Equipment pickup method and terminal equipment
Patent Information
- Application Number
- CN202410358523.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-03-26
- Publication Date
- 2025-10-03
Smart Images
Figure CN120751319A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of terminal technology, and in particular to a method for device sound pickup and a terminal device. Background Art
[0002] Sound pickup technology is widely used in various audio and video scenarios, and it mainly uses microphone arrays to pick up sound. Currently, in traditional wireless headphones, sound pickup is usually achieved by deploying omnidirectional microphones or omnidirectional microphone arrays. However, omnidirectional microphones respond to sound signals uniformly in all directions in space and cannot enhance sounds in a specific direction, thus easily retaining ambient noise. Unlike omnidirectional microphones, directional microphones have specific spatial directionality and can be used to enhance sound signals in a specific direction, improving the sound pickup effect.
[0003] Considering that in current scenarios such as calls and headphone reception, the sound pickup effect of headphones is still easily affected by environmental noise, how to use directional microphones to enhance sound signals in a specific direction and improve the sound pickup effect of headphones has become a topic worthy of research. Summary of the Invention
[0004] The embodiments of the present application provide a method for device sound pickup and a terminal device, which solves the problem of poor sound pickup effect of headphones by combining a MEMS microphone array structure of an omnidirectional microphone and a directional microphone with a back-end algorithm.
[0005] In a first aspect, a method for picking up sound by a device is provided, which is applied to a terminal device, wherein the terminal device includes a micro-electromechanical system (MEMS) microphone array, the MEMS microphone array includes an omnidirectional microphone and a directional microphone, and the direction of maximum acoustic signal gain of the MEMS microphone array points in a first direction and a second direction. The method includes:
[0006] Acquire a first acoustic signal acquired by the omnidirectional microphone and a second acoustic signal acquired by the directional microphone;
[0007] transforming the first acoustic signal in the time domain into a third acoustic signal in the time-frequency domain, and transforming the second acoustic signal in the time domain into a fourth acoustic signal in the time-frequency domain using a short-time Fourier algorithm;
[0008] Filter coefficients are obtained according to the third acoustic signal and the fourth acoustic signal, wherein the filter coefficients ensure that the direction in which the acoustic signal after filtering has the maximum gain is the first direction and the direction in which the acoustic signal has the maximum attenuation is the second direction.
[0009] It should be understood that the direction from the wireless headset to the human mouth can specifically refer to the direction from the center point of the bottom of the wireless headset to the center point of the human mouth, or the direction from the second sound inlet of the directional microphone in the wireless headset (corresponding to the sound inlet 2 in the embodiment of this article) to the center point of the human mouth, etc.
[0010] The method for device sound pickup provided by this implementation combines hardware and software improvements, wherein the hardware improvements include applying a MEMS microphone array including an omnidirectional microphone and a directional microphone in the terminal device, wherein the omnidirectional microphone and the directional microphone can both be MEMS microphones. The sound pickup direction of the MEMS microphone array can correspond to the direction in which the acoustic signal gain of the microphone array is the largest (or the direction in which the acoustic signal is enhanced the most), that is, the first direction and the second direction, and the acoustic signal enhancement direction is determined by the MEMS microphone array structure. In order to reduce the influence of acoustic signals in other spatial directions on the sound pickup effect of the device, on the basis of the MEMS microphone array, the first acoustic signal and the second acoustic signal obtained by the omnidirectional microphone and the directional microphone are combined, and algorithm processing is used to achieve that the direction in which the acoustic signal gain of the terminal device is the largest is unidirectional (that is, the first direction).
[0011] In combination with the first aspect, in some implementations of the first aspect, the terminal device is a wireless headset, and the first direction is the direction of the wireless headset pointing to the human mouth when the user wears the wireless headset.
[0012] It should be understood that wireless headphones are relatively small in size. Although traditional MEMS microphones are relatively large in size, with the development of MEMS technology, MEMS microphones have gradually become smaller. Based on this feature, the method provided in this implementation will apply a MEMS microphone array including omnidirectional microphones and directional microphones to wireless headphones.
[0013] In combination with the first aspect, in some implementations of the first aspect, obtaining a filter coefficient based on the third acoustic signal and the fourth acoustic signal includes:
[0014] performing dereverberation processing on the third sound signal to obtain a fifth sound signal; and
[0015] performing dereverberation processing on the fourth sound signal to obtain a sixth sound signal; wherein the dereverberation processing is used to remove ambient noise in the third sound signal and / or the fourth sound signal;
[0016] The filter coefficient is obtained according to the fifth sound signal and the sixth sound signal.
[0017] In combination with the first aspect, in some implementations of the first aspect, obtaining the filter coefficient according to the fifth sound signal and the sixth sound signal specifically includes:
[0018] performing amplitude compensation on the fifth sound signal to obtain a seventh sound signal; and
[0019] performing amplitude compensation and phase compensation on the sixth sound signal to obtain an eighth sound signal, wherein the amplitude compensation is used to compensate for a sound signal amplitude loss caused by a device structure and environmental objects of the terminal device, and the phase compensation is performed so that a phase difference between the seventh sound signal and the eighth sound signal is a preset phase difference;
[0020] The filter coefficient is obtained according to the seventh sound signal and the eighth sound signal.
[0021] In combination with the first aspect, in some implementations of the first aspect, obtaining the filter coefficient according to the seventh sound signal and the eighth sound signal specifically includes:
[0022] The filter coefficient is obtained according to the seventh sound signal, the eighth sound signal and a steering vector, where the steering vector is:
[0023]
[0024] in, θ represents the direction of the acoustic signal gain of the MEMS microphone array, j 2 = -1, ω = 2πf, f is the frequency point of the acoustic signal in the frequency domain, τ0 is the delay of the plane wave reaching two adjacent microphones when incident from the 0° direction, ⊙ is the Hadamard product of the matrix, and c(ω,θ) is the correction factor obtained by measuring acoustic signals of different frequencies incident from different directions;
[0025] The filter coefficient h(ω) is:
[0026] h(ω)=A H (AA H ) -1 b
[0027] Where H is the matrix conjugate transpose, b=[1 0] T , T represents matrix transpose; θ max Indicates the direction of the acoustic signal gain, θ max Indicates the direction where the acoustic signal attenuation is greatest.
[0028] In conjunction with the first aspect, in some implementations of the first aspect, the method further includes:
[0029] performing post-processing and noise reduction on the filtered acoustic signal;
[0030] The post-processed noise-reduced acoustic signal is transformed into the time domain using an inverse short-time Fourier transform.
[0031] In combination with the first aspect, in certain implementations of the first aspect, the MEMS microphone array satisfies that when the frequency is a preset frequency, the difference in sound pressure levels between the maximum angle of acoustic signal gain and the maximum angle of attenuation of the omnidirectional microphone is less than a first threshold, and the difference in sound pressure levels between the maximum angle of acoustic signal gain and the maximum angle of attenuation of the directional microphone is equal to or greater than a second threshold.
[0032] In combination with the first aspect, in some implementations of the first aspect, the preset frequency is 1 kHz, the first threshold is 2 dB, and the second threshold is 15 dB.
[0033] In combination with the first aspect, in certain implementations of the first aspect, the MEMS microphone array includes an omnidirectional microphone and a directional microphone, the omnidirectional microphone is arranged at the upper end of the wireless headset, the upper end is the end close to the sound outlet of the wireless headset, and the directional microphone is arranged at the bottom end of the wireless headset. When the wireless headset is worn on the human ear, the direction of the maximum acoustic signal gain of the MEMS microphone array is close to or along the direction of the human mouth pointed by the wireless headset.
[0034] In combination with the first aspect, in some implementations of the first aspect, the first direction is 310°, and the second direction is 130°.
[0035] In a second aspect, a wireless headset is provided, comprising a MEMS microphone array, wherein the MEMS microphone array comprises an omnidirectional microphone and a directional microphone, the omnidirectional microphone being arranged at the upper end of the wireless headset, the upper end being the end close to the sound outlet of the wireless headset, and the directional microphone being arranged at the bottom end of the wireless headset, and the direction in which the acoustic signal gain of the MEMS microphone array is maximum is close to or along the direction in which the wireless headset points to the direction of the human mouth when the wireless headset is worn on the human ear.
[0036] In combination with the second aspect, in some implementations of the second aspect, the directional microphone includes a first sound inlet and a second sound inlet, the first sound inlet is arranged on the outer shell of the bottom end of the wireless headset, and the second sound inlet is arranged on the outer shell of the bottom end of the wireless headset, the first sound inlet and the second sound inlet are connected through a right-angle pipe, and the right-angle pipe includes a first pipe perpendicular to the direction of the earphone handle of the wireless headset, and a second pipe and a third pipe parallel to the direction of the earphone handle.
[0037] In combination with the second aspect, in certain implementations of the second aspect, the omnidirectional microphone includes a third sound inlet arranged in the outer shell of the earphone handle at the upper end of the wireless earphone, and the third sound inlet is connected to a sound inlet duct perpendicular to the direction of the earphone handle.
[0038] In a third aspect, a terminal device is provided, including:
[0039] processor;
[0040] Memory;
[0041] The memory stores a computer program, which includes instructions. When the instructions are executed by the processor, the terminal device executes the method described in any one of the implementations of the first aspect.
[0042] In a fourth aspect, a chip system is provided, which includes a processing circuit, a receiving pin and a transmitting pin; wherein the receiving pin, the transmitting pin and the processing circuit communicate with each other through an internal connection path, and the processing circuit executes the method described in any one of the implementation methods in the first aspect above to control the receiving pin to receive signals and control the transmitting pin to send signals.
[0043] In a fifth aspect, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer-executable program instructions, and when the computer-executable program instructions are executed on a computer, the computer executes the method as described in any one of the implementation methods of the first aspect above.
[0044] In a sixth aspect, a computer program product is provided, comprising a computer program code, which, when executed on a computer, causes the computer to execute the method described in any one of the implementations of the first aspect. BRIEF DESCRIPTION OF THE DRAWINGS
[0045] Figure 1A and Figure 1B They are respectively a directivity diagram formed by an omnidirectional microphone and a directivity diagram formed by a directional microphone provided in an embodiment of the present application.
[0046] Figure 2 A schematic diagram of an earphone provided in an embodiment of the present application.
[0047] Figure 3 A schematic diagram of the partially enlarged structure of an earphone provided in an embodiment of the present application.
[0048] Figure 4A and Figure 4B A schematic diagram of the measured directivity of a MEMS microphone array provided in an embodiment of the present application.
[0049] Figure 5A and Figure 5B A schematic diagram of the acoustic signal enhancement direction of the wireless headset provided in an embodiment of the present application.
[0050] Figure 6 A schematic flowchart of the algorithm processing process involved in the implementation of a device sound pickup method provided in an embodiment of the present application.
[0051] Figure 7 A schematic diagram of a physical model of a simplified MEMS microphone array structure in a headset provided in an embodiment of the present application.
[0052] Figures 8A to 8C Schematic diagram of the theoretical value of the phase difference between the sound signals of the omnidirectional microphone and the directional microphone in the MEMS microphone array when the sound frequencies provided in the embodiment of the present application are 2KHz, 4KHz and 6KHz.
[0053] Figure 9 The above diagrams are the actual headphone sound pickup directivity diagram and the theoretical directivity diagram when the sound frequency is 1 kHz using the device sound pickup method provided in the embodiment of the present application.
[0054] Figure 10 A schematic flowchart of another method for picking up sound by a device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0055] It should be noted that the terms used in the implementation methods section of the embodiments of the present application are only used to explain the specific embodiments of the present application, and are not intended to limit the present application. In the description of the embodiments of the present application, unless otherwise specified, " / " means or, for example, A / B can mean A or B; "and / or" in this article is merely a way to describe the association relationship of associated obstacles, indicating that there can be three relationships, for example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, in the description of the embodiments of the present application, unless otherwise specified, "multiple" means two or more than two, "at least one" and "one or more" mean one, two or more than two.
[0056] In the following, the terms "first" and "second" are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of technical features indicated. Therefore, the definition of "first" and "second" features may explicitly or implicitly include one or more of the features.
[0057] References to "one embodiment" or "some embodiments" in this specification mean that a particular feature, structure, or characteristic described in conjunction with that embodiment is included in one or more embodiments of the present application. Thus, phrases such as "in one embodiment," "in some embodiments," "in other embodiments," and "in yet other embodiments" appearing in various places in this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "including," "comprising," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.
[0058] The following specific embodiments are used to describe the technical solution of the present application in detail. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.
[0059] In order to better understand the method for device sound pickup provided in the embodiments of the present application, the following first introduces the terms or concepts that may be involved in the following embodiments.
[0060] 1. Omnidirectional microphone
[0061] An omnidirectional microphone is a microphone that has uniform sensitivity to sound signals (or voice signals) in all directions of space. It can collect sound signals in any direction.
[0062] 2. Directional microphone
[0063] A directional microphone is sensitive only to sound signals in a specific direction. This means it can only pick up sound signals from that specific direction. The directional microphone in the embodiments of this application can be a figure-of-eight directional microphone. The pickup shape of a figure-of-eight directional microphone resembles the number "8," and is also known as a dual-cardioid microphone or a bidirectional microphone.
[0064] 3. Beam
[0065] The spatial direction in which a microphone responds to an acoustic signal can be represented by a beam, that is, the beam can represent the microphone's sensitivity to acoustic signals in a specific direction. The direction in which the beam is pointing indicates that the microphone is more sensitive to acoustic signals in that direction.
[0066] like Figure 1A Figure 2 shows the beam pattern of an omnidirectional microphone. The beam pattern of an omnidirectional microphone is close to a circle, with the boundary of the circle at 0 decibels (dB), indicating a maximum concavity of 0 dB. This indicates that an omnidirectional microphone is sensitive to sound signals from all directions in space, not just in a specific direction, and has poor directivity.
[0067] like Figure 1B The figure shows a schematic diagram of the beam directivity formed by an 8-shaped directional microphone. The beam shape of the directional microphone is similar to the number "8". For example, in the 0° and 180° directions in the figure, the directional beam can reach 0dB, that is, the sound signals in the 0° and 180° directions are not attenuated. However, there is suppression for the sound signals in other directions. For example, the maximum depression depth corresponding to the 90° and 270° directions in the figure is -40dB, and the sound signal attenuation is relatively large. Among them, the maximum depression depth can be used to indicate the degree of attenuation of the sound signal by the directional microphone in a specific direction, so as to indicate the degree of attenuation of the sound signal in the directional microphone. Figure 1B For example, the maximum depression degree of the figure-8 directional microphone is -40dB, which indicates that the directional microphone does not collect sound signals in this specific direction.
[0068] 4. Micro electro mechanical system (MEMS) microphone
[0069] MEMS microphone is also called silicon microphone. MEMS microphone is an acoustic-to-electric converter made based on MEMS technology, which has the characteristics of good frequency response characteristics and low noise. The general structure of a MEMS microphone is to use a printed circuit board (PCB) and a shell to form a microphone package. A MEMS chip and an application specific integrated circuit (ASIC) chip electrically connected to it are integrated on the PCB board inside, and the signal is transmitted through bonding gold wires. Among them, the MEMS chip includes a substrate and a diaphragm and a backplate fixed on the substrate. The diaphragm and the backplate constitute a capacitor and are integrated on a silicon wafer. When picking up sound, the sound enters the microphone through the sound hole and acts on the diaphragm of the MEMS chip. Through the vibration of the diaphragm, the distance between the diaphragm and the backplate is changed, thereby converting the sound signal into an electrical signal. The omnidirectional microphone and the directional microphone in the embodiments of the present application can both be MEMS microphones.
[0070] Combined with the introduction in the background technology, in the current wireless headset call scenario, users still face the problem of poor call quality when the ambient noise is strong, which puts higher requirements on the sound pickup characteristics of wireless headsets. The microphones used in traditional wireless headsets are usually omnidirectional microphones, which make it difficult to enhance the sound signal in a specific direction. In recent years, with the development of MEMS technology, MEMS microphones tend to be smaller, making it possible to use them in small electronic products. In the method for device sound pickup provided in the embodiment of the present application, the MEMS microphone array formed by the omnidirectional microphone and the directional microphone is used as an improved hardware structure in the headset.
[0071] However, because directional microphones have an "8"-shaped spatial directivity, two sound holes are generally required on the surface of the device. This requires careful design of the specific structure of the device to ensure that the microphone's pickup characteristics are maintained while also ensuring compatibility with other components in the device. Considering the restrictions on the number of microphones in consumer electronics and the impact of the structural characteristics of directional microphones themselves (requiring two sound holes) on the product's appearance, the number of directional microphones should not be too large. In addition, although directional microphones themselves have an "8"-shaped spatial directivity, in headphone pickup scenarios, it is often necessary to only enhance sound in a single direction.
[0072] In view of this, an embodiment of the present application provides a method for device sound pickup. By applying a MEMS microphone array including an omnidirectional microphone and a directional microphone in headphones, combined with a back-end design algorithm, it not only meets the volume and appearance requirements of the headphones for the device, but also achieves the headphone's enhanced sound pickup effect for sound signals in a one-way specific direction. Specifically, the method for device sound pickup provided by an embodiment of the present application can achieve that when a user wears two headphones, the two headphones can only enhance the sound signals pointing in the direction of the human mouth, without enhancing the sound signals in other directions, that is, filtering out environmental noise and mainly picking up the voice of the human mouth sound source, thereby improving the sound pickup effect in the headphone call scenario and enhancing the user experience.
[0073] It should be noted that the embodiment of the present application, which enhances the sound signal pointing in the direction of the human mouth without enhancing the sound signals in other directions, may also include not attenuating the sound signal pointing in the direction of the human mouth, but attenuating the sound signals in other directions.
[0074] For example, Figure 2 , which is a schematic diagram of an earphone provided in an embodiment of the present application.
[0075] in, Figure 2 The invention will be described using an in-ear wireless headset as an example. The headset may be a Bluetooth headset, for example. In addition, the device sound pickup method provided in the embodiment of the present application can also be applied to other types of headsets, such as headphones, earphones, etc., and the embodiment of the present application does not limit this.
[0076] Combine Figure 2 Wireless earphones can be equipped with a MEMS microphone array including an omnidirectional microphone and a directional microphone as the front-end device for collecting sound signals. Among them, the omnidirectional microphone can be set at the upper end of the earphone handle (the end close to the earphone sound outlet) and close to the inside of the shell on one side of the earphone handle. The length direction of the corresponding sound inlet can be perpendicular to the length direction of the earphone handle, and the sound input direction along the sound inlet direction is also perpendicular to the length direction of the earphone handle (see Figure 2The structure shown by the circular dotted line frame at the upper end of the earphone stem). The directional microphone can be an 8-shaped directional microphone, which can be arranged at the bottom of the earphone stem (such as Figure 2 the position shown by the rectangular dotted line frame at the bottom of the earphone stem), for example, closely attached to the housing at the bottom of the earphone stem. The front and back surfaces of the 8-shaped directional microphone can be perpendicular to the length direction of the earphone stem (see Figure 2 the structure shown by the rectangular dotted line frame at the bottom of the earphone stem). The 8-shaped directional microphone has two sound inlet ports, denoted as sound inlet port 1 and sound inlet port 2. Sound inlet port 1 can be opened at the rear of the 8-shaped directional microphone and led out through a right-angled pipe to communicate with sound inlet port 2.
[0077] In some embodiments, the voice enhancement direction of the 8-shaped directional microphone can be the connection direction of sound inlet port 1 and sound inlet port 2. In order to obtain better voice enhancement effect, the strongest pickup angle of the 8-shaped directional microphone can also be aligned with the connection line of sound inlet port 1 and sound inlet port 2. For example, if the strongest pickup angles of the 8-shaped directional microphone are 130° and 310° (see Figure 4B ), then the connection line of sound inlet port 1 and sound inlet port 2 can be aligned with the direction where the 130° and 310° are located.
[0078] In some embodiments, the distance between the omnidirectional microphone and the directional microphone can be represented by δ. In common earphone forms, the value range of this distance can be, for example: 10mm < δ < 20mm. However, in earphones of different sizes, the value of δ can be flexibly set to adapt the omnidirectional microphone and the directional microphone to other structures in the earphone. The embodiments of the present application do not limit this.
[0079] In some embodiments, as [[ID=I6]] Figure 3 shown, the connection of the right-angled pipe connecting sound inlet port 1 and sound inlet port 2 can include a first pipe connected to sound inlet port 1 and perpendicular to the length direction of the earphone stem, a second pipe connected to the first pipe and perpendicular to it, and a third pipe connected to the second pipe and communicating with it through an intermediate component area, where the second pipe and the third pipe can be parallel to the length direction of the earphone.
[0080] As an example, in common earphone forms, the length range of the first pipe, the second pipe and the third pipe can satisfy, for example: 10mm < L1 + L2 + L3 < 20mm. Where, L1 is the length of the first pipe, L2 is the length of the second pipe, and L3 is the length of the third pipe. However, in earphones of different sizes, L1, L2 and L3 can be flexibly set respectively to adapt the right-angled pipe to other structures in the earphone. The embodiments of the present application do not limit this.
[0081] It should be noted that Figure 2 and Figure 3Only one earphone structure is used as an example for explanation. For the other earphone, the positions of the omnidirectional microphone and the directional microphone can be based on the central axis of the human head and the Figure 2 The structure shown is a mirror image layout, and this article will not provide a detailed description of the structure of the other earphone.
[0082] It should also be noted that Figure 2 The positions of the omnidirectional microphone and the directional microphone in the earphones are shown as examples only. In actual applications, the positions of the omnidirectional microphone and the directional microphone in the earphones may not be limited to these. In some embodiments, the positions of the omnidirectional microphone and the directional microphone in the earphones only need to meet the following layout criteria: when the frequency of the sound source is a preset frequency, the sound pressure level difference between the strongest sound pickup angle and the weakest sound pickup angle of the omnidirectional microphone is less than a first threshold, and the sound pressure level difference between the strongest sound pickup angle and the weakest sound pickup angle of the directional microphone is greater than or equal to a second threshold. In other words, in the MEMS microphone array of the earphones, if the omnidirectional microphone and the directional microphone meet the aforementioned layout criteria, then the specific positions of the omnidirectional microphone and the directional microphone in the earphones may not be specifically limited.
[0083] As an example, the preset frequency in the layout standard can be 1KHz, the first threshold can be 4dB, and the second threshold can be 9dB. That is, the layout of the omnidirectional microphone and the directional microphone only needs to meet the following standards: when the frequency of the sound source is 1KHz, the sound pressure level difference between the strongest pickup angle and the weakest pickup angle of the omnidirectional microphone is less than 4dB, and the sound pressure level difference between the strongest pickup angle and the weakest pickup angle of the directional microphone is equal to or greater than 9dB.
[0084] Taking the above layout standards as an example, Figure 2 The MEMS microphone array shown in the figure is measured, and the measured results are as follows Figure 4A and Figure 4B As shown. Among them, Figure 4A This diagram shows the directivity of an omnidirectional microphone in a headphone MEMS microphone array, measured at a 1 kHz sound source frequency. Under these test conditions, the omnidirectional microphone maintains essentially consistent directivity in all directions, with signal attenuation near 0dB at both the strongest and weakest pickup angles, and the sound pressure level difference between the two being less than 4dB. Figure 4BThe figure shows the directivity diagram of the directional microphone in the MEMS microphone array of the headset at a sound source frequency of 1KHz. It can be concluded that under this test condition, the directional microphone has the strongest sound pickup angles of 130° and 310°, and the corresponding sound signal attenuation is 0dB. The weakest sound pickup angles are 30° and 210°, and the corresponding sound signal attenuation is -15dB. In other words, the sound pressure level difference between the strongest and weakest sound pickup angles is greater than 9dB. Figure 2 The MEMS microphone array structure provided by the embodiment can meet the layout standards.
[0085] For example, Figure 5A As shown in the figure, it is a schematic diagram of the direction of acoustic signal enhancement of the wireless earphones on both sides when the human ear wears the wireless earphones under the earphone structure provided in the above embodiment. After applying the above-mentioned MEMS microphone array structure to the earphones, the wireless earphones can enhance the acoustic signals emitted from the direction of the human mouth after being worn. However, since the figure-8 directional microphone can enhance the acoustic signals in both directions, in the earphone call scenario, if only relying on the improvements in the hardware structure introduced above, then in addition to enhancing the voice signals emitted by the human mouth, the wireless earphones will also enhance the noise signals in the environment, and ultimately still cannot obtain a good earphone call effect. Therefore, it is necessary to combine the back-end algorithm to make the earphones only enhance the acoustic signals emitted by the human mouth, without enhancing the acoustic signals in other directions, that is, to achieve the following Figure 5B Shown is the effect of enhancing a unidirectional acoustic signal directed towards the human mouth.
[0086] The following describes the back-end algorithm processing involved in the device sound pickup method provided in the embodiment of the present application in conjunction with the accompanying drawings. Figure 6 The figure is a schematic flow chart of the algorithm processing process involved in the implementation of a device sound pickup method provided in an embodiment of the present application. The execution body of this process may include the sound collection module and processing module in the headset, and specifically may include the following steps:
[0087] S601A, omnidirectional microphone acquires the first sound signal.
[0088] S601B: The directional microphone obtains a second sound signal.
[0089] Among them, the omnidirectional microphone and directional microphone here can correspond to the above Figure 2 Omnidirectional microphones and directional microphones in the MEMS microphone array in the embodiment.
[0090] In some embodiments, the first acoustic signal may be a signal corresponding to an input sound acquired by an omnidirectional microphone, and the second acoustic signal may be a signal corresponding to an input sound acquired by a directional microphone. The directional microphone may be an 8-way directional microphone.
[0091] S602A: Frame the first sound signal.
[0092] S602B: Frame the second sound signal.
[0093] S603A: transform the framed acoustic signal corresponding to the omnidirectional microphone from the time domain into a third acoustic signal in the time-frequency domain.
[0094] S603B: transform the framed acoustic signal corresponding to the directional microphone from the time domain into a fourth acoustic signal in the time-frequency domain.
[0095] In some embodiments, the acoustic signals corresponding to the omnidirectional microphone and the acoustic signals corresponding to the directional microphone can be transformed from the time domain to the time-frequency domain using a short-time Fourier transform. The specific calculation method will be described below and will not be described in detail here.
[0096] S604A: Perform dereverberation processing on the third sound signal to obtain a fifth sound signal.
[0097] S604B: Perform dereverberation processing on the fourth sound signal to obtain a sixth sound signal.
[0098] In some embodiments, a reverberation algorithm may be used to perform dereverberation processing on the third sound signal corresponding to the omnidirectional microphone in the time-frequency domain and the fourth sound signal corresponding to the directional microphone in the time-frequency domain.
[0099] S605 , performing amplitude compensation on the fifth sound signal to obtain a seventh sound signal corresponding to the omnidirectional microphone after compensation; performing amplitude compensation and phase compensation on the sixth sound signal to obtain an eighth sound signal corresponding to the directional microphone.
[0100] In some embodiments, the amplitude compensation factor can be used to compensate the acoustic signal corresponding to the omnidirectional microphone in the time-frequency domain for amplitude. Furthermore, the amplitude compensation factor and phase compensation factor can be used to compensate the acoustic signal corresponding to the directional microphone in the time-frequency domain for amplitude and phase, respectively. The specific calculation process will be described below and will not be detailed here.
[0101] It should be understood that by performing amplitude compensation on the acoustic signal, it is possible to reduce actual signal amplitude loss caused by device structure and environmental objects, making the subsequently processed acoustic signal closer to the theoretical value. By performing phase compensation on the acoustic signal corresponding to the directional microphone, the phase difference between the acoustic signal corresponding to the directional microphone and the acoustic signal corresponding to the omnidirectional microphone can be set to a specific phase difference, thereby achieving acoustic signal enhancement results at specific angles.
[0102] Optionally, in some embodiments, the directivity of the MEMS microphone array may be corrected using a microphone directivity correction factor.
[0103] S606: Design a steering vector to obtain filter coefficients according to the seventh sound signal and the eighth sound signal.
[0104] S607: Process the acoustic signal in the time-frequency domain using a filter to obtain a filtered signal.
[0105] In some embodiments, a filter can be set according to the filter coefficient, and then the seventh sound signal after amplitude compensation and the eighth sound signal after amplitude compensation and phase compensation can be filtered using the filter. Specifically, after obtaining the filter, the seventh sound signal and the eighth sound signal in the time-frequency domain can be respectively compared with the filter coefficient h by frame. t (ω) and the multiplication results are superimposed to obtain a single-channel output signal containing the array algorithm processing result. The process of obtaining the filter coefficients can be found in the following description and will not be described in detail here.
[0106] S608: Perform post-processing and noise reduction on the filtered sound signal based on a statistical method.
[0107] In some embodiments, after obtaining a single-channel output signal, a single-channel post-processing noise reduction module based on a statistical model can be cascaded at the back end to perform noise reduction on the aforementioned acquired signal. The noise reduction module can be, for example, a noise reduction module based on Wiener filtering, or a noise reduction module based on logarithmic minimum mean square error, etc.
[0108] S609: transform the post-processed noise-reduced acoustic signal into the time domain according to an inverse short-time Fourier transform.
[0109] Afterwards, the acoustic signal in the time domain can be output.
[0110] It should be noted that by conducting actual measurements on the device that combines the MEMS microphone array and the back-end algorithm involved in the sound pickup method provided in the embodiment of the present application, an excellent unidirectional sound signal enhancement effect can be obtained, that is, the headphones that use the MEMS microphone array and the back-end algorithm can enhance the sound signal in the direction of the human mouth, while significantly reducing the influence of sound signals in other spatial directions on the pickup of human mouth sound signals.
[0111] For ease of understanding, the following combines a simplified physical model and a comparison of measured results with simulation results. Taking headphones with the above-mentioned MEMS microphone array structure as an example, the back-end algorithm processing process and the presented effects involved in the implementation of the device pickup method provided in the embodiment of the present application are introduced.
[0112] For example, Figure 7As shown, it is a schematic diagram of a physical model of a simplified MEMS microphone array structure in headphones provided by an embodiment of the present application. Among them, the omnidirectional microphone and the directional microphone are abstracted as black circles in the figure. The distance between the omnidirectional microphone and the directional microphone is δ, and the direction of the sound signal enhancement of the directional microphone is θ. Since the enhancement direction of the figure-8 directional microphone is bidirectional, for the sake of convenience, θ1 and θ2 are used here to represent the two angles corresponding to the sound signal enhancement. As shown in FIG. Figure 7 As shown in the figure-8 directional microphone directional diagram on the right, the acoustic signal enhancement direction of the directional microphone can be set to θ1 = 130° and θ2 = 310°.
[0113] by Figure 7 Taking the physical model shown in the figure as an example, after the omnidirectional microphone and the directional microphone pick up the sound signal, they can be converted from the time domain to the time-frequency domain through short-time Fourier transform. The sound signal after the short-time Fourier transform can be calculated using the following formula (1-1):
[0114]
[0115] Among them, y m,t (ω) represents the acoustic signal in the time-frequency domain after short-time Fourier transform; m represents the microphone array element number, which can be 1 or 2; t represents the sequence number of each frame of the acoustic signal after the acoustic signal is divided into T frames, which can be 1, 2, ..., T, where T is an integer greater than or equal to 2; x m,t is the tth frame of the speech signal collected by the mth microphone array element under ideal conditions; τ0=δ / c is the plane wave from 0 ° The delay between the two adjacent microphones when the incident direction is j 2 =-1,ω=2πf.
[0116] Afterwards, the acoustic signal y can be rectified by the dereverberation algorithm. m,t (ω) performs dereverberation processing to remove background noise and reverberation caused by sound reflections in the space from the sound signal. Exemplarily, the dereverberation algorithm may be a WPE (weighted predicition error) algorithm. The sound signal after dereverberation by the WPE algorithm may be as shown in the following formula (1-2):
[0117] y′ m,t (ω)=WPE(y m,t (ω)) (1-2)
[0118] Among them, y′ m,t (ω) represents the acoustic signal after dereverberation.
[0119] In some embodiments, in order to reduce the loss of the sound signal caused by the spatial environment or the device structure, the sound signal after dereverberation can also be amplitude compensated. For example, the following formula (1-3) can be used to calculate the sound signal after amplitude compensation for the sound signal picked up by the omnidirectional microphone, and the following formula (1-4) can be used to calculate the sound signal after amplitude compensation for the sound signal picked up by the directional microphone:
[0120]
[0121]
[0122] Among them, y 1，t (ω) represents the acoustic signal after amplitude compensation of the acoustic signal picked up by the omnidirectional microphone; y″ 2,t (ω) represents the acoustic signal after amplitude compensation of the acoustic signal picked up by the omnidirectional microphone; ω is the frequency point in the acoustic signal in the frequency domain; is the amplitude compensation factor corresponding to the omnidirectional microphone; is the amplitude compensation factor corresponding to the directional microphone; and It can be obtained by comparing the deviation between the measured value and the theoretical value of the acoustic signal at the maximum incident direction; is the phase compensation factor; e is a natural constant.
[0123] It should be understood that by performing amplitude compensation on the sound signal, the amplitude loss caused by the earphone structure, environmental factors, etc. during the transmission process can be reduced, making the sound signal closer to the theoretical value without external factors.
[0124] In some embodiments, phase compensation may be performed on the acoustic signal corresponding to the omnidirectional microphone and the acoustic signal corresponding to the directional microphone using the following formula (1-5):
[0125] y″ m,t (ω)=[y″ 1，t (ω),y″ 2,t (ω)] T =d(ω,θ)c(ω,θ)x″ m,t (ω) (1-5)
[0126] Among them, y m,t (ω) is the acoustic signal after phase compensation; is the steering vector; x″ m,t (ω) is the acoustic signal collected by the mth microphone array element in real conditions; c(ω,θ) is the amplitude correction factor of acoustic signals of different frequencies incident from different directions obtained in actual measurements.
[0127] It should be understood that by performing phase compensation on the acoustic signals in the time-frequency domain corresponding to the omnidirectional microphone and the directional microphone respectively, two acoustic signals with a specific phase difference can be obtained, which facilitates the subsequent enhancement of the acoustic signal in a specific direction.
[0128] For example, Figures 8A to 8C The following are schematic diagrams showing the theoretical values of the phase difference between the acoustic signals of the omnidirectional microphone and the directional microphone in the MEMS microphone array at sound frequencies of 2 kHz, 4 kHz, and 6 kHz, respectively. The compensation factor used in the embodiments of the present application for phase compensation of the acoustic signals of the omnidirectional microphone and the directional microphone can be obtained based on the theoretical value of the phase difference between the acoustic signals of the omnidirectional microphone and the directional microphone.
[0129] After that, the filter coefficients can be solved by the null pit method. In the embodiment of the present application, the filter coefficients can meet the following conditions: the acoustic signal enhancement direction θ is 310° (i.e. Figure 7 θ2 direction in the middle), and the specified attenuation direction is 130° (i.e. Figure 7 The gain at θ1 in the direction of the human mouth is 0. In other words, the filter coefficients must satisfy the requirement that the filter maximizes the acoustic signal only in the direction of the human mouth, without enhancing the acoustic signal in other directions. For example, the filter coefficients can be calculated using the following formulas (1-6) to (1-10):
[0130]
[0131]
[0132] Formula (1-6) and formula (1-7) can be uniformly expressed by the following formula (1-8):
[0133] Ah t (ω)=b (1-8)
[0134] in, b=[1 0] T .
[0135] Solving the above formula (1-8) can obtain the filter coefficient, which is shown in the following formula (1-9):
[0136] h t (ω)=A H (AA H ) -1 b (1-9)
[0137] Among them, h t (ω) is the filter coefficient; H is the matrix conjugate transpose.
[0138] In some embodiments, a corresponding filter can be designed based on the filter coefficients obtained by calculation. After obtaining the filter, the acoustic signal in the time-frequency domain can be compared with the filter coefficients h by frame. t (ω) and the multiplied frame results are superimposed to obtain a single-channel output signal containing the array algorithm processing result.
[0139] In some embodiments, after obtaining the above-mentioned single-channel output signal, a single-channel post-processing noise reduction module based on a statistical model can be cascaded at the back end to perform noise reduction on the aforementioned acquired signal. The noise reduction module can be, for example, a noise reduction module based on Wiener filtering, or a noise reduction module based on logarithmic minimum mean square error, etc.
[0140] It should be understood that the noise reduction effect can be further enhanced through the processing of the post-processing noise reduction module.
[0141] In some embodiments, after post-processing and noise reduction of the single-channel output signal in the time-frequency domain, an optimized time-domain speech signal can be output by inverse short-time Fourier transform to obtain a final output signal.
[0142] For example, Figure 9 As shown, when the sound frequency is 1KHz, the headphone pickup directivity diagram actually measured using the device sound pickup method provided in the embodiment of the present application and the theoretical directivity diagram.
[0143] in, Figure 9 The left figure is a schematic diagram of theoretical directivity. This theoretical directivity can be based on the MEMS microphone array structure model of omnidirectional microphones and directional microphones, and the acoustic signal enhancement direction is set to θ = 310 ° , obtained through software simulation at a sound frequency of 1 kHz. This simulates an ideal noise-free (completely anechoic) environment with no interference from other equipment structures or environmental objects. This simulates the sound pickup directivity achieved by a MEMS microphone array structure consisting of omnidirectional and directional microphones, combined with the back-end algorithm described above. Under these test conditions, the target directivity pattern is a first-order cardioid.
[0144] Figure 9 The figure on the right is a schematic diagram of the measured directivity. For example, the measurement environment can be: in a fully anechoic room, the test sound source is white noise, the MEMS microphone array is fixed on a pre-set turntable, and the turntable is controlled by a motor to rotate in 10° steps. After each rotation, the sound pickup signal of the MEMS microphone array is collected; then, a sound pickup directivity diagram representing the relationship between the sound pickup signal and the angle is obtained. According to Figure 9The results shown in the directivity diagram on the right indicate that the measured directivity is also a first-order cardioid directivity, and the sound pressure level at the null angle (130°) of this cardioid directivity diagram is more than 20dB lower than that at the maximum gain angle (310°).
[0145] Therefore, based on the aforementioned MEMS microphone array structure and corresponding back-end algorithm provided by the embodiment of the present application, it is possible to enhance unidirectional sound signals, and the sound pressure level difference between the maximum pickup gain angle and the minimum pickup gain angle is greater than a certain threshold. In other words, according to the device sound pickup method provided by the embodiment of the present application, the headset can enhance the sound pickup effect of the sound signal pointing in the direction of the human mouth, while reducing the gain of the sound signal in other directions, thereby reducing the interference of ambient noise on the voice signal in the headset call scenario.
[0146] For example, Figure 10 As shown, it is a schematic flow chart of a method for sound pickup by a device provided in an embodiment of the present application. The execution subject of the process may be a terminal device, which may include a MEMS microphone array, which may include an omnidirectional microphone and a directional microphone, and the direction of the maximum acoustic signal gain of the MEMS microphone array points to the first direction and the second direction. Among them, the first direction may correspond to the direction of the maximum acoustic signal enhancement mentioned above, such as the direction from the wireless headset to the human mouth (the corresponding angle is such as 310°), and the second direction may correspond to the direction 180° to the first direction mentioned above, such as the direction from the human mouth to the wireless headset (the corresponding angle is such as 130°). The process of this method may specifically include the following steps:
[0147] S1001: Acquire a first sound signal picked up by an omnidirectional microphone and a second sound signal picked up by a directional microphone.
[0148] S1002 : Using a short-time Fourier algorithm, transform the first acoustic signal in the time domain into a third acoustic signal in the time-frequency domain, and transform the second acoustic signal in the time domain into a fourth acoustic signal in the time-frequency domain.
[0149] S1003 : Obtain filter coefficients according to the third acoustic signal and the fourth acoustic signal, wherein the filter coefficients ensure that the direction in which the acoustic signal gain after filtering is the maximum is the first direction, and the direction in which the acoustic signal attenuation is the maximum is the second direction.
[0150] In some embodiments, the terminal device is a wireless headset, and the first direction is the direction from the wireless headset to the user's mouth when the user wears the wireless headset.
[0151] Among them, the direction from the wireless headset to the human mouth can specifically refer to the direction from the center point of the bottom of the wireless headset to the center point of the human mouth, or the direction from the second sound inlet of the directional microphone in the wireless headset (corresponding to the sound inlet 2 above) to the center point of the human mouth, etc.
[0152] In some embodiments, obtaining the filter coefficients based on the third sound signal and the fourth sound signal includes: performing dereverberation processing on the third sound signal to obtain a fifth sound signal; and performing dereverberation processing on the fourth sound signal to obtain a sixth sound signal; wherein the dereverberation processing is used to remove ambient noise in the third sound signal and / or the fourth sound signal; and obtaining the filter coefficients based on the fifth sound signal and the sixth sound signal.
[0153] In some embodiments, obtaining the filter coefficient based on the fifth sound signal and the sixth sound signal specifically includes: performing amplitude compensation on the fifth sound signal to obtain a seventh sound signal; and performing amplitude compensation and phase compensation on the sixth sound signal to obtain an eighth sound signal, wherein the amplitude compensation is used to compensate for the sound signal amplitude loss caused by the device structure and environmental objects of the terminal device, and the phase compensation makes the phase difference between the seventh sound signal and the eighth sound signal a preset phase difference; and obtaining the filter coefficient based on the seventh sound signal and the eighth sound signal.
[0154] In some embodiments, obtaining the filter coefficients according to the seventh sound signal and the eighth sound signal specifically includes obtaining the filter coefficients according to the seventh sound signal, the eighth sound signal, and a steering vector, where the steering vector is: in, θ represents the direction of the acoustic signal gain of the MEMS microphone array, j 2 =-1, ω = 2πf, f is the frequency point of the sound signal in the frequency domain, τ0 is the delay of the plane wave reaching two adjacent microphones when it is incident from the 0° direction, ⊙ is the Hadamard product of the matrix, c(ω,θ) is the correction factor obtained when the sound signals of different frequencies are incident from different directions; the filter coefficient h(ω) is: h(ω) = A H (AA H ) -1 b; where H is the matrix conjugate transpose, b=[1 0] T , T represents matrix transpose; θ max Indicates the direction of the acoustic signal gain (corresponding to Figure 7 θ2 direction in the max Indicates the direction in which the acoustic signal attenuates the most (corresponding to Figure 7 θ1 direction in the image).
[0155] In some embodiments, the method further includes: performing post-processing and noise reduction on the filtered acoustic signal; and transforming the post-processed and noise-reduced acoustic signal into a time domain using an inverse short-time Fourier transform.
[0156] In some embodiments, the MEMS microphone array satisfies that when the frequency is a preset frequency, the difference in sound pressure levels between the maximum angle of acoustic signal gain and the maximum angle of attenuation of the omnidirectional microphone is less than a first threshold, and the difference in sound pressure levels between the maximum angle of acoustic signal gain and the maximum angle of attenuation of the directional microphone is equal to or greater than a second threshold.
[0157] In some embodiments, the preset frequency is 1 KHz, the first threshold is 2 dB, and the second threshold is 15 dB.
[0158] In some embodiments, the MEMS microphone array includes an omnidirectional microphone and a directional microphone. The omnidirectional microphone is arranged at the upper end of the wireless headset, and the upper end is the end close to the sound outlet of the wireless headset. The directional microphone is arranged at the bottom end of the wireless headset. The direction of the maximum acoustic signal gain of the MEMS microphone array is close to or along the direction of the human mouth when the wireless headset is worn on the human ear.
[0159] In some embodiments, the first direction is 310° and the second direction is 130°.
[0160] In addition, an embodiment of the present application also provides a wireless headset, including a MEMS microphone array, wherein the MEMS microphone array includes an omnidirectional microphone and a directional microphone, the omnidirectional microphone is arranged at the upper end of the wireless headset, the upper end is the end close to the sound outlet of the wireless headset, and the directional microphone is arranged at the bottom end of the wireless headset. When the direction of the acoustic signal gain of the MEMS microphone array is maximum, the wireless headset points towards the direction of the human mouth when the wireless headset is worn on the human ear.
[0161] In some embodiments, the directional microphone includes a first sound inlet and a second sound inlet, the first sound inlet is arranged on the outer shell of the bottom end of the wireless headset, and the second sound inlet is arranged on the outer shell of the bottom end of the wireless headset, the first sound inlet and the second sound inlet are connected through a right-angle pipe, and the right-angle pipe includes a first pipe perpendicular to the direction of the earphone handle of the wireless headset, and a second pipe and a third pipe parallel to the direction of the earphone handle.
[0162] In some embodiments, the omnidirectional microphone includes a third sound inlet arranged in the outer shell of the earphone handle at the upper end of the wireless earphone, and the third sound inlet is connected to a sound inlet pipe perpendicular to the direction of the earphone handle.
[0163] According to the method for device sound pickup provided in the embodiment of the present application, by deploying a MEMS microphone array including an omnidirectional microphone and a directional microphone in the earphone, combined with a back-end algorithm, the earphone can enhance the sound signal in a single direction pointing to the human mouth, and the sound pressure level difference between the maximum gain angle and the minimum gain angle of the MEMS microphone array is greater than or equal to a certain threshold, which further optimizes the sound pickup effect in a single direction and improves the user's earphone usage experience.
[0164] It should be noted that the above embodiments of this application use headphones, especially wireless headphones, as an example to introduce the device structure of a MEMS microphone array including an omnidirectional microphone and a directional microphone, and the process of applying a back-end algorithm to achieve unidirectional acoustic signal enhancement. However, in actual applications, the MEMS microphone array structure (or hybrid microphone array structure) and corresponding algorithm provided in the embodiments of this application can also be applied to other terminal devices, and the embodiments of this application are not limited to this.
[0165] Based on the same technical concept, an embodiment of the present application also provides a terminal device, including a processor; a memory; the memory stores a computer program, and the computer program includes instructions. When the instructions are executed by the processor, the electronic device performs one or more steps in any of the above methods.
[0166] Based on the same technical concept, an embodiment of the present application also provides a chip system, which includes: a processing circuit, a receiving pin and a transmitting pin; wherein, the receiving pin, the transmitting pin and the processing circuit communicate with each other through an internal connection path, and the processing circuit executes one or more steps in any of the above methods to control the receiving pin to receive signals and control the transmitting pin to send signals.
[0167] Based on the same technical concept, an embodiment of the present application also provides a computer-readable storage medium, which stores computer-executable program instructions. When the computer-executable program instructions are executed on a computer, the computer or processor executes one or more steps in any of the above methods.
[0168] Based on the same technical concept, an embodiment of the present application also provides a computer program product containing instructions, wherein the computer program product includes computer program code. When the computer program code is run on a computer, the computer or processor executes one or more steps in any of the above methods.
[0169] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted via the computer-readable storage medium. The computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center via a wired (e.g., coaxial cable, optical fiber, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) method. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrated. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid state drive (SSD)).
[0170] Those skilled in the art will appreciate that all or part of the process steps in the above-described method embodiments can be implemented by a computer program instructing the relevant hardware. The program can be stored in a computer-readable storage medium, and when executed, the program can include the process steps in the above-described method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM or random access memory (RAM), magnetic disks, or optical disks.
[0171] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions within the technical scope disclosed in the present invention should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.
Claims
1. A method for picking up sound by a device, characterized in that: Applied to a terminal device, the terminal device includes a micro-electromechanical system (MEMS) microphone array, the MEMS microphone array includes an omnidirectional microphone and a directional microphone, and the direction of maximum acoustic signal gain of the MEMS microphone array points to a first direction and a second direction, the method comprising: Acquiring a first sound signal picked up by the omnidirectional microphone and a second sound signal picked up by the directional microphone; transforming the first acoustic signal in the time domain into a third acoustic signal in the time-frequency domain, and transforming the second acoustic signal in the time domain into a fourth acoustic signal in the time-frequency domain using a short-time Fourier algorithm; Filter coefficients are obtained according to the third acoustic signal and the fourth acoustic signal, wherein the filter coefficients ensure that the direction in which the acoustic signal after filtering has the maximum gain is the first direction and the direction in which the acoustic signal has the maximum attenuation is the second direction.
2. The method according to claim 1, characterized in that The terminal device is a wireless headset, and the first direction is the direction from the wireless headset to the user's mouth when the user wears the wireless headset.
3. The method according to claim 1 or 2, characterized in that The obtaining of filter coefficients according to the third sound signal and the fourth sound signal includes: performing dereverberation processing on the third sound signal to obtain a fifth sound signal; and performing dereverberation processing on the fourth sound signal to obtain a sixth sound signal; wherein the dereverberation processing is used to remove ambient noise in the third sound signal and / or the fourth sound signal; The filter coefficient is obtained according to the fifth sound signal and the sixth sound signal.
4. The method according to claim 3, characterized in that The obtaining of the filter coefficient according to the fifth sound signal and the sixth sound signal specifically includes: performing amplitude compensation on the fifth sound signal to obtain a seventh sound signal; and performing amplitude compensation and phase compensation on the sixth sound signal to obtain an eighth sound signal, wherein the amplitude compensation is used to compensate for a sound signal amplitude loss caused by a device structure and environmental objects of the terminal device, and the phase compensation is performed so that a phase difference between the seventh sound signal and the eighth sound signal is a preset phase difference; The filter coefficient is obtained according to the seventh sound signal and the eighth sound signal.
5. The method according to claim 4, characterized in that The obtaining of the filter coefficient according to the seventh sound signal and the eighth sound signal specifically includes: The filter coefficient is obtained according to the seventh sound signal, the eighth sound signal and a steering vector, where the steering vector is: in, θ represents the direction of the acoustic signal gain of the MEMS microphone array, j 2 = -1, ω = 2πf, f is the frequency point of the acoustic signal in the frequency domain, τ0 is the delay of the plane wave reaching two adjacent microphones when incident from the 0° direction, ⊙ is the Hadamard product of the matrix, and c(ω,θ) is the correction factor obtained by measuring acoustic signals of different frequencies incident from different directions; The filter coefficient h(ω) is: h(ω)=A H (AA H ) -1 b Where H is the matrix conjugate transpose, b=[1 0] T , T represents matrix transpose; θ max Indicates the direction of the acoustic signal gain, θ max Indicates the direction where the acoustic signal attenuation is greatest.
6. The method according to any one of claims 1 to 5, characterized in that The method further comprises: performing post-processing and noise reduction on the filtered acoustic signal; The post-processed noise-reduced acoustic signal is transformed into the time domain using an inverse short-time Fourier transform.
7. The method according to any one of claims 1 to 6, characterized in that The MEMS microphone array satisfies that when the frequency is a preset frequency, the difference in sound pressure level between the maximum angle of acoustic signal gain and the maximum angle of attenuation of the omnidirectional microphone is less than a first threshold, and the difference in sound pressure level between the maximum angle of acoustic signal gain and the maximum angle of attenuation of the directional microphone is equal to or greater than a second threshold.
8. The method according to claim 7, characterized in that The preset frequency is 1 KHz, the first threshold is 2 dB, and the second threshold is 15 dB.
9. The method according to any one of claims 2 to 8, characterized in that The MEMS microphone array includes an omnidirectional microphone and a directional microphone. The omnidirectional microphone is arranged at the upper end of the wireless headset, and the upper end is the end close to the sound outlet of the wireless headset. The directional microphone is arranged at the bottom end of the wireless headset. When the wireless headset is worn on the human ear, the direction of the acoustic signal gain of the MEMS microphone array is close to or along the direction of the human mouth.
10. The method according to any one of claims 1 to 9, characterized in that The first direction is 310°, and the second direction is 130°.
11. A wireless headset, characterized in that: It includes a MEMS microphone array, which includes an omnidirectional microphone and a directional microphone. The omnidirectional microphone is arranged at the upper end of the wireless headset, and the upper end is the end close to the sound outlet of the wireless headset. The directional microphone is arranged at the bottom end of the wireless headset. When the wireless headset is worn on the human ear, the direction in which the acoustic signal gain of the MEMS microphone array is maximum is close to or along the direction in which the wireless headset points to the direction of the human mouth.
12. The wireless headset according to claim 11, wherein: The directional microphone includes a first sound inlet and a second sound inlet. The first sound inlet is arranged on the outer shell of the bottom end of the wireless headset, and the second sound inlet is arranged on the outer shell of the bottom end of the wireless headset. The first sound inlet and the second sound inlet are connected through a right-angle pipe. The right-angle pipe includes a first pipe perpendicular to the direction of the earphone handle of the wireless headset, and a second pipe and a third pipe parallel to the direction of the earphone handle.
13. The wireless headset according to claim 11 or 12, characterized in that: The omnidirectional microphone includes a third sound inlet arranged in the outer shell of the earphone handle at the upper end of the wireless earphone, and the third sound inlet is connected to the sound inlet pipe perpendicular to the direction of the earphone handle.
14. A terminal device, characterized in that: include: processor; Memory; The memory stores a computer program, which includes instructions. When the instructions are executed by the processor, the electronic device performs the method according to any one of claims 1 to 10.
15. A chip system, characterized in that: The chip system includes a processing circuit, a receiving pin and a transmitting pin; wherein the receiving pin, the transmitting pin and the processing circuit communicate with each other through an internal connection path, and the processing circuit executes the method described in any one of claims 1 to 10 to control the receiving pin to receive a signal and control the transmitting pin to send a signal.
16. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer-executable program instructions, which, when executed on a computer, enable the computer to perform the method according to any one of claims 1 to 10.