Microphone array noise suppression method, system, hearing aid, and storage medium

Abstract

The application relates to a microphone array noise suppression method, system, hearing aid and storage medium, wherein the method comprises the following steps: pre-processing source voice signals obtained by a microphone array to obtain audio signals; the microphone array comprises at least three omnidirectional microphones; the audio signals are decomposed to obtain oblique forward audio signals and oblique backward audio signals opposite to the oblique forward audio signals in direction; the oblique forward audio signals refer to audio signals located in the forward direction of the microphone array and deviated from the center line of the microphone array; the oblique forward audio signals and the oblique backward audio signals are combined through an adaptive algorithm to output target audio signals subjected to noise suppression. Through the application, the audio signals can be decomposed into the oblique forward audio signals and the oblique backward audio signals, and then the adaptive filtering and weighting combination are used for noise suppression, so that the range of noise suppression is expanded, and the problem that the limited noise suppression range affects the noise reduction effect is solved.

Description

Technical Field

[0001] This application relates to the field of speech signal processing technology, and in particular to a microphone array noise suppression method, system, hearing aid, and storage medium. Background Technology

[0002] A hearing aid is an assistive device that helps people hear sounds. It amplifies sound signals to varying degrees and then feeds them back into the ear for use by people with hearing loss, thereby compensating for their hearing loss. In noisy environments, where there is interference from various sounds, the noise reduction function of a hearing aid is very important, as it relates to whether the user can hear and understand sounds clearly.

[0003] Most hearing aids currently use dual-microphone arrays to adaptively separate the required speech signal from ambient noise. However, for single-ear hearing aids, this only achieves a limited range of background noise suppression. In noisy environments, noise sources are widespread in the background sound sources, and the limited noise suppression range makes effective noise suppression impossible, thus affecting the noise reduction effect.

[0004] There is currently no effective solution to the problem that the limited noise suppression range in related technologies affects the noise reduction effect. Summary of the Invention

[0005] This embodiment provides a microphone array noise suppression method, system, hearing aid, and storage medium to address the problem that the limited noise suppression range in related technologies affects the noise reduction effect.

[0006] In a first aspect, this embodiment provides a microphone array noise suppression method, characterized by comprising:

[0007] The source speech signal acquired by the microphone array is preprocessed to obtain an audio signal; the microphone array includes at least three omnidirectional microphones;

[0008] The audio signal is decomposed to obtain a forward-facing audio signal and a backward-facing audio signal that is opposite in direction to the forward-facing audio signal; the forward-facing audio signal refers to the audio signal located in front of the microphone array and whose direction is offset from the center line of the microphone array.

[0009] An adaptive algorithm is used to weight and combine the forward and backward audio signals to output a target audio signal that has been noise-suppressed.

[0010] In some embodiments, the step of acquiring the source speech signal through a microphone array and preprocessing it to obtain the audio signal includes:

[0011] The source speech signal in the external environment is picked up through the omnidirectional microphone in the microphone array;

[0012] The source speech signal is processed by Fast Fourier Transform to obtain the audio signal.

[0013] In some embodiments, decomposing the audio signal to obtain a forward-facing audio signal and a backward-facing audio signal in the opposite direction to the forward-facing audio signal includes:

[0014] The audio signal is filtered by the beamforming of the microphone array based on two opposite preset directions to obtain the forward-facing audio signal and the backward-facing audio signal.

[0015] In some embodiments, the step of weighting and combining the forward and backward audio signals using an adaptive algorithm to output a noise-suppressed target audio signal includes:

[0016] Based on the feedback from the output audio signal, the adaptive weights are iteratively updated through adaptive filtering.

[0017] Based on the adaptive weights, the oblique backward audio signal is filtered out from the oblique forward audio signal to obtain the target audio signal.

[0018] In some embodiments, when the microphone array consists of three omnidirectional microphones, they are arranged in an equilateral triangle.

[0019] In some embodiments, the method further includes: when the location of the noise source in the audio signal is known, suppressing the noise using a preset filter.

[0020] In some embodiments, the suppression of the noise by a preset filter includes:

[0021] The preset filter is generated using the zero-point limiting method to suppress a specific amount of noise.

[0022] Secondly, this embodiment provides a microphone array noise suppression system, characterized in that it includes: an audio signal acquisition module, an audio signal decomposition module, and an adaptive filtering output module;

[0023] The audio signal acquisition module is used to preprocess the source speech signal acquired by the microphone array to obtain an audio signal; the microphone array includes at least three omnidirectional microphones;

[0024] The audio signal decomposition module is used to decompose the audio signal to obtain a forward-facing audio signal and a backward-facing audio signal with the opposite direction to the forward-facing audio signal; the forward-facing audio signal refers to the audio signal located in front of the microphone array and whose direction is offset from the center line of the microphone array.

[0025] The adaptive filtering output module is used to perform weighted combination of the oblique forward audio signal and the oblique backward audio signal using an adaptive algorithm, and output the target audio signal after noise suppression.

[0026] Thirdly, this embodiment provides a hearing aid, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the microphone array noise suppression method described in the first aspect above.

[0027] Fourthly, this embodiment provides a storage medium storing a computer program that, when executed by a processor, implements the microphone array noise suppression method described in the first aspect above.

[0028] Compared with related technologies, the microphone array noise suppression method, system, hearing aid, and storage medium provided in this embodiment preprocess the source speech signal acquired by the microphone array to obtain an audio signal; the microphone array includes at least three omnidirectional microphones; the audio signal is decomposed to obtain a forward-facing audio signal and a backward-facing audio signal with the opposite direction to the forward-facing audio signal; the forward-facing audio signal refers to the audio signal located in front of the microphone array and offset from the center line of the microphone array; the forward-facing audio signal and the backward-facing audio signal are weighted and combined using an adaptive algorithm to output a target audio signal with noise suppression. This method can decompose the audio signal into forward-facing and backward-facing audio signals, and then perform noise suppression through adaptive filtering and weighted combination, thereby expanding the range of noise suppression and solving the problem that the limited noise suppression range affects the noise reduction effect.

[0029] Details of one or more embodiments of this application are set forth in the following drawings and description to make other features, objects and advantages of this application more readily apparent. Attached Figure Description

[0030] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments of this application and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0031] Figure 1 This is a hardware structure block diagram of the terminal of the microphone array noise suppression method in one embodiment;

[0032] Figure 2 This is a flowchart of a microphone array noise suppression method in one embodiment;

[0033] Figure 3 This is a schematic diagram of the overall process of a microphone array noise suppression method in one embodiment;

[0034] Figure 4 This is a schematic diagram of an arbitrary microphone array distribution model in one embodiment;

[0035] Figure 5 This is a schematic diagram of an equilateral triangle microphone array distribution model in one embodiment;

[0036] Figure 6 This is a beammap of an oblique audio signal in one embodiment;

[0037] Figure 7 This is a flowchart of a microphone array noise suppression method in a preferred embodiment;

[0038] Figure 8 This is a block diagram of a microphone array noise suppression system in one embodiment.

[0039] In the diagram: 102, processor; 104, memory; 106, transmission device; 108, input / output device; 10, audio signal acquisition module; 20, audio signal decomposition module; 30, adaptive filtering output module. Detailed Implementation

[0040] To better understand the purpose, technical solution, and advantages of this application, the application is described and illustrated below in conjunction with the accompanying drawings and embodiments.

[0041] Unless otherwise defined, the technical or scientific terms used in this application shall have the general meaning as understood by one of ordinary skill in the art to which this application pertains. Words such as “a,” “an,” “an,” “the,” “the,” and “these,” used in this application, do not indicate quantitative limitation and may be singular or plural. The terms “comprising,” “including,” “having,” and any variations thereof used in this application are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or device that comprises a series of steps or modules (units) is not limited to the listed steps or modules (units) but may include steps or modules (units) not listed, or may include other steps or modules (units) inherent to such processes, methods, products, or devices. The terms “connected,” “linked,” and “coupled,” used in this application, are not limited to physical or mechanical connections but may include electrical connections, whether direct or indirect. The term “multiple” used in this application refers to two or more. The "and / or" operator describes the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: A alone, A and B simultaneously, and B alone. Typically, the character " / " indicates that the objects before and after it are in an "or" relationship. The terms "first," "second," and "third," etc., used in this application are merely for distinguishing similar objects and do not represent a specific ordering of the objects.

[0042] The method embodiments provided in this example can be executed in a terminal (such as a hearing aid), a computer, or a similar computing device. For example, it can run on a terminal. Figure 1 This is a hardware structure block diagram of the terminal of the microphone array noise suppression method in this embodiment. For example... Figure 1 As shown, a terminal may include one or more ( Figure 1 Only one is shown in the diagram. A processor 102 and a memory 104 for storing data are also included. The processor 102 may be, but is not limited to, a microprocessor (MCU) or a programmable logic device (FPGA). The terminal may also include a transmission device 106 for communication functions and an input / output device 108. Those skilled in the art will understand that… Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the terminal described above. For example, the terminal may also include components that are larger than... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown include, for example, a microphone array that picks up the source speech signal and a playback component that transmits the target audio signal to the user's ear canal.

[0043] The memory 104 can be used to store computer programs, such as application software programs and modules, like the computer program corresponding to the microphone array noise suppression method in this embodiment. The processor 102 executes various functional applications and data processing by running the computer programs stored in the memory 104, thereby implementing the aforementioned method. The memory 104 may include high-speed random access memory and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor 102, and these remote memories can be connected to the terminal via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0044] The transmission device 106 is used to receive or send data via a network. This network includes a wireless network provided by the terminal's communication provider. In one example, the transmission device 106 includes a Network Interface Controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission device 106 can be a Radio Frequency (RF) module used for wireless communication with the Internet.

[0045] A hearing aid is an assistive device that helps people hear sounds. It amplifies sound signals to varying degrees and then feeds them back into the ear for use by people with hearing loss, thereby compensating for their hearing loss. In noisy environments, where various sounds interfere, the noise reduction function of a hearing aid becomes very important, as it relates to whether the user can hear and understand sounds clearly.

[0046] Most hearing aids currently use dual-microphone arrays to adaptively separate the required speech signal from ambient noise. However, for single-ear hearing aids, this only achieves a limited range of background noise suppression. In noisy environments, noise sources are widespread in the background sound sources, and the limited noise suppression range makes effective noise suppression impossible, thus affecting the noise reduction effect.

[0047] To address the above issues, the following embodiments provide a microphone array noise suppression method, system, hearing aid, and storage medium. This method can acquire source speech signals through a microphone array including at least three omnidirectional microphones, decompose the audio signals into forward and backward audio signals, and then perform noise suppression through adaptive filtering and weighted combination, effectively expanding the range of noise suppression.

[0048] This embodiment provides a microphone array noise suppression method. Figure 2 This is a flowchart of the microphone array noise suppression method in this embodiment, as follows: Figure 2 As shown, the method includes the following steps:

[0049] Step S210: Preprocess the source speech signal acquired by the microphone array to obtain an audio signal; the microphone array includes at least three omnidirectional microphones.

[0050] Specifically, a microphone array is an array formed by arranging a group of omnidirectional microphones located at different positions in space according to a certain shape rule. It is a device for spatial sampling of sound signals propagating in space. An omnidirectional microphone is a microphone that has the same sensitivity to all angles and can pick up source speech signals evenly from all directions. For a microphone array including at least three microphones, the microphones can be arbitrarily distributed, such as linear arrays and planar arrays, and the distribution interval is not limited.

[0051] The source speech signal in the external environment is picked up by multiple omnidirectional microphones in the microphone array. The source speech signal includes the desired speech signal and noise in the environment. The audio signal is obtained after preprocessing the source speech signal. The preprocessing includes, but is not limited to, fast Fourier transform and other speech signal enhancement processes.

[0052] Step S220: Decompose the audio signal to obtain a forward-facing audio signal and a backward-facing audio signal that is opposite in direction to the forward-facing audio signal; the forward-facing audio signal refers to the audio signal located in front of the microphone array and whose direction is offset from the center line of the microphone array.

[0053] Specifically, the microphone array is focused both obliquely forward and obliquely backward, filtering and decomposing the input audio signal into obliquely forward and obliquely backward audio signals. The obliquely forward audio signal is located obliquely forward of the microphone array, while the obliquely backward audio signal, opposite to the forward audio signal, is located obliquely backward. Furthermore, the oblique audio signal is offset from the center line of the microphone array. Taking the angle directly in front of the microphone array as 0° as an example, the oblique forward direction of the microphone array typically refers to an angle range of 0° to 90° and 270° to 0°. For example, with a hearing aid worn in the right ear, focusing on a 330° forward audio signal and a 150° backward audio signal means offsetting the microphone array's center line by 30°.

[0054] For microphone arrays of different numbers and distributions, the audio signal can be adaptively decomposed into forward-facing audio signals in different directions and backward-facing audio signals in the opposite direction within the forward angle range of the microphone array. The angles of the forward-facing audio signals include, but are not limited to, 30° and 330°, and the corresponding angles of the backward-facing audio signals are 210° and 150°, respectively.

[0055] Step S230: The forward and backward audio signals are weighted and combined using an adaptive algorithm to output the target audio signal with noise suppression.

[0056] Specifically, the weights of the weighted combination of the forward and backward audio signals are iteratively optimized to remove the portion of the forward audio signal related to the backward audio signal through adaptive filtering. After inverse Fast Fourier Transform, the target audio signal with noise suppression is output. The adaptive filtering algorithms include, but are not limited to, LMS (Least Mean Square), NLMS (Normalized Least Mean Square), LS (The Derive of Least-Square), and RLS (Recursive Least Squares).

[0057] For the oblique forward and oblique backward audio signals obtained by decomposing microphone arrays of different numbers and distributions, adaptive noise reduction can be performed based on the oblique forward and backward audio signals in different directions, which can flexibly obtain a wider range of adaptive noise reduction suppression.

[0058] Figure 3 This is a schematic diagram of the overall process of the microphone array noise suppression method in this embodiment. Taking a microphone array including three omnidirectional microphones as an example, as follows... Figure 3 As shown, the source speech signal is picked up by each omnidirectional microphone in the microphone array, and the audio signal is obtained after passing through Fast Fourier Transform. By focusing the microphone array in the forward and backward directions respectively, the input audio signal is filtered and decomposed into forward and backward audio signals. The forward and backward audio signals are weighted and combined through an adaptive algorithm to obtain the output audio signal, and then the target audio signal is output after passing through the inverse Fast Fourier Transform.

[0059] The above steps use a microphone array with at least three omnidirectional microphones to pick up the source speech signal in the environment, decompose the audio signal into forward and backward audio signals, and then use adaptive filtering to weightedly combine the forward and backward audio signals to suppress noise in the backward audio signal. Compared with the noise suppression of dual-microphone arrays in existing hearing aids, this method can flexibly decompose the audio signal based on different microphone arrays to obtain forward and backward audio signals in different directions for adaptive noise reduction, effectively expanding the range of noise suppression and solving the problem that the limited noise suppression range affects the noise reduction effect.

[0060] In some embodiments, the preprocessing of the source speech signal acquired by the microphone array in step S210 to obtain the audio signal can be achieved through the following steps:

[0061] Step S211: Pick up the source speech signal in the external environment through the omnidirectional microphone in the microphone array.

[0062] Specifically, the source speech signal picked up by the microphone array includes the desired speech signal and ambient noise. For monohearing aids with built-in microphone arrays, especially in noisy and complex external environments, the source speech signal picked up includes ambient noise, which seriously affects the experience of users wearing hearing aids.

[0063] Step S212: Perform Fast Fourier Transform on the source speech signal to obtain the audio signal.

[0064] Specifically, the source speech signal is processed by a Fast Fourier Transform (FFT) to convert it from the time domain to the frequency domain, thus obtaining the audio signal. Additionally, before the FFT, the source speech signal can be enhanced through processes such as frame segmentation and windowing.

[0065] In this embodiment, the source speech signal acquired by the microphone array is preprocessed to obtain an audio signal in the frequency domain, so that the audio signal can be processed for a wider range of noise suppression in subsequent steps.

[0066] In some embodiments, the audio signal is decomposed in step S220 to obtain a forward-facing audio signal and a backward-facing audio signal in the opposite direction to the forward-facing audio signal. This can be achieved through the following steps:

[0067] Based on two opposite preset directions, the audio signal is filtered through the beamforming of the microphone array to obtain a forward-facing audio signal and a backward-facing audio signal.

[0068] Specifically, in the forward range of the microphone array, based on the selected preset direction, a preset filter is generated through the beamforming of the microphone array to filter the audio signal and obtain the forward audio signal. In addition, in the backward range of the microphone array, based on another selected opposite preset direction, a preset filter is also generated to filter the audio signal and obtain the backward audio signal.

[0069] For microphone arrays of varying numbers and distributions, the audio signal can be adaptively decomposed into forward-facing audio signals in different directions and backward-facing audio signals in the opposite direction within the forward-facing angle range of the microphone array, thereby achieving speech focusing in different directions. For example, if the direction of the forward-facing audio signal is offset from the centerline of the microphone array, i.e., at an angle of 330° or 30°, then the speech in the forward-facing direction of the microphone array can be focused.

[0070] In this embodiment, by using a preset filter to filter the input audio signal into oblique forward audio signal and oblique backward audio signal, speech focusing in different directions can be achieved, which has a high degree of freedom in noise reduction.

[0071] In some embodiments, the above step S230, which uses an adaptive algorithm to weight and combine the forward and backward audio signals to output a noise-suppressed target audio signal, can be achieved through the following steps:

[0072] In step S231, the adaptive weights are iteratively updated through adaptive filtering based on the feedback from the output audio signal.

[0073] Step S232: Based on adaptive weights, filter out the backward audio signal from the forward audio signal to obtain the target audio signal.

[0074] Specifically, the forward and backward audio signals are weighted and combined. The backward audio signal and related noise are adaptively filtered out from the forward audio signal, ensuring that the beam nulls and poles are always aligned with the noise source. Noise reduction is achieved by focusing the forward audio signal and filtering out related parts of the backward audio signal. After an inverse fast Fourier transform, the target audio signal is finally obtained. Simultaneously, the filtered audio signal is used as external feedback to iteratively optimize the adaptive weights using an adaptive filtering algorithm. For example, if the forward audio signal angle is 330° and the backward audio signal angle is 150°, after adaptive weighting and combination, the forward focusing angle remains unchanged, while the noise suppression angle can adaptively change from 90° to 210°.

[0075] Adaptive filtering algorithms include, but are not limited to, LMS, NLMS, LS and RLS. Considering the robustness and efficiency of the algorithm, the RLS algorithm is the preferred algorithm.

[0076] The following describes the calculation process of the RLS algorithm when the omnidirectional microphones in the microphone array are arranged in an equilateral triangle:

[0077]

[0078] Z(m,k)=X(m,k)-W(m-1,k)·Y(m,k),

[0079] W(m,k)=W(m-1,k)+g(m,k)·Z(m,k) * ,

[0080] P(m,k)=λ -1 ·P(m-1,k)-λ -1 ·g(m,k)·Y(m,k) * P(m-1,k), where the forward and backward audio signals are defined as X(m,k) and Y(m,k) respectively, m is the frame index, and k is the frequency domain subband index; g(m,k) is the gain vector; the adaptive weight is defined as W(m,k), and the output audio signal is Z(m,k); P(m,k) represents the intermediate variable matrix; λ represents the forgetting factor, used to enhance the stability of the algorithm; * represents the conjugate operation.

[0081] In this embodiment, the adaptive algorithm continuously updates the weights of the weighted combination, ensuring that the beam nulls and poles are always aligned with the noise source, thus achieving a wider adaptive noise suppression range.

[0082] The following embodiments further illustrate the distribution design of omnidirectional microphones in a microphone array.

[0083] In some of these embodiments, the omnidirectional microphones in the microphone array are arbitrarily distributed.

[0084] Specifically, any number of microphones in the microphone array can be distributed arbitrarily, such as linear arrays and planar arrays, and there is no restriction on the distribution interval.

[0085] Suppose there are Q omnidirectional microphones on a plane, and the coordinates of the position r of each microphone q in the coordinate system are... q Defined as:

[0086] r q =[x q ,y q ]=ρ q [cosφ q ,sinφ q ],q=1,2,…,Q,

[0087] Where, ρ q φ represents the polar diameter of microphone q in the polar coordinate system. q This represents the polar angle of microphone q in the polar coordinate system. Figure 4 This is a schematic diagram of an arbitrary microphone array distribution model in this embodiment. Figure 4 Only two omnidirectional microphones are shown, with coordinates (x1, y1) and (x2, y2) respectively.

[0088] For a plane wave incident in the θ direction, the incident beam l is defined as follows:

[0089] l = [l x ,l y ]=-2πf / c[cosθ,sinθ],

[0090] Where f is the frequency of the plane wave and c is the speed of sound.

[0091] With the origin position as a reference, the delay of each microphone in receiving the plane wave is as follows:

[0092]

[0093] Where, τ0=ρ q / c;ω=2πf。

[0094] The delay vector of the microphone array is defined as:

[0095]

[0096] Define a set of filter coefficients H Q (ω) represents the filter response coefficient corresponding to the Qth microphone, from which the frequency response of the beam is obtained as:

[0097]

[0098] Where B(ω,θ) represents the frequency response of a plane wave incident at an angle θ with a frequency of ω / 2π.

[0099] Considering the practical use of hearing aids, the required signal will come from directly in front, therefore the following distortion-free conditions should always be met:

[0100] B(ω,θ=0°)=1.

[0101] When some noise comes from θ i When the angle is reached, the following zero-point constraint is applied to attenuate noise:

[0102]

[0103] in, N is the noise level; θ i =θ1,θ2,…,θ N (N≤Q-1),H0(ω)=1,Q is the number of microphones.

[0104] There are N+1 unknown vector equations that coexist with zero-point constraints and distortion-free conditions. For an array with Q microphones, the maximum number of noises that can be suppressed is Q-1.

[0105] When N+1 < Q and the location of the noise source is unknown, the filter coefficients h(ω) are calculated using the least 2 norm as follows:

[0106] h(ω)=D H (DD H ) -1 β,

[0107] in, D H This represents the conjugate transpose of the delay matrix.

[0108] In this embodiment, for any microphone array with an arbitrary distribution and an unlimited number of microphones, the frequency response of the microphone array to plane wave incident can be obtained. After the above zero-point constraint and least norm 2 calculation, the filter coefficients of the oblique audio signal can be obtained for noise suppression.

[0109] In some of these embodiments, when the microphone array consists of three omnidirectional microphones, they are arranged in an equilateral triangle.

[0110] Figure 5 This is a schematic diagram of the equilateral triangle microphone array distribution model in this embodiment, as shown below. Figure 5 As shown, the three omnidirectional microphones are spaced equidistantly by a distance d. In the coordinate system, the coordinates of the three microphones (x1, y1), (x2, y2), and (x3, y3) are respectively:

[0111] The structural advantage of an equilateral triangle array lies in the stability of its array shape. This array can maintain stable beamforming for focusing in any direction, and the array's directivity index remains consistent when focusing at different angles. At the same time, this microphone array can better suppress white noise in white noise scenarios.

[0112] An equilateral triangular microphone array, with an additional microphone compared to a traditional dual-microphone array hearing aid, increases the noise reduction freedom of the hearing aid algorithm, enabling simultaneous speech focusing in any direction.

[0113] Figure 6 This is the beammap of the oblique audio signal in this embodiment, such as... Figure 6 As shown, in an equilateral triangular microphone array, the left and right images represent beamforms with a focusing angle of 330° forward and 150° backward, respectively. By setting θ1 to 330° and θ2 to 150° in the above calculations, the filter coefficients corresponding to the beamform in the left image can be obtained; similarly, by setting θ1 to 150° and θ2 to 330°, the filter coefficients corresponding to the beamform in the right image can be obtained.

[0114] Figure 6The multiple cardioids in the two beam diagrams represent different frequencies of the incident plane wave, respectively. The frequency response to different orientations demonstrates the good frequency invariance of the equilateral triangular microphone array. 0dB, -10dB, -20dB, and -30dB represent different frequency responses.

[0115] Taking a single hearing aid worn in the right ear as an example, considering the noise attenuation effect of the "head shadow" obstruction, when the forward audio signal angle is 330°, its noise attenuation range is 60° to 240°. Adding the noise reduction range due to head obstruction, the final noise reduction range is 60° to 270°.

[0116] This embodiment provides a method for decomposing audio signals into oblique forward and backward audio signals when the microphone array includes at least three omnidirectional microphones, thereby achieving speech focusing in any direction. A wider noise reduction range and speech focusing capability are obtained through the oblique audio signals. Furthermore, the stability of the equilateral triangular microphone array structure enables better noise reduction performance and a wider noise reduction range.

[0117] In the above embodiments, a wider adaptive noise suppression was achieved for situations where the noise source in the environment is unknown. In the following embodiments, a scheme for noise suppression in any direction can be achieved by zero-point limitation when the noise source in the audio signal is known.

[0118] In some of these embodiments, when the location of the noise source in the audio signal is known, the noise is suppressed by a preset filter.

[0119] Furthermore, noise is suppressed by using a preset filter, including: generating a preset filter using the zero-point limiting method to suppress a specific amount of noise.

[0120] Specifically, based on the distortion-free condition B(ω,θ=0°)=1 and the following zero-point constraint in the above embodiments:

[0121]

[0122] in, N is the number of plane waves; θ i =θ1,θ2,…,θ N (Q≤Q-1),H0(ω)=1, where Q is the number of microphones.

[0123] There are N+1 unknown vector equations that coexist with zero-point constraints and distortion-free conditions. For a microphone array with Q microphones, the maximum number of noise sources that can be suppressed is Q-1. That is, when the microphone array includes three omnidirectional microphones, the estimation and suppression of multiple noise sources in any direction can be achieved by using the phase difference between any two microphones near the noise source or by using a filter generated by the zero-point constraint method.

[0124] The present embodiment will now be described and illustrated through preferred embodiments.

[0125] Figure 7 This is a flowchart of the microphone array noise suppression method according to a preferred embodiment, as follows: Figure 7 As shown, the microphone array noise suppression method includes the following steps:

[0126] Step S710: Perform a fast Fourier transform on the source speech signal acquired by the microphone array to obtain an audio signal; wherein the microphone array includes three omnidirectional microphones arranged in an equilateral triangle.

[0127] In step S720, the audio signal is filtered by beamforming of the microphone array based on two opposite preset directions to obtain a forward-facing audio signal and a backward-facing audio signal.

[0128] In step S730, under the feedback of the output audio signal, the adaptive weights are iteratively updated through adaptive filtering; based on the adaptive weights, the oblique backward audio signal is filtered out from the oblique forward audio signal, so that the beam null and pole are always aligned with the noise source, and the target audio signal is obtained.

[0129] Step S740: When the location of the noise source in the audio signal is known, a filter is generated to suppress the noise using the zero-point constraint method.

[0130] The method provided in this preferred embodiment leverages the structural advantages of an equilateral triangle array and increases the noise reduction freedom of the hearing aid algorithm, enabling simultaneous speech focusing in any direction. By decomposing the audio signal into forward and backward audio signals, and then using adaptive filtering and weighted combination of these signals, noise suppression is achieved by suppressing the backward audio signal. Compared to the noise suppression of dual-microphone arrays in existing hearing aids, this method flexibly decomposes the audio signal based on different microphone arrays to obtain forward and backward audio signals from different directions for adaptive noise reduction. This effectively expands the noise suppression range and solves the problem of limited noise suppression range affecting noise reduction performance.

[0131] Furthermore, taking the deployment of the microphone array and noise suppression algorithm in this preferred embodiment on a chip with a main frequency of 15.36 MHz as an example, it actually requires 12216 cycles. This results in low power consumption on the chip, which is suitable for the low power endurance configuration of hearing aids.

[0132] The noise suppression method based on an equilateral triangle microphone array in this preferred embodiment can meet the needs of hearing aid users in terms of speech focusing ability, noise attenuation range, and low power consumption and durability of hearing aids.

[0133] It should be noted that the steps shown in the above flowchart or the flowchart in the accompanying figures can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here. For example, step S740 is for cases where the location of the noise source is known, and is not limited to any specific step in which it is executed.

[0134] This embodiment also provides a microphone array noise suppression system for implementing the above embodiments and preferred embodiments; details already described will not be repeated. The terms "module," "unit," "subunit," etc., used below refer to combinations of software and / or hardware that perform a predetermined function. Although the system described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0135] Figure 8 This is a structural block diagram of the microphone array noise suppression system in this embodiment, as shown below. Figure 8 As shown, the system includes: an audio signal acquisition module 10, an audio signal decomposition module 20, and an adaptive filtering output module 30.

[0136] The audio signal acquisition module 10 is used to preprocess the source speech signal acquired by the microphone array to obtain an audio signal; the microphone array includes at least three omnidirectional microphones.

[0137] The audio signal decomposition module 20 is used to decompose the audio signal to obtain a forward-facing audio signal and a backward-facing audio signal that is opposite in direction to the forward-facing audio signal; the forward-facing audio signal refers to the audio signal located in front of the microphone array and whose direction is offset from the center line of the microphone array.

[0138] The adaptive filtering output module 30 is used to perform weighted combination of the oblique forward audio signal and the oblique backward audio signal through an adaptive algorithm, and output the target audio signal after noise suppression.

[0139] The system provided in this embodiment picks up source speech signals from the environment using a microphone array with at least two three-way microphones, decomposes the audio signals into forward and backward audio signals, and then uses adaptive filtering to weightedly combine the forward and backward audio signals to suppress noise in the backward audio signal. Compared with the noise suppression of dual-microphone arrays in existing hearing aids, this system can flexibly decompose audio signals based on different microphone arrays to obtain forward and backward audio signals in different directions for adaptive noise reduction, effectively expanding the range of noise suppression and solving the problem that the limited noise suppression range affects the noise reduction effect.

[0140] In some embodiments, the audio signal acquisition module 10 is further configured to pick up source speech signals in the external environment through omnidirectional microphones in the microphone array; and to perform fast Fourier transform processing on the source speech signals to obtain audio signals.

[0141] In some embodiments, the audio signal decomposition module 20 is further configured to filter the audio signal based on two opposite preset directions by beamforming of the microphone array to obtain a forward-facing audio signal and a backward-facing audio signal.

[0142] In some embodiments, the adaptive filtering output module 30 is further configured to update the adaptive weights iteratively through adaptive filtering based on the feedback of the output audio signal; and to filter out the backward audio signal from the forward audio signal based on the adaptive weights to obtain the target audio signal.

[0143] In some embodiments, the system further includes an arbitrary direction noise suppression module, used to suppress noise by means of a preset filter when the direction of the noise source in the audio signal is known.

[0144] It should be noted that the above modules can be functional modules or program modules, and can be implemented through software or hardware. For modules implemented through hardware, the above modules can reside in the same processor; or the above modules can be located in different processors in any combination.

[0145] This embodiment also provides a hearing aid, including a memory and a processor, wherein the memory stores a computer program and the processor is configured to run the computer program to perform the steps in any of the above method embodiments.

[0146] Optionally, the hearing aid may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor and the input / output device is connected to the processor.

[0147] It should be noted that the specific examples in this embodiment can refer to the examples described in the above embodiments and optional implementations, and will not be repeated in this embodiment.

[0148] Furthermore, in conjunction with the microphone array noise suppression method provided in the above embodiments, this embodiment can also provide a storage medium for implementation. The storage medium stores a computer program; when executed by a processor, the computer program implements any of the microphone array noise suppression methods described in the above embodiments.

[0149] It should be understood that the specific embodiments described herein are merely illustrative of the application and not intended to limit it. All other embodiments derived by those skilled in the art based on the embodiments provided in this application without inventive effort are within the scope of protection of this application.

[0150] Obviously, the accompanying drawings are merely some examples or embodiments of this application. Those skilled in the art can apply this application to other similar situations based on these drawings without any creative effort. Furthermore, it is understood that although the work done in this development process may be complex and lengthy, for those skilled in the art, certain design, manufacturing, or production modifications made based on the technical content disclosed in this application are merely conventional technical means and should not be considered as insufficient disclosure of this application.

[0151] The term "embodiment" in this application refers to a specific feature, structure, or characteristic described in connection with an embodiment that may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily imply the same embodiment, nor does it imply that it is mutually exclusive with or alternative to other embodiments. It will be clearly or implicitly understood by those skilled in the art that the embodiments described in this application may be combined with other embodiments without conflict.

[0152] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of patent protection. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the appended claims.

Claims

1. A microphone array noise suppression method, characterized in that, include: The source speech signal acquired by the microphone array is preprocessed to obtain the audio signal; The microphone array includes at least three omnidirectional microphones; The audio signal is decomposed to obtain a forward-facing audio signal and a backward-facing audio signal that is opposite in direction to the forward-facing audio signal; the forward-facing audio signal refers to the audio signal located in front of the microphone array and whose direction is offset from the center line of the microphone array. This includes the following steps: The audio signal is filtered by the beamforming of the microphone array based on two opposite preset directions to obtain the forward-facing audio signal and the backward-facing audio signal. An adaptive algorithm is used to weight and combine the forward and backward audio signals to output a noise-suppressed target audio signal; this includes the following steps: Based on the feedback from the output audio signal, the adaptive weights are iteratively updated through adaptive filtering. Based on the adaptive weights, the oblique backward audio signal is filtered out from the oblique forward audio signal to obtain the target audio signal.

2. The microphone array noise suppression method according to claim 1, characterized in that, The preprocessing of the source speech signal acquired by the microphone array to obtain the audio signal includes: The source speech signal in the external environment is picked up through the omnidirectional microphone in the microphone array; The source speech signal is processed by Fast Fourier Transform to obtain the audio signal.

3. The microphone array noise suppression method according to claim 1, characterized in that, When the microphone array consists of three omnidirectional microphones, they are arranged in an equilateral triangle.

4. The microphone array noise suppression method according to claim 1, characterized in that, Also includes: When the location of the noise source in the audio signal is known, the noise is suppressed by a preset filter.

5. The microphone array noise suppression method according to claim 4, characterized in that, The noise suppression via a preset filter includes: The preset filter is generated using the zero-point limiting method to suppress a specific amount of noise.

6. A microphone array noise suppression system, characterized in that, include: Audio signal acquisition module, audio signal decomposition module, and adaptive filtering output module; The audio signal acquisition module is used to preprocess the source speech signal acquired by the microphone array to obtain an audio signal; the microphone array includes at least three omnidirectional microphones; The audio signal decomposition module is used to decompose the audio signal to obtain a forward-facing audio signal and a backward-facing audio signal with the opposite direction to the forward-facing audio signal; the forward-facing audio signal refers to the audio signal located in front of the microphone array and whose direction is offset from the center line of the microphone array. This includes the following steps: The audio signal is filtered by the beamforming of the microphone array based on two opposite preset directions to obtain the forward-facing audio signal and the backward-facing audio signal. The adaptive filtering output module is used to perform a weighted combination of the oblique forward audio signal and the oblique backward audio signal using an adaptive algorithm, and output a target audio signal with noise suppression; wherein the following steps are included: Based on the feedback from the output audio signal, the adaptive weights are iteratively updated through adaptive filtering. Based on the adaptive weights, the oblique backward audio signal is filtered out from the oblique forward audio signal to obtain the target audio signal.

7. A hearing aid, comprising a memory and a processor, characterized in that, The memory stores a computer program, and the processor is configured to run the computer program to perform the microphone array noise suppression method according to any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the microphone array noise suppression method according to any one of claims 1 to 5.

Images