedforward noise signal, that is, the noise signal outside the active noise cancellation device; when the audio acquisition device is a feedback microphone, the acquired noise signal is a feedback noise signal, that is, the noise signal after the external environmental noise enters the active noise cancellation device.

[0116] When the audio acquisition device is a feedforward microphone and a feedback microphone, step 203 is: calculating the signal power of the feedforward noise signal, the feedback noise signal, and the secondary sound source drive signal at the current moment; step 205 is: performing binarization processing on the feedforward noise signal, the feedback noise signal, and the secondary sound source drive signal to obtain a binarized feedforward noise signal, a binarized feedback noise signal, and a binarized secondary sound source drive signal; step 207 is: calculating the phase difference between the binarized feedforward noise signal and the binarized secondary sound source drive signal, and aligning the phases of the binarized feedforward noise signal and the binarized secondary sound source drive signal according to the phase difference; calculating the phase difference between the binarized feedback noise signal and the binarized secondary sound source drive signal, and aligning the phases of the binarized feedback noise signal and the binarized secondary sound source drive signal according to the phase difference; step 209 is: performing a correlation operation on the phase-aligned binarized feedforward noise signal and the binarized secondary sound source drive signal to obtain the feedforward audio signal correlation value at the current moment; performing a correlation operation on the phase-aligned binarized feedback noise signal and the binarized secondary sound source drive signal to obtain the feedback audio signal correlation value at the current moment; step 211 is: if any one of the feedforward audio signal correlation value and the feedback audio signal correlation value at the current moment is greater than a preset audio signal correlation threshold, and the signal power of any one of the feedforward noise signal, the feedback noise signal, and the secondary sound source drive signal at the current moment is greater than a preset power threshold, it is determined that there is a single-frequency signal at the current moment, and the power of the single-frequency signal at the current moment increases, resulting in howling.

[0117] Of course, in hybrid active noise cancellation, this solution can also be implemented only using the noise signal collected by the feedforward microphone, or only using the noise signal collected by the feedback microphone. The present invention is not limited.

[0118] Figure 6 It is the simulation result of the howling pre-detection method of the present invention, and the simulation is performed using a feedforward active noise cancellation device.

[0119] The figure respectively shows the output of the active noise cancellation device, the output of the correlation operation, the output of the zero-crossing detection (delay estimation), and the result after phase alignment. It can be seen from the output of the active noise cancellation device that whistling occurs at the sampling point 6000 and the power increases rapidly. It can be seen from the output of the correlation operation that the correlation value reaches the peak before the output of the active noise cancellation device saturates. It can be seen from the output of the zero-crossing detection that when the whistling starts, the phase difference between the output of the active noise cancellation device, that is, the secondary sound source drive signal and the noise signal, is stable. By aligning the phase through zero-crossing detection and correlation operation, it is possible to predict the possible whistling before it occurs.

[0120] For the output of the zero-crossing detection, the solid curve is the position of the current zero-crossing point of the secondary sound source drive signal relative to the zero-crossing point of the previous noise signal, and the dashed line is the position of the current zero-crossing point of the noise signal relative to the zero-crossing point of the previous secondary sound source drive signal. It can be seen that the phase position of the zero-crossing point is stable at about the sampling point 5000 (before the whistling occurs) and after the whistling occurs. For example, in the figure, the zero-crossing point of the secondary sound source drive signal is 8 - 9 sampling points away from the zero-crossing point of the previous noise signal, and the zero-crossing point of the noise signal relative to the zero-crossing point of the previous secondary sound source drive signal is stable at 2 sampling points. Therefore, only by delaying the secondary sound source drive signal by 2 sampling points can the microphone be aligned, or by delaying the noise signal by 8 - 9 sampling points can the secondary sound source drive signal be aligned.

[0121] It can be seen from the above simulation results that by using the whistling pre-detection method of the present invention, it is possible to quickly and accurately determine whether whistling will occur before it occurs, so as to perform control and adjustment operations on the possible whistling in advance.

[0122] The present invention also provides a whistling control method for an active noise cancellation device. The active noise cancellation device includes an audio acquisition device and a noise cancellation device, as Figure 7 shown. The whistling control method uses the whistling pre-detection method of the present invention to detect whether whistling will occur, that is, steps 301 - 313 are the same as steps 201 - 213 in Figure 3 . When it is detected at step 313 that whistling will occur at the current moment, the gain value of the secondary sound source is reduced, so that the output signal of the audio playback device can be reduced, and thus whistling can be controlled.

[0123] The present invention also provides a whistling pre-detection device, as Figure 8As shown in the figure, the howling pre-detection device includes an audio acquisition module 40, an active noise reduction module 50, a power detection module 60, a correlation operation module 70, and a control module 80. The correlation operation module 70 includes a binarization calculation unit 701, a phase alignment unit 703, and a correlation calculation unit 705. The audio acquisition module 40 acquires noise signals; the active noise reduction module 50 performs operations on the noise signals acquired by the audio acquisition device 40 to obtain a secondary sound source drive signal; the power detection module 60 calculates the signal powers of the noise signal and the secondary sound source drive signal at the current moment; the binarization calculation unit 701 in the correlation operation module 70 performs binarization processing on the noise signal and the secondary sound source drive signal to obtain a binarized noise signal and a binarized secondary sound source drive signal; the phase alignment unit 703 calculates the phase difference between the binarized noise signal and the binarized secondary sound source drive signal, and aligns the phases of the binarized noise signal and the binarized secondary sound source drive signal according to the phase difference; the correlation calculation unit 705 performs a correlation operation on the phase-aligned binarized noise signal and the binarized secondary sound source drive signal to obtain the audio signal correlation value at the current moment; the control module 80 receives the audio signal correlation value at the current moment and the signal powers of the noise signal and the secondary sound source drive signal at the current moment. If the audio signal correlation value at the current moment is greater than a preset audio signal correlation threshold, and the signal power value of any one of the noise signal and the secondary sound source drive signal at the current moment is greater than a preset power threshold, it is determined that there is a single-frequency signal at the current moment, and the single-frequency signal increases at the current moment, and it is determined that howling will occur at the current moment. Among them, the single-frequency signal is the single-frequency point signal of the howling that will occur.

[0124] The audio signal correlation threshold and the power threshold are usually given according to test experience. Among them, the power threshold may include a noise signal power threshold and / or a secondary sound source signal power threshold. The two may be the same value or different values, and both are given according to test experience.

[0125] The howling pre-detection device of the present invention, by adding a power detection module 60 and a correlation operation module 70, after acquiring the noise signal and the secondary sound source drive signal, the power detection module 60 calculates the power of the corresponding sound source signal, and at the same time the correlation operation module 70 calculates the correlation value of the noise signal and the secondary sound source drive signal, and can quickly determine whether a howling signal will be generated. Binarize the noise signal and the secondary sound source drive signal, then align the phases of the binarized noise signal and the secondary sound source drive signal, and then calculate the correlation value of the phase-aligned noise signal and the secondary sound source drive signal, which can reduce the data storage amount, reduce the operation amount of howling detection, and ensure the howling detection effect. The detection is fast and effective, and the howling pre-signal is captured before the howling is formed, so as to suppress the generation of howling.

[0126] In one embodiment, the noise signal collected by the audio acquisition device 40 is a time-domain audio amplitude value, and the secondary sound source driving signal played by the noise reduction device 50 is a time-domain audio amplitude value. By directly processing and calculating the time-domain audio amplitude values of the audio acquisition device 40 and the noise reduction device 50, the operation complexity can be reduced.

[0127] In one embodiment, the binarization calculation unit 701 may include a first judgment unit and a second judgment unit. The first judgment unit performs binarization processing on the noise signal. When the time-domain audio amplitude value of the noise is greater than 0, the binarized noise signal is 1; otherwise, the binarized noise signal is -1. The second judgment unit performs binarization processing on the secondary sound source driving signal. When the time-domain audio amplitude value of the secondary sound source driving signal is greater than 0, the binarized secondary sound source driving signal is 1; otherwise, the binarized secondary sound source driving signal is -1. Binarization generally refers to quantifying the time-domain audio amplitude value of the audio acquisition device and then fixing it to 1 bit, and performing 1-bit quantization on the time-domain audio amplitude value output by the active noise reduction device. Through this processing, the storage amount and the amount of operation can be further reduced, the operation speed can be improved, and thus the sensitivity of howling pre-detection can be improved.

[0128] In one embodiment, as Figure 9 shown, the phase alignment unit 703 includes a zero-crossing detection unit 7031, a first buffer unit 7033, and a second buffer unit 7035. The zero-crossing detection unit 7031 detects the zero-crossing points of the binarized noise signal and the binarized secondary sound source driving signal, and obtains the phase difference between the binarized noise signal and the binarized secondary sound source driving signal according to the difference of the zero-crossing points. The first buffer unit 7033 buffers the binarized noise signal. The second buffer unit 7035 buffers the binarized secondary sound source driving signal. The zero-crossing detection unit 7031 also controls the storage of the binarized noise signal in the first buffer unit 7033 and / or controls the storage of the binarized secondary sound source driving signal in the second buffer unit 7035 according to the phase difference to achieve the phase alignment of the binarized noise signal and the binarized secondary sound source driving signal. Using the zero-crossing detection unit to achieve the phase alignment of the binarized noise signal and the binarized secondary sound source driving signal has a small amount of operation and further improves the sensitivity of howling pre-detection.

[0129] In one embodiment, in order to further reduce the operation of the RAM and the amount of operation, the correlation calculation unit 70 may adopt a recursive form to perform the correlation value calculation of the binarized noise signal and the binarized secondary sound source driving signal. Specifically, it can be: the correlation value at the current moment is the correlation value at the previous moment plus the product of the binarized noise signal and the binarized secondary sound source driving signal at the current moment, and then subtract the product of the binarized noise signal and the binarized secondary sound source driving signal at the Lth moment before, where L is the length of the correlation calculation, that is, the number of sampling points in one correlation calculation.

[0130] In one embodiment, the audio acquisition device can be a feedforward microphone in an active noise cancellation system, a feedback microphone, or both a feedforward microphone and a feedback microphone. The feedforward microphone is usually arranged outside the device and is used to collect the noise signal outside the howling pre-detection device. The feedback microphone is usually arranged inside the device and is used to collect the noise signal of the external noise entering the howling pre-detection device.

[0131] The present invention also provides a howling control device, as Figure 10 shown. The howling control device includes the howling pre-detection device of the present invention, and further includes a gain value update module 90. When the howling pre-detection device detects that howling will occur at the current moment, the control module 80 configures a gain update value to the gain value update module 90 to reduce the gain value of the audio playback module 50, so as to reduce the gain value of the secondary sound source signal. Specifically, it can be to reduce the time-domain amplitude value of the secondary sound source signal, thereby achieving the effect of controlling the generation of howling.

[0132] In one embodiment, the gain value update unit 90 can be a digital gain value update unit or an analog gain value update unit. That is, the control module 80 generates a new gain value, which can be configured onto a digital signal or an analog signal.

[0133] Figure 11 FIG. 800 is a circuit block diagram of a howling pre-detection circuit according to a preferred embodiment of the present invention, including a feedforward microphone 401, a feedback microphone 402, a processor 501, a power detection circuit 601, a correlation operation circuit 701, and a controller 801.

[0134] The feedforward microphone 401 is used to collect external noise, and the feedback microphone 402 is used to collect internal noise. The processor 501 is an active noise cancellation module, including two filters (FILTER), which respectively process the data of the feedforward microphone 401 and the feedback microphone 402, and are used to generate a signal with the same amplitude and opposite phase to the noise. After being played through the speaker, the noise is cancelled. The echo cancellation module (AFC) operates on the output signal of the processor 501 and the error signal after cancellation by the reference microphone 401 to generate a compensated echo signal to eliminate the influence of the echo on the feedforward microphone signal 401. In a specific implementation, the processor 501 can only perform active noise cancellation processing on the noise collected by the feedforward microphone 401 and then output a secondary sound source drive signal, or only perform active noise cancellation processing on the noise collected by the feedback microphone 402 and then output a secondary sound source drive signal, or perform active noise cancellation processing on the noise signals collected by the feedforward microphone 401 and the feedback microphone 402 at the same time and then output a secondary sound source drive signal.

[0135] When the secondary sound source drive signal is output after performing active noise reduction processing on the noise collected by the front feed microphone 401, the relevant arithmetic circuit 701 performs binary processing, zero-crossing detection, phase alignment, and correlation calculation on the noise signal collected by the reference microphone 401 and the secondary sound source drive signal output by the processor 501, and then outputs a correlation value. The power detection circuit 601 performs power calculation on the noise signal collected by the reference microphone 401 and the secondary sound source drive signal output by the processor, and then outputs a power value. The correlation value and the power value are output to the controller 801. The controller 801 judges that if both the correlation value and the power value output to the controller are greater than the corresponding thresholds, it is judged that howling may occur, and the controller 801 can execute the control to suppress howling.

[0136] When the secondary sound source drive signal is output after performing active noise reduction processing on the noise collected by the back feed microphone 402, the calculation process is similar to the above and will not be elaborated here.

[0137] When the secondary sound source drive signal is output after performing active noise reduction processing on the noise collected by the front feed microphone 401 and the noise collected by the back feed microphone 402 at the same time, the power detection circuit 601 calculates the signal power of the front feed noise signal, the back feed noise signal, and the secondary sound source drive signal at the current moment; the relevant arithmetic circuit 701 performs binary processing on the front feed noise signal, the back feed noise signal, and the secondary sound source drive signal output by the processor 501 to obtain the binary front feed noise signal, the binary back feed noise signal, and the binary secondary sound source drive signal; calculates the phase difference between the binary front feed noise signal and the binary secondary sound source drive signal, and aligns the phases of the binary front feed noise signal and the binary secondary sound source drive signal according to the phase difference; calculates the phase difference between the binary back feed noise signal and the binary secondary sound source drive signal, and aligns the phases of the binary back feed noise signal and the binary secondary sound source drive signal according to the phase difference; performs correlation operation on the phase-aligned binary front feed noise signal and the binary secondary sound source drive signal to obtain the correlation value of the front feed audio signal at the current moment; performs correlation operation on the phase-aligned binary back feed noise signal and the binary secondary sound source drive signal to obtain the correlation value of the back feed audio signal at the current moment; the controller 801 judges that if any one of the correlation value of the front feed audio signal at the current moment and the correlation value of the back feed audio signal at the current moment is greater than the preset audio signal correlation threshold, and the signal power of any one of the front feed noise signal, the back feed noise signal, and the secondary sound source drive signal at the current moment is greater than the preset power threshold, it is judged that there is a single-frequency signal at the current moment, and the power of the single-frequency signal at the current moment increases, and howling will occur.

[0138] Figure 12 and 13 are two preferred howling control processes.

[0139] Such as Figure 12As shown, after detecting that howling may occur, the control module 801 generates a new gain value, which is configured onto the gain value update circuit 901 of the digital signal (D_GAIN). The new gain value can be optionally 6 dB, 3 dB, etc. smaller than the initial gain value, so that the output signal becomes smaller and howling is suppressed.

[0140] As Figure 13 shown, the control module generates a new gain value, which is configured onto the gain value update unit 902 of the analog signal (A_GAIN). Similarly, the gain value can be optionally 6 dB, 3 dB, etc. smaller than the initial gain, so that the output signal becomes smaller and howling is suppressed.

[0141] The present invention also provides an active noise reduction chip that can execute the howling pre-detection method or the howling control method of the present invention. By performing binarization processing to obtain a binarized curve, obtaining the zero-crossing values of the binarized noise signal and the binarized secondary sound source drive signal, performing phase alignment, calculating the correlation value of the two signals, and combining the signal power value to confirm whether the current signal frequency point will generate howling, it can reduce the computational complexity of howling detection of the chip and ensure the howling detection effect.

[0142] The present invention also provides an active noise reduction earphone, which includes the howling pre-detection device of the present invention, or the howling control device of the present invention, or the active noise reduction chip of the present invention. By adopting the howling control device of the present invention, the prediction of howling is realized, the formation of howling is prevented, and the user experience is improved.

[0143] The present invention also provides a storage medium that stores a program, wherein the program is used to be executed to implement the howling pre-detection method or the howling control method of the present invention.

[0144] It should be noted that when the present invention uses digital numbers to refer to certain specific method steps, it is only for the purpose of convenient and brief description, and by no means uses letters or numbers to limit the order of these method steps. Those skilled in the art can understand that the order of the relevant method steps should be determined by the technology itself and should not be inappropriately restricted due to the existence of step numbers.

[0145] Those skilled in the art can understand that on the premise of no conflict, the above preferred solutions can be freely combined and superimposed.

[0146] It should be understood that the above embodiments are only exemplary and not restrictive. Without departing from the basic principle of the present invention, various obvious or equivalent modifications or substitutions made by those skilled in the art to the above details will be included within the scope of the claims of the present invention.

Claims

1. A pre-whistle detection method for an active noise reduction device, where the active noise reduction device includes an audio acquisition device and a noise reduction device. The audio acquisition device is used to acquire ambient noise signals, and the noise reduction device is used to perform operations on the ambient noise signals acquired by the audio acquisition device to obtain a secondary sound source drive signal. The secondary sound source drive signal is provided to an audio playback device for playback to cancel the ambient noise and thus achieve active noise reduction. It is characterized in that, The pre-whistle detection method includes the steps of: Receiving the noise signal acquired by the audio acquisition device and the secondary sound source drive signal output by the noise reduction device; Calculating the signal power of the noise signal and the secondary sound source drive signal at the current moment; Performing binarization processing on the noise signal and the secondary sound source drive signal to obtain a binarized noise signal and a binarized secondary sound source drive signal; Calculating the phase difference between the binarized noise signal and the binarized secondary sound source drive signal, and aligning the phases of the binarized noise signal and the binarized secondary sound source drive signal according to the phase difference; Performing a correlation operation on the phase-aligned binarized noise signal and binarized secondary sound source drive signal to obtain the audio signal correlation value at the current moment; When the audio signal correlation value at the current moment is greater than a preset audio signal correlation threshold, and the signal power of any one of the noise signal and the secondary sound source drive signal at the current moment is greater than a preset power threshold, it is determined that there is a single-frequency signal at the current moment, and the power of the single-frequency signal increases at the current moment, which will generate a whistle, where the single-frequency signal is a single-frequency point signal that will generate a whistle.

2. The pre-whistling detection method according to claim 1, wherein The noise signal is the noise time-domain audio amplitude value acquired by the audio acquisition device, and the secondary sound source drive signal is the time-domain audio amplitude value of the secondary sound source drive signal output by the noise reduction device.

3. The pre-whistling detection method according to claim 2, wherein Performing binarization processing on the noise signal and the secondary sound source drive signal includes: When the noise time-domain audio amplitude value is greater than 0, the binarized noise signal is 1; otherwise, the binarized noise signal is -1; When the time-domain audio amplitude value of the secondary sound source drive signal is greater than 0, the binarized secondary sound source drive signal is 1; otherwise, the binarized secondary sound source drive signal is -1.

4. The pre-whistling detection method according to claim 1, characterized in that, Calculating the phase difference between the binarized noise signal and the binarized secondary sound source drive signal, and aligning the phases of the binarized noise signal and the binarized secondary sound source drive signal according to the phase difference includes: Calculating the zero-crossing points of the binarized noise signal and the binarized secondary sound source drive signal, and obtaining the phase difference between the binarized noise signal and the binarized secondary sound source drive signal according to the difference of the zero-crossing points; Controlling the caching of the binarized noise signal and / or the binarized secondary sound source drive signal according to the phase difference to achieve phase alignment of the binarized noise signal and the binarized secondary sound source drive signal.

5. The pre-whistling detection method according to claim 1, characterized in that, Performing a correlation operation on the phase-aligned binarized noise signal and binarized secondary sound source drive signal to obtain the audio signal correlation value at the current moment includes: The relevant value at the current moment is the relevant value at the previous moment plus the product of the binary noise signal and the binary secondary sound source drive signal at the current moment, and then minus the product of the binary noise signal and the binary secondary sound source drive signal at the L-th moment before, where L is the length value of the relevant operation.

6. The pre-whistling detection method according to any one of claims 1-5, characterized in that, The noise signal includes a feedforward noise signal and / or a feedback noise signal. The feedforward noise signal is the noise signal outside the active noise reduction device; the feedback noise signal is the noise signal of the external noise signal entering the active noise reduction device. When the noise signal is the feedforward noise signal and the feedback noise signal. Calculate the signal power of the feedforward noise signal, the feedback noise signal, and the secondary sound source drive signal at the current moment. Perform binary processing on the feedforward noise signal, the feedback noise signal, and the secondary sound source drive signal to obtain a binary feedforward noise signal, a binary feedback noise signal, and a binary secondary sound source drive signal. Calculate the phase difference between the binary feedforward noise signal and the binary secondary sound source drive signal, and align the phases of the binary feedforward noise signal and the binary secondary sound source drive signal according to the phase difference. Calculate the phase difference between the binary feedback noise signal and the binary secondary sound source drive signal, and align the phases of the binary feedback noise signal and the binary secondary sound source drive signal according to the phase difference. Perform a correlation operation on the phase-aligned binary feedforward noise signal and the binary secondary sound source drive signal to obtain the relevant value of the feedforward audio signal at the current moment. Perform a correlation operation on the phase-aligned binary feedback noise signal and the binary secondary sound source drive signal to obtain the relevant value of the feedback audio signal at the current moment. When any one of the relevant values of the feedforward audio signal and the feedback audio signal at the current moment is greater than a preset audio signal correlation threshold, and the signal power of any one of the feedforward noise signal, the feedback noise signal, and the secondary sound source drive signal at the current moment is greater than a preset power threshold, it is determined that there is a single-frequency signal at the current moment, and the power of the single-frequency signal at the current moment increases, resulting in a howling sound.

7. A howling control method for an active noise reduction device, the active noise reduction device comprising an audio acquisition device and a noise reduction device, characterized in that, Adopt the howling pre-detection method described in any one of claims 1-6 to detect whether a howling sound will occur. When it is detected that a howling sound will occur at the current moment, reduce the gain value of the secondary sound source to control the howling sound.

8. A whistling pre-detection device, characterized in that, It includes an audio acquisition module, an active noise reduction module, a power detection module, a correlation operation module, and a control module. The audio acquisition module is used to acquire the noise signal. The active noise reduction module is used to perform operations on the noise signal acquired by the audio acquisition module to obtain a secondary sound source drive signal. The power detection module is used to calculate the signal power of the noise signal and the secondary sound source drive signal at the current moment. The correlation operation module includes a binary calculation unit, a phase alignment unit, and a correlation calculation unit. The binary calculation unit performs binary processing on the noise signal and the secondary sound source drive signal to obtain a binary noise signal and a binary secondary sound source drive signal. The phase alignment unit calculates the phase difference between the binary noise signal and the binary secondary sound source drive signal, and aligns the phases of the binary noise signal and the binary secondary sound source drive signal according to the phase difference; The correlation calculation unit performs a correlation operation on the phase-aligned binary noise signal and the binary secondary sound source drive signal to obtain the audio signal correlation value at the current moment; The control module is configured to determine that there is a single-frequency signal at the current moment and the single-frequency signal increases and whistling will occur when the audio signal correlation value at the current moment is greater than a preset audio signal correlation threshold and the signal power value of any one of the noise signal and the secondary sound source drive signal at the current moment is greater than a preset power threshold, where the single-frequency signal is the single-frequency point signal that will cause whistling.

9. The pre-whistling detection device according to claim 8, characterized in that, The noise signal collected by the audio acquisition module is the noise time-domain audio amplitude value, and the secondary sound source drive signal output by the active noise reduction module is the time-domain audio amplitude value of the secondary sound source drive signal.

10. The pre-whistling detection device according to claim 9, characterized in that, The binary calculation unit includes a first judgment unit and a second judgment unit. The first judgment unit is configured to perform binary processing on the noise signal. When the noise time-domain audio amplitude value is greater than 0, the binary noise signal is 1; otherwise, the binary noise signal is -1. The second judgment unit is configured to perform binary processing on the secondary sound source drive signal. When the time-domain audio amplitude value of the secondary sound source drive signal is greater than 0, the binary secondary sound source drive signal is 1. Otherwise, the binary secondary sound source drive signal is -1.

11. The pre-whistling detection device according to claim 8, characterized in that, The phase alignment unit includes a zero-crossing detection unit, a first buffer unit, and a second buffer unit. The zero-crossing detection unit calculates the zero-crossing points of the binary noise signal and the binary secondary sound source drive signal, and obtains the phase difference between the binary noise signal and the binary secondary sound source drive signal according to the difference between the zero-crossing points. The first buffer unit is used to buffer the binary noise signal. The second buffer unit is used to buffer the binary secondary sound source drive signal. The zero-crossing detection unit also controls the storage of the binary noise signal in the first buffer unit and / or controls the storage of the binary secondary sound source drive signal in the second buffer unit according to the phase difference to achieve the phase alignment of the binary noise signal and the binary secondary sound source drive signal.

12. The pre-whistling detection device according to claim 8, wherein The correlation calculation unit calculates the correlation value at the current moment as the correlation value at the previous moment plus the product of the binary noise signal and the binary secondary sound source drive signal at the current moment, and then subtracts the product of the binary noise signal and the binary secondary sound source drive signal at the L-th moment before, where the L is the length of the correlation calculation.

13. The pre-whistling detection device according to any one of claims 8-12, characterized in that, The audio acquisition module includes a front-feed microphone and / or a rear-feed microphone. The front-feed microphone is disposed outside the whistling pre-detection device and is used to collect the noise signal outside the whistling pre-detection device. The rear-feed microphone is disposed inside the howling pre-detection device and is used to collect the noise signal of the external noise signal entering the howling pre-detection device.

14. A howling control device, characterized in that, It includes the howling pre-detection device according to any one of claims 8-13, and further includes a gain value update module. When the howling pre-detection device detects that howling will occur at the current moment, the control module configures a gain update value to the gain value update module to reduce the gain value of the secondary sound source drive signal.

15. The howling control device according to claim 14, characterized in that, The gain value update module is a digital gain value update unit or an analog gain value update unit.

16. An active noise cancellation chip, characterized in that, It is capable of executing the howling pre-detection method according to any one of claims 1-6 or the howling control method according to claim 7.

17. An active noise-canceling headphone, characterized in that, It includes the howling pre-detection device according to any one of claims 8-13, or the howling control device according to any one of claims 14-15, or the active noise reduction chip according to claim 16.

18. A storage medium, characterized in that, The storage medium stores a program, wherein the program is used to be executed to implement the howling pre-detection method according to any one of claims 1-6 or the howling control method according to claim 7.

Images