Sound collection device and method for removing wind noise

Abstract

An embodiment of the present application provides a sound collection device and a method for removing wind noise, by performing high-pass filtering on a first audio signal collected by a first microphone without a wind noise protection structure to obtain a first high-frequency signal, performing low-pass filtering on a second audio signal collected by a second microphone with a wind noise protection structure to obtain a second low-frequency signal, performing time sequence processing on the first high-frequency signal and the second low-frequency signal and fusing the aligned first high-frequency signal and the second low-frequency signal to obtain a target audio signal, which can filter out wind noise while ensuring the audio quality of the target audio signal.

Description

Technical Field

[0001] The present application relates to the technical field of noise reduction, and in particular to a sound collection device and a method for removing wind noise. Background Art

[0002] In outdoor scenes, wind noise caused by wind is a more difficult problem for equipment collection. Currently, there are mainly two types of wind noise reduction solutions:

[0003] The first is a purely physical solution, which physically separates the sound waves and wind by adding an acoustic structure design to the device, thereby reducing the impact of wind noise. However, when the existing noise reduction solution adds an acoustic structure design to the device, it usually causes the acoustic front cavity of the device to increase, which deteriorates the high-frequency collection performance of the microphone and affects the audio quality.

[0004] The second is the algorithm data processing solution, which estimates wind noise by extracting features from the signals collected by the microphone, and reduces wind noise by separating features. However, when strong wind noise causes clipping and distortion, noise reduction through algorithm data processing has problems such as overlapping of language and distortion spectrum and difficulty in feature separation, resulting in poor restoration after noise reduction.

[0005] Therefore, the audio obtained based on the above two types of wind noise reduction solutions has the problem of insufficient audio quality and cannot meet the user's demand for audio quality. Summary of the invention

[0006] The purpose of this application is to provide a sound collection device and a method for removing wind noise, so as to achieve the goal of removing wind noise while ensuring the audio quality of the restored audio signal. The specific technical solution is as follows:

[0007] In a first aspect, the present application provides a sound collection device, comprising: a housing, a first microphone, a second microphone and a sound processing component arranged in the housing;

[0008] A sound cavity is provided in the shell, and a sound pickup hole connected to the sound cavity is provided on the surface of the shell; the sound cavity is filled with windproof material;

[0009] The first microphone is arranged at the sound inlet end of the sound cavity for the sound entering the sound pickup hole, and the second microphone is arranged at the sound outlet end of the sound cavity for the sound entering the sound pickup hole;

[0010] The sound processing component is used to obtain a first audio signal collected by the first microphone, perform high-pass filtering on the first audio signal to obtain a first high-frequency signal, and obtain a second audio signal collected by the second microphone, perform low-pass filtering on the second audio signal to obtain a second low-frequency signal; perform timing alignment processing on the first high-frequency signal and the second low-frequency signal, and fuse the aligned first high-frequency signal and the second low-frequency signal to obtain a target audio signal.

[0011] Optionally, the sound pickup hole includes a first sound pickup hole and a plurality of second sound pickup holes;

[0012] The first microphone is arranged at the sound inlet end of the sound cavity for the sound entering the first sound pickup hole;

[0013] The second microphone is arranged at the sound output end of the sound cavity where the sound entering the second sound pickup hole is output.

[0014] Optionally, the plurality of second sound pickup holes are symmetrically arranged around the first sound pickup hole, and the sound cavity has a symmetrical structure with a line connecting the second microphone and the first sound pickup hole as a symmetry axis.

[0015] Optionally, a first circuit board is disposed on a side of the acoustic cavity close to the first sound pickup hole, the first microphone is disposed on a side of the first circuit board away from the first sound pickup hole, and a position of the first microphone corresponds to a position of the first sound pickup hole;

[0016] A second circuit board is provided on a side of the sound cavity away from the second sound pickup hole, and the second microphone is arranged on a side of the second circuit board away from the second sound pickup hole; a sound outlet is provided on a side of the first cavity away from the second sound pickup hole, and the position of the second microphone corresponds to the position of the sound outlet;

[0017] The sound processing component includes a first encoding processor, a second encoding processor and a main control processor respectively connected to the first encoding processor and the second encoding processor; the first encoding processor is integrated in the first circuit board, and is used to obtain a first audio signal collected by the first microphone, and high-pass filter the first audio signal to obtain a first high-frequency signal; the second encoding processor is integrated in the second circuit board, and is used to obtain a second audio signal collected by the second microphone, and low-pass filter the second audio signal to obtain a second low-frequency signal; the main control processor is used to perform timing alignment processing on the first high-frequency signal and the second low-frequency signal, and fuse the aligned first high-frequency signal and the second low-frequency signal to obtain a target audio signal.

[0018] Optionally, a first sealing member is provided on an inner wall of one side of the sound cavity where the first sound pickup hole and / or the second sound pickup hole is opened, and the first sealing member has a first through hole corresponding to the position of the first sound pickup hole and / or the second sound pickup hole;

[0019] A second sealing member is provided between the acoustic cavity and the second circuit board, and the second sealing member has a second through hole corresponding to the position of the sound outlet.

[0020] Optionally, the sound collecting device further includes a waterproof and breathable membrane that isolates the sound pickup hole from the sound cavity.

[0021] Optionally, the side walls of the acoustic cavity are streamlined or stepped.

[0022] In a second aspect, the present application provides a method for removing wind noise, which is applied to a sound collection device, wherein the sound collection device includes a first microphone and a second microphone, wherein the first microphone does not have a wind noise prevention structure, and the second microphone has a wind noise prevention structure, and the method includes:

[0023] Acquire a first audio signal collected by the first microphone and a second audio signal collected by the second microphone;

[0024] Performing high-pass filtering on the first audio signal to obtain a first high-frequency signal;

[0025] Performing low-pass filtering on the second audio signal to obtain a second low-frequency signal;

[0026] Performing timing alignment processing on the first high-frequency signal and the second low-frequency signal;

[0027] The aligned first high-frequency signal and the second low-frequency signal are fused to obtain a target audio signal.

[0028] Optionally, the performing timing alignment processing on the first high-frequency signal and the second low-frequency signal includes:

[0029] Based on a first delay amount, the first high-frequency signal is delayed so that the delayed first high-frequency signal is aligned with the second low-frequency signal; the first delay amount is the transmission delay difference of sound transmitted to the first microphone and the second microphone.

[0030] Optionally, when the sound collection device is the sound collection device of any one of the first aspects, the cutoff frequencies for high-pass filtering the first audio signal and low-pass filtering the second audio signal are determined based on the following formula:

[0031]

[0032] in, is the cut-off frequency, is the speed of sound propagation, is the damping ratio of the windproof material, is the circumference of a circle, is the radius of the pickup hole, is the volume of the acoustic cavity, is the thickness of the sound pickup hole, is the diameter of the sound pickup hole, It is the natural vibration frequency of the sound collection device when it is undamped.

[0033] Optionally, when the sound collection device is the sound collection device of any one of the first aspects, the first time delay is determined based on the following formula:

[0034]

[0035] in, is the first delay amount, is the sound propagation path length from the sound pickup hole to the second microphone, is the sound propagation path length from the sound pickup hole to the first microphone, c is the speed of sound, To preset the sound signal frequency, is the damping ratio of the windproof material, is the resonant frequency of the acoustic cavity.

[0036] Beneficial effects of this application:

[0037] The sound collection device and method for removing wind noise provided by the present application take into account that although the first microphone without a wind noise protection structure has the problem that the collected low-frequency signal is easily drowned by wind noise, it can achieve better high-frequency collection performance. Although the high-frequency collection performance of the second microphone with a wind noise protection structure will be affected, it can achieve better low-frequency collection performance. Based on the above-mentioned characteristics, the low-frequency signal collected by the first microphone is filtered out, and only the high-quality high-frequency signal is retained. The high-frequency signal collected by the second microphone is filtered out, and only the high-quality low-frequency signal is retained. Then, the retained high-frequency signal and low-frequency signal are fused at high and low frequencies. This can remove wind noise and, combined with sound fusion technology, restore the most realistic sound effect, thereby ensuring the audio quality of the restored audio signal.

[0038] Of course, implementing any product or method of the present application does not necessarily require achieving all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application, and for ordinary technicians in this field, other embodiments can also be obtained based on these drawings.

[0040] Figure 1 The overall logical architecture diagram of the sound collection device and the method for removing wind noise provided in the embodiment of the present application;

[0041] FIG2 (a) is a schematic diagram of a structure of a sound collection device provided in an embodiment of the present application;

[0042] FIG2( b ) is another schematic diagram of the structure of the sound collection device provided in an embodiment of the present application;

[0043] Figure 3 A schematic diagram of a hardware link in a sound collection device provided in an embodiment of the present application;

[0044] Figure 4 A schematic diagram of an audio processing process performed by a hardware link in a sound collection device provided in an embodiment of the present application;

[0045] Figure 5 A flow chart of a method for removing wind noise provided in an embodiment of the present application;

[0046] FIG6( a ) is a schematic diagram of an initial audio signal collected by the first microphone and the second microphone in FIG2( a ) according to an embodiment of the present application;

[0047] FIG6( b ) is a schematic diagram of a filtered audio signal after filtering the audio signal in FIG6( a );

[0048] FIG6( c ) is a schematic diagram of a target audio signal obtained by fusing the high and low frequency signals in FIG6( b ).

[0049] Reference numerals:

[0050] Shell -1, sound cavity -11, sound pickup hole -12, first sound pickup hole -12a, second sound pickup hole -12b, sound outlet -111, first microphone -21, second microphone -22, windproof material -3, first circuit board -41, first opening -411, second circuit board -42, second opening -421, first seal -51, second seal -52, waterproof breathable membrane -6. DETAILED DESCRIPTION

[0051] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field based on the present application belong to the scope of protection of the present application.

[0052] For ease of understanding, the following explains the terms and nouns that will be involved in one or more subsequent embodiments of this application:

[0053] Sound pickup: the process of collecting sound. Usually a microphone is used to collect acoustic fluctuations, convert them into electrical signals, and then sample them. The sampled electrical signals are then recorded and stored.

[0054] Wind noise: A professional term for aerodynamic noise, which is generated by the interaction between objects moving in the flow field, or the interaction between fluids caused by the turbulent motion of the fluid itself. The wind noise in this application mainly refers to the noise caused by turbulence at the microphone collection end when collecting sound.

[0055] High-pass filter: A filtering method whose rule is: high-frequency signals can pass normally, while low-frequency signals below the set critical value (cut-off frequency) are blocked and weakened. The degree of blocking and weakening will vary according to different frequencies and different filtering procedures and filtering purposes. High-pass filtering is sometimes also called low-cut filter.

[0056] Low-pass filtering: The opposite of high-pass filtering, its rule is: low-frequency signals can pass normally, while high-frequency signals above the set critical value are blocked and weakened.

[0057] Spectrum fusion: In the embodiment of the present application, spectrum fusion means that two microphones collect sounds of different frequencies respectively, and then align the high and low frequency sounds through an algorithm before fusing the high and low frequency sounds.

[0058] At present, in order to avoid the impact of wind noise on the sound pickup process, there are mainly two types of wind noise reduction solutions in the industry:

[0059] The first type is a purely physical noise reduction solution, which reduces the noise caused by the turbulent movement of wind at the microphone collection end by adding an acoustic structure design to the sound collection device. Common acoustic structures include windshields, windproof cotton, etc.

[0060] However, adding an acoustic structure design usually causes the acoustic front cavity of the microphone to be lengthened, thereby deteriorating the high-frequency collection performance of the microphone.

[0061] In a related technology, in order to reduce the impact of the elongated acoustic front cavity on the high-frequency collection performance of the microphone, multiple pickup holes are set for the same microphone, so that each pickup hole is connected to the microphone through an independent audio transmission channel, and the acoustic front cavity of the microphone is jointly constituted by the various audio transmission channels. Since the various audio transmission channels are in a parallel structure, the equivalent acoustic capacity of the acoustic front cavity of the microphone is the parallel value of the acoustic capacity of the various audio transmission channels, and the equivalent acoustic capacity of the acoustic front cavity is reduced, thereby reducing the impact of the elongated acoustic front cavity on the high-frequency collection performance of the microphone.

[0062] However, the problem with this related technology is that no matter how many audio transmission channels are added, the parallel value of the sound capacitance of each audio transmission channel cannot reach 0. In other words, no matter how many audio transmission channels are added to reduce the equivalent sound capacitance of the microphone acoustic front cavity, the high-frequency collection performance of the microphone cannot reach the high-frequency collection performance that can be achieved when the microphone is set close to the surface of the shell, and the sound collected by the microphone still has the problem of poor audio quality. In addition, with the increase of audio transmission channels in the device, the requirements for the product appearance length and volume increase significantly, which has a greater impact on the product inner diameter.

[0063] The second category is a pure algorithm solution, which generally uses dual microphones for sound collection, and identifies wind noise by the difference in wind noise between the dual microphones, thereby filtering out the wind noise.

[0064] In a related technology, the input frequency domain signals of the dual microphones are obtained respectively, and based on the irrelevance between the input frequency domain signals of the dual microphones, the input energy and wind noise energy of the first microphone and the input energy and wind noise energy of the second microphone are estimated. Then, based on a preset noise reduction gain algorithm, the noise reduction gains of the dual microphones are calculated according to the input energy and wind noise energy of the first microphone and the input energy and wind noise energy of the second microphone respectively. Finally, the input frequency domain signals are denoised according to the noise reduction gains to obtain the output frequency domain signals of the dual microphones.

[0065] However, in actual applications, it was found that this related technology has the following problems: 1. The consistency of the microphone has a great impact on the algorithm; 2. The wind sound must be collected with great difference relative to the two microphones (that is, this related technology has requirements for the direction of the wind sound); 3. The algorithm processing causes great damage to the effective signal.

[0066] In response to the problems existing in the above-mentioned purely physical noise reduction solutions and purely algorithmic noise reduction solutions, the present application superimposes the acoustic physical noise reduction structure and the audio processing algorithm, and provides a wind noise removal solution that integrates software and hardware, in order to achieve more realistic sound restoration in sound collection scenarios where wind noise exists.

[0067] The core concept of this application is as follows Figure 1 As shown, at the hardware level, the present application collects sound signals through two different microphones, one of which has no wind noise protection structure and the other has a wind noise protection structure.

[0068] Since wind noise is reflected as a low-frequency signal in the sound signal, for a microphone without a wind noise protection structure, although the wind noise will cause the low-frequency signal it collects to be significantly submerged, the microphone can collect high-quality high-frequency signals. For a microphone with a wind noise protection structure, although the added wind noise protection structure will affect its high-frequency collection performance, it enables the microphone to collect high-quality low-frequency signals.

[0069] Based on the above characteristics of the two microphones, the present application processes the sound signals collected by the two microphones as follows at the software level to obtain high-quality sound signals without wind noise: for the audio signals collected by the microphone without wind noise protection structure, the present application filters out the low-frequency signals and does not use them, and only retains the high-quality high-frequency signals; for the audio signals collected by the microphone with wind noise protection structure, the present application filters out the high-frequency signals and does not use them, and only retains the high-quality low-frequency signals. Then, the spectrum of the remaining high-frequency and low-frequency signals is fused, so that the most realistic sound effect can be restored while removing the wind noise, combining the sound fusion technology, and ensuring the audio quality of the restored audio signal.

[0070] Based on the above concept, the present application provides a sound collection device, which is described below in conjunction with a specific embodiment. Referring to FIG. 2 (a) and FIG. 2 (b), the sound collection device includes: a housing 1, a first microphone 21 disposed in the housing 1, a second microphone 22, and a sound processing component.

[0071] The housing 1 is provided with a sound cavity 11, which is filled with a windproof material 3, and the surface of the housing 1 is provided with a sound pickup hole 12 connected to the sound cavity 11. The first microphone 21 is arranged at the sound input end of the sound cavity 11 for the sound entering the sound pickup hole 12, and the second microphone 22 is arranged at the sound output end of the sound cavity 11 for the sound entering the sound pickup hole 12.

[0072] The windproof material 3 specifically refers to a material that allows sound to pass through and can buffer air turbulence to reduce wind noise. For example, the windproof material 3 can be made of windproof cotton, and the material and density of the windproof cotton can be selected in combination with actual scenarios, so as to achieve a better wind noise removal effect while avoiding the windproof cotton from affecting the transmission performance of the sound in the sound cavity 11.

[0073] Based on the above structure, the sound outside the sound collection device can enter the sound cavity 11 through the sound pickup hole 12, and be further transmitted from the position connected to the sound pickup hole 12 in the sound cavity 11 (sound input end) to the deep part of the sound cavity 11 (sound output end) until it is collected by the microphone.

[0074] The first microphone 21 mentioned above can be understood as Figure 1 The first microphone 21 is disposed at the sound inlet end of the sound cavity 11, so that the sound entering the sound pickup hole 12 can be collected by the first microphone 21 without passing through the windproof material 3 filled in the sound cavity 11, and the acoustic front cavity of the first microphone 21 is very short, so it has better high-frequency collection performance.

[0075] The second microphone 22 can be understood as Figure 1 The microphone with wind noise protection structure referred to in the figure. Since the second microphone 22 is arranged at the sound output end of the sound cavity 11, the sound entering the sound pickup hole 12 needs to pass through the windproof material 3 filled in the sound cavity 11 before it can be further collected by the second microphone 22. The airflow is greatly attenuated when passing through the windproof material 3, which can achieve the effect of filtering out wind noise, while the low-frequency sound signal can be smoothly collected by the second microphone 22 through the windproof material 3 due to its good diffraction ability, so that the second microphone 22 has good low-frequency collection performance.

[0076] The processing process of the sound signals collected by the first microphone 21 and the second microphone 22 is specifically implemented by the sound processing component. First, the sound processing component needs to obtain the first audio signal collected by the first microphone 21, perform high-pass filtering on the first audio signal to obtain a first high-frequency signal, and obtain the second audio signal collected by the second microphone 22, perform low-frequency filtering on the second audio signal, and obtain a second low-frequency signal.

[0077] The above low-pass filtering and high-pass filtering process can be implemented based on existing filtering methods, which are not described in detail in this application. When filtering, high-pass filtering can be performed on the first audio signal and low-pass filtering can be performed on the second audio signal based on the same cutoff frequency.

[0078] Exemplarily, to determine the cutoff frequency, after designing the basic acoustic structure of the sound collection device, the first microphone 21 and the second microphone 22 can be used to collect sound signals respectively, and then the collected sound signals can be subjected to actual spectrum analysis. Combined with the situation where the sound signal collected by the first microphone 21 is drowned by wind noise, and the attenuation of the sound performance collected by the second microphone 22 in the high frequency band, a suitable cutoff frequency can be set through experimental verification to ensure the extraction of high-quality first high-frequency signal and second low-frequency signal.

[0079] After obtaining the first high frequency signal and the second low frequency signal, the sound processing component needs to perform timing alignment processing on the first high frequency signal and the second low frequency signal, and fuse the aligned first high frequency signal and the second low frequency signal to obtain the target audio signal.

[0080] Specifically, since the first microphone 21 and the second microphone 22 are arranged at different positions in the sound collection device, and the sound entering the sound collection hole 12 needs to pass through the windproof material 3 before reaching the second microphone 22, the transmission time required for the sound entering the sound collection hole 12 to be transmitted to the first microphone 21 and the second microphone 22 is different. In other words, the first audio signal collected by the first microphone 21 and the second audio signal collected by the second microphone 22 do not match in timing.

[0081] Therefore, the sound processing component needs to first perform timing alignment processing on the first high-frequency signal and the second low-frequency signal so that the processed signals match in timing, and then fuse the processed first high-frequency signal and the second low-frequency signal to restore the target audio signal with wind noise removed and high audio quality.

[0082] In practical applications, the existing channel alignment algorithm for multi-microphone sound collection equipment can be combined to achieve timing alignment processing for the first high-frequency signal and the second low-frequency signal.

[0083] As an example, since the present application configures a wind noise protection structure for the second microphone 22, the time it takes for the sound to be transmitted from the sound pickup hole 12 to the second microphone 22 will be longer than the time it takes for the sound to be transmitted from the sound pickup hole 12 to the first microphone 21. Therefore, after designing the basic acoustic structure of the sound collection device, the sound signals can be collected by the first microphone 21 and the second microphone 22 respectively, and the transmission delay difference of the sound entering the sound pickup hole 12 to the first microphone 21 and the second microphone 22 can be actually measured, and the measured transmission delay difference can be recorded as the first delay amount. After subsequently obtaining the first high-frequency signal and the second low-frequency signal, the first high-frequency signal is delayed based on the first delay amount, so that the delayed first high-frequency signal and the second low-frequency signal can be time-aligned, which can greatly simplify the process of timing alignment processing of the first high-frequency signal and the second low-frequency signal.

[0084] The above content describes the basic functions to be implemented by the sound processing component. At the hardware level, the above sound processing component can specifically include an ADC (Analog-to-Digital Converter) processor (encoding processor) and a main control processor. Figure 3 and Figure 4The hardware link of the sound collection device provided in this application and the process of implementing audio processing based on the hardware link are exemplarily described in Figure 3 The sound collection device specifically includes two encoding processors, namely, encoding processor 1 (first encoding processor) and encoding processor 2 (second encoding processor). Encoding processor 1 is connected to microphone 1 (first microphone 21), encoding processor 2 is connected to microphone 2 (second microphone 22), and both encoding processor 1 and encoding processor 2 are connected to the main control processor.

[0085] The encoding processor is used to perform A / D (Analog / Digital, analog signal to digital signal) signal conversion on the sound collected by the microphone, and can filter the converted digital signal. Specifically, a filter with a cutoff frequency of f1 can be configured in the encoding processor 1, and the encoding processor 1 performs A / D signal conversion on the sound signal collected by microphone 1, and the converted audio signal is high-pass filtered through the filter configured therein to extract the high-frequency signal collected by microphone 1; and a filter with a cutoff frequency of f2 can be configured in the encoding processor 2, and the encoding processor 2 performs A / D signal conversion on the sound signal collected by microphone 2, and the converted audio signal is low-pass filtered through the filter configured therein to extract the low-frequency signal collected by microphone 2. Among them, the cutoff frequencies f1 and f2 can be equal, and their specific settings can refer to the description in the previous text.

[0086] The main control processor may be a DSP (Digital Signal Processing) processor or an ARM (Advanced RISC Machine, a processing architecture) processor, etc., for performing timing alignment processing on the high-frequency signal obtained by filtering the encoding processor 1 and the low-frequency signal obtained by filtering the encoding processor 2, and performing high- and low-frequency fusion on the aligned signals to obtain the target audio signal. Exemplarily, in combination with the above description, the transmission delay difference of the sound entering the sound pickup hole 12 to the first microphone 21 and the second microphone 22 can be determined in advance, and the transmission delay difference can be used as the first delay amount T, and the delay time length is configured as T in the main control processor. After obtaining the high-frequency signal obtained by the encoding processor 1 based on high-pass filtering, the main control processor adds a delay to the high-frequency signal based on the configured delay time length T, so as to ensure that the delayed high-frequency signal is time-aligned with the low-frequency signal obtained by the encoding processor 2 based on low-pass filtering.

[0087] Based on the above description, it can be seen that the sound collection device provided by the embodiment of the present application has a sound cavity 11 filled with a windproof material 3 in the shell 1, wherein the first microphone 21 is arranged at the sound inlet end of the sound cavity 11, and the sound entering the sound pickup hole 12 can be collected by the first microphone 21 without passing through the windproof material 3 filled in the sound cavity 11, thereby ensuring that the first microphone 21 has good high-frequency collection performance; wherein the second microphone 22 is arranged at the sound outlet end of the sound cavity 11, and the sound entering the sound pickup hole 12 needs to be buffered by the windproof material 3 in the sound cavity 11 before it can be collected by the second microphone 22, thereby achieving a noise reduction effect, and making the second microphone 22 have good low-frequency collection performance. When the sound processing component performs sound processing on the sound signals collected by the first microphone 21 and the second microphone 22, the high-quality high-frequency signal collected by the first microphone 21 is obtained by high-pass filtering, and the high-quality low-frequency signal collected by the second microphone 22 is obtained by low-pass filtering, and then the obtained high and low frequency signals are subjected to spectrum fusion processing, so that the target audio signal with high audio quality can be restored while filtering out wind noise.

[0088] In the structure shown in FIG. 2 (a) and FIG. 2 (b), the sound pickup hole 12 opened on the housing 1 includes: a first sound pickup hole 12a and a plurality of second sound pickup holes 12b, a first microphone 21 is arranged at the sound inlet end of the sound cavity 11 for the sound entering the first sound pickup hole 12a, and a second microphone 22 is arranged at the sound outlet end of the sound entering the second sound pickup hole 12b for the sound entering the sound cavity 11. In this structure, the sound entering the first sound pickup hole 12a will be directly collected by the first microphone 21, and the sound entering each second microphone 22 will be transmitted through the windproof material 3 filled in the sound cavity 11 respectively until it is collected by the second microphone 22.

[0089] As an example, the first microphone 21 and the second microphone 22 can be coaxially arranged, and the second sound pickup holes 12b opened for the second microphone 22 can be symmetrically arranged around the first sound pickup hole 12a, and the sound cavity 11 has a symmetrical structure with the line connecting the second microphone 22 and the first sound pickup hole 12a as the symmetry axis.

[0090] The structure of the sound cavity 11 can refer to the schematic diagram in FIG2 (a), and its side wall can be designed to be streamlined, or it can refer to the schematic diagram in FIG2 (b), and its side wall can be designed to be stepped. When these two designs are adopted, the length of the acoustic front cavity formed by the second microphone 22 relative to each second sound pickup hole 12b is relatively long, which can play a good buffering effect on the airflow entering the sound cavity 11 from the second sound pickup hole 12b, so as to achieve the purpose of filtering out wind noise, thereby helping to ensure the low-frequency collection performance of the second microphone 22.

[0091] Based on the above structure, the sound transmission path of the external sound transmitted to the second microphone 22 through each second sound pickup hole 12b is symmetrical with the sound transmission path of the external sound transmitted to the first microphone 21 through the first sound pickup hole 12a as the symmetry axis, which helps to ensure the sound consistency between the first microphone 21 and the second microphone 22, thereby helping to ensure the sound effect of the target audio performance is finally restored.

[0092] Furthermore, the first microphone 21 and the second microphone 22 need to be specifically arranged on a circuit board inside the sound collection device so as to further process the collected sound. In the structures shown in FIG2 (a) and FIG2 (b), a first circuit board 41 is arranged on the side of the sound cavity 11 close to the first sound pickup hole 12a, and the first microphone 21 is arranged on the side of the first circuit board 41 away from the first sound pickup hole 12a, and the position of the first microphone 21 corresponds to the position of the first sound pickup hole 12a; a second circuit board 42 is arranged on the side of the sound cavity 11 away from the second sound pickup hole 12b, and the second microphone 22 is arranged on the side of the second circuit board 42 away from the second sound pickup hole 12b, and a sound outlet 111 is opened on the side of the first cavity away from the second sound pickup hole 12b, and the position of the second microphone 22 corresponds to the position of the sound outlet 111.

[0093] In actual applications, a first opening 411 can be opened on the first circuit board 41 to expose the sound inlet of the first microphone 21 to prevent the first circuit board 41 from affecting the sound collection of the first microphone 21, and a second opening 421 corresponding to the position of the second microphone 22 can be opened on the second circuit board 42 to expose the sound inlet of the second microphone 22.

[0094] The first circuit board 41 and the second circuit board 42 mentioned above can be specifically PCB (Printed Circuit Board) or FPC (Flexible Printed Circuit) etc. In the structures shown in FIG2 (a) and FIG2 (b), the encoding processor 1 mentioned above can be specifically integrated on the first circuit board 41, and the encoding processor 2 can be integrated on the second circuit board 42. The encoding processor 1 and the encoding processor 2 can be connected to the main control processor configured in the sound collection device via the first circuit board 41 and the second circuit board 42 respectively.

[0095] In the structure shown in Figures 2 (a) and 2 (b), a first sealing member 51 is provided on the inner wall of one side of the acoustic cavity 11 on which the first sound pickup hole 12a and the second sound pickup hole 12b are opened. The first sealing member 51 has a through hole corresponding to the position of the first sound pickup hole 12a, and a through hole corresponding to the position of the second sound pickup hole 12b; a second sealing member 52 is provided between the acoustic cavity 11 and the second circuit board 42, and the second sealing member 52 has a through hole corresponding to the position of the sound outlet 111.

[0096] In this structure, the gap between the inner wall of the shell 1 and the first circuit board 41 is sealed by the first sealing member 51 to ensure that the sound entering the first sound pickup hole 12a can be collected by the first microphone 21 arranged on the first circuit board 41, which can improve the sound pickup quality of the first microphone 21. Correspondingly, the gap between the outer wall of the sound cavity 11 and the second circuit board 42 is sealed by the second sealing member 52 to ensure that the sound entering the sound cavity 11 through the second sound pickup hole 12b and transmitted to the sound outlet 111 can be collected by the second microphone 22 arranged on the second circuit board 42, which can improve the sound pickup quality of the second microphone 22, thereby helping to improve the audio quality of the target audio signal finally obtained.

[0097] The structures shown in FIG2 (a) and FIG2 (b) are further provided with a waterproof breathable membrane 6 that isolates the sound pickup hole 12 from the sound cavity 11. The waterproof breathable membrane 6 can prevent external water vapor, dust, etc. from entering the sound cavity 11 through the sound pickup hole 12 and causing damage to components such as microphones and circuit boards therein, thereby improving the component performance of the sound collection device.

[0098] Based on the same inventive concept, the present application also provides a method for removing wind noise, which is applied to a sound collection device, wherein the sound collection device includes a first microphone and a second microphone, wherein the first microphone does not have a wind noise prevention structure, and the second microphone has a wind noise prevention structure, see Figure 5 , the method comprising:

[0099] Step S101: acquiring a first audio signal collected by a first microphone and a second audio signal collected by a second microphone.

[0100] Step S102: performing high-pass filtering on the first audio signal to obtain a first high-frequency signal.

[0101] Step S103: low-pass filter the second audio signal to obtain a second low-frequency signal.

[0102] Step S104: performing timing alignment processing on the first high-frequency signal and the second low-frequency signal.

[0103] Step S105: fusing the aligned first high-frequency signal and the second low-frequency signal to obtain a target audio signal.

[0104] As an example, the method for removing wind noise provided in the present application can be applied to the sound collection device provided in any of the aforementioned embodiments of the present application. Naturally, the method for removing wind noise can also be applied to sound collection devices of other structures, as long as the sound collection device includes both a microphone with a wind noise protection structure and a microphone without a wind noise protection structure.

[0105] The contents of the above-mentioned steps S101 to S105 are all mentioned in the above-mentioned description of the sound collection device embodiment, and reference may be made to the above description.

[0106] The method for removing wind noise provided in the present application takes into account that although the first microphone without a wind noise protection structure has the problem that the collected low-frequency signal is easily drowned by wind noise, it can achieve better high-frequency collection performance. Although the high-frequency collection performance of the second microphone with a wind noise protection structure will be affected, it can achieve better low-frequency collection performance. Based on the above characteristics, the low-frequency signal collected by the first microphone is filtered out, and only the high-quality high-frequency signal is retained. The high-frequency signal collected by the second microphone is filtered out, and only the high-quality low-frequency signal is retained. Then, the retained high-frequency signal and low-frequency signal are fused at high and low frequencies. This can remove wind noise and restore the most realistic sound effect in combination with sound fusion technology, thereby ensuring the audio quality of the restored audio signal.

[0107] In one embodiment of the present application, the aforementioned step S104 specifically includes: based on a first delay amount, delaying the first high-frequency signal so that the delayed first high-frequency signal is aligned with the second low-frequency signal; the first delay amount is the transmission delay difference between the sound transmission to the first microphone and the second microphone.

[0108] The method of adding a time delay to the first high-frequency signal to achieve timing alignment between the first high-frequency signal and the first low-frequency signal has also been mentioned in the above description of the sound collection device embodiment and will not be described in detail here.

[0109] Further, when the method for removing wind noise provided in the embodiment of the present application is specifically applied to the sound collection device provided in the aforementioned embodiment of the present application, the present application provides a calculation formula for determining the first delay amount, which is specifically expressed as follows:

[0110]

[0111] in, is the first delay, is the sound propagation path length from the sound pickup hole to the second microphone, is the sound propagation path length from the sound pickup hole to the first microphone, c is the speed of sound, To preset the sound signal frequency, is the damping ratio, is the resonant frequency of the acoustic cavity.

[0112] When the sound collection device adopts the structure shown in Figure 2 (a) or Figure 2 (b), Specifically refers to the sound propagation path length from the second sound pickup hole 12b to the second microphone 22, Specifically, it is the length of the sound propagation path from the first sound pickup hole 12 a to the first microphone 21 . and The specific results can be obtained through actual measurement.

[0113] The above It can be set in combination with the actual application scenario. For example, Set to 1kHz (kilohertz). and All of these can be obtained through actual measurement.

[0114] When the method for removing wind noise provided in the present application is actually applied, the corresponding first delay amount can be calculated based on the above formula according to the actual structure of the sound collection device. After the first high-frequency signal and the second low-frequency signal are obtained based on the above steps S101 to S103, the first high-frequency signal is added with a duration of The timing alignment process can be completed by reducing the delay.

[0115] Further, the present application has mentioned in the foregoing that high-pass filtering of the first audio signal and low-pass filtering of the second audio signal can be performed based on the same cutoff frequency. When the method for removing wind noise provided in the embodiment of the present application is specifically applied to the sound collection device provided in the foregoing embodiment of the present application, the present application provides a calculation formula for determining the cutoff frequency, which is specifically expressed as follows:

[0116]

[0117] in, is the cut-off frequency, is the speed of sound propagation, is the damping ratio of the windproof material, is the circumference of a circle, is the radius of the pickup hole, is the volume of the vocal cavity, is the thickness of the pickup hole, is the diameter of the sound pickup hole, It is the natural vibration frequency of the sound collection device when it is undamped.

[0118] The above As shown in FIG2 (a), it can be specifically understood as the distance from the outer wall of the housing to the bottom of the sound pickup hole. It can be understood as the natural vibration frequency of the acoustic system provided by the sound collection device when it is undamped. For example, for the structure shown in Figure 2 (a) and Figure 2 (b), Specifically, it refers to the natural vibration frequency of the sound collecting device when the sound cavity 11 is not filled with the windproof material 3 . , and etc. can be obtained through actual measurement.

[0119] When the method for removing wind noise provided in the present application is applied in practice, the corresponding cutoff frequency can be calculated based on the above formula according to the actual structure of the sound collection device. , and then based on the cutoff frequency The filtering processes involved in step S102 and step S103 are performed.

[0120] In order to determine the effects that can be achieved by the sound collection device and the method for removing wind noise provided in the above embodiment of the present application, the present application actually applies the method for removing wind noise provided in the above embodiment to the structure shown in FIG2 (a), and experiments are conducted on the process of restoring the target audio signal. The experimental results are described below in conjunction with FIG6 (a) to FIG6 (b).

[0121] Figure 6 (a) shows the initial signal collected by the first microphone without wind noise protection structure (i.e., the direct blowing structure part shown in the figure), and the initial signal collected by the second microphone with wind noise protection structure (i.e., the wind noise reduction structure part shown in the figure). Part A in the figure is the high-frequency part of the signal collected by the wind noise reduction structure. It can be seen that this part of the high-frequency signal is significantly attenuated; Part B in the figure is the low-frequency part of the signal collected by the wind noise reduction structure. This part of the low-frequency signal basically has no attenuation problem and is less affected by the wind noise reduction structure. Part C in the figure is the high-frequency part of the signal collected by the direct blowing structure. This part of the high-frequency signal is basically not affected by wind noise; Part D in the figure is the low-frequency part of the signal collected by the direct blowing structure, in which there is obvious wind noise interference.

[0122] Figure 6 (b) shows the high-frequency signal obtained after high-pass filtering the initial signal collected by the first microphone (the direct blowing structure part), that is, the signal in area B of the figure; and the low-frequency signal obtained after low-pass filtering the initial signal collected by the second microphone (the wind noise protection structure part), that is, the signal in area C of the figure. It can be seen that both of the two signals obtained after filtering have good audio quality.

[0123] Figure 6 (c) shows the target audio signal obtained by spectrally fusing the low-frequency signal and the high-frequency signal in Figure 6 (b). Taking the signal in the E area as an example, it can be seen that the target audio signal combines the advantages of the audio signal of the direct blowing structure and the audio signal of the wind noise reduction structure, the wind noise is filtered out, and the audio quality is significantly improved.

[0124] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0125] Each embodiment in this specification is described in a related manner, and the same or similar parts between the embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0126] The above description is only a preferred embodiment of the present application and is not intended to limit the protection scope of the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application are included in the protection scope of the present application.

Claims

1. A sound collection device, characterized in that it includes: A housing, a first microphone, a second microphone and a sound processing component disposed in the housing; A sound cavity is provided in the shell, and a sound pickup hole connected to the sound cavity is opened on the surface of the shell; the sound cavity is filled with windproof material; The first microphone is arranged at the sound inlet end of the sound cavity for the sound entering the sound pickup hole, and the second microphone is arranged at the sound outlet end of the sound entering the sound pickup hole; the sound entering the sound pickup hole reaches the second microphone after passing through the windproof material; The sound processing component is used to obtain a first audio signal collected by the first microphone, perform high-pass filtering on the first audio signal to obtain a first high-frequency signal, and obtain a second audio signal collected by the second microphone, perform low-pass filtering on the second audio signal to obtain a second low-frequency signal; perform timing alignment processing on the first high-frequency signal and the second low-frequency signal, and fuse the aligned first high-frequency signal and the second low-frequency signal to obtain a target audio signal.

2. The sound collection device according to claim 1, characterized in that: The sound pickup holes include a first sound pickup hole and a plurality of second sound pickup holes; The first microphone is arranged at the sound inlet end of the sound cavity for the sound entering the first sound pickup hole; The second microphone is arranged at the sound output end of the sound cavity where the sound entering the second sound pickup hole is output.

3. The sound collection device according to claim 2, characterized in that: The plurality of second sound pickup holes are symmetrically arranged around the first sound pickup hole, and the sound cavity is symmetrically structured with the line connecting the second microphone and the first sound pickup hole as the symmetry axis.

4. The sound collection device according to claim 3, characterized in that: A first circuit board is arranged on a side of the sound cavity close to the first sound pickup hole, and the first microphone is arranged on a side of the first circuit board away from the first sound pickup hole, and the position of the first microphone corresponds to the position of the first sound pickup hole; A second circuit board is provided on a side of the sound cavity away from the second sound pickup hole, and the second microphone is arranged on a side of the second circuit board away from the second sound pickup hole; a sound outlet is provided on a side of the sound cavity away from the second sound pickup hole, and the position of the second microphone corresponds to the position of the sound outlet; The sound processing component includes a first encoding processor, a second encoding processor and a main control processor respectively connected to the first encoding processor and the second encoding processor; the first encoding processor is integrated in the first circuit board, and is used to obtain a first audio signal collected by the first microphone, and high-pass filter the first audio signal to obtain a first high-frequency signal; the second encoding processor is integrated in the second circuit board, and is used to obtain a second audio signal collected by the second microphone, and low-pass filter the second audio signal to obtain a second low-frequency signal; the main control processor is used to perform timing alignment processing on the first high-frequency signal and the second low-frequency signal, and fuse the aligned first high-frequency signal and the second low-frequency signal to obtain a target audio signal.

5. The sound collection device according to claim 4, characterized in that: A first sealing member is provided on the inner wall of one side of the sound cavity where the first sound pickup hole and / or the second sound pickup hole are opened, and the first sealing member has a first through hole corresponding to the position of the first sound pickup hole and / or the second sound pickup hole; A second sealing member is provided between the acoustic cavity and the second circuit board, and the second sealing member has a second through hole corresponding to the position of the sound outlet.

6. The sound collection device according to any one of claims 1 to 5, characterized in that: The sound collecting device also includes a waterproof and breathable membrane that isolates the sound pickup hole from the sound cavity.

7. The sound collection device according to any one of claims 1 to 5, characterized in that: The side wall of the acoustic cavity is streamlined or stepped.

8. A method for removing wind noise, characterized in that: Applied to the sound collection device according to any one of claims 1 to 7, the method comprises: Acquire a first audio signal collected by the first microphone and a second audio signal collected by the second microphone; Performing high-pass filtering on the first audio signal to obtain a first high-frequency signal; Performing low-pass filtering on the second audio signal to obtain a second low-frequency signal; Performing timing alignment processing on the first high-frequency signal and the second low-frequency signal; The aligned first high-frequency signal and the second low-frequency signal are fused to obtain a target audio signal.

9. The method according to claim 8, characterized in that The performing timing alignment processing on the first high-frequency signal and the second low-frequency signal includes: Based on a first delay amount, the first high-frequency signal is delayed so that the delayed first high-frequency signal is aligned with the second low-frequency signal; the first delay amount is the transmission delay difference of sound transmitted to the first microphone and the second microphone.

10. The method according to claim 8, characterized in that The cutoff frequencies for high-pass filtering the first audio signal and low-pass filtering the second audio signal are determined based on the following formula: ； in, is the cut-off frequency, is the speed of sound propagation, is the damping ratio of the windproof material, is the circumference of a circle, is the radius of the pickup hole, is the volume of the acoustic cavity, is the thickness of the sound pickup hole, is the diameter of the sound pickup hole, It is the natural vibration frequency of the sound collection device when it is undamped.

11. The method according to claim 9, characterized in that The first delay amount is determined based on the following formula: ； in, is the first delay amount, is the sound propagation path length from the sound pickup hole to the second microphone, is the sound propagation path length from the sound pickup hole to the first microphone, c is the speed of sound, To preset the sound signal frequency, is the damping ratio of the windproof material, is the resonant frequency of the acoustic cavity.

Images