Method for reducing noise of multiple microphone cores of microphone

Abstract

The invention discloses a multi-microphone-core noise reduction method for a microphone, and the method comprises the steps: obtaining a multi-channel sound signal through a plurality of microphone cores of the microphone, and transmitting the sound signal to a beam forming system; the beam forming system can output processed multi-channel sound signals and transmit the processed multi-channel sound signals to the filtering control system and the self-adaptive noise elimination system. The filtering control system can output processed multichannel sound signals and transmit the processed multichannel sound signals to the self-adaptive noise elimination system and the single-channel noise reduction system; the adaptive noise elimination system can convert multi-channel sound signals into single-channel sound signals and transmit the single-channel sound signals to the single-channel noise reduction system, background noise can be effectively suppressed through technical means of beam forming, filtering control, adaptive noise elimination and the like, voice signal definition and quality are improved, and the noise reduction effect is improved. And the requirements of modern communication and audio processing equipment are met.

Description

Technical Field

[0001] The present invention relates to the field of audio processing technology, and in particular to a method for reducing noise of a microphone using multiple microphone cores. Background Art

[0002] With the continuous development of modern communication technology and audio processing technology, microphones are increasingly used as audio input devices in various electronic devices. However, in actual applications, microphones are often affected by environmental noise, resulting in reduced recording or communication quality. Especially in noisy environments, such as outdoors, public places or workplaces, how to effectively eliminate or reduce background noise has become an urgent problem to be solved.

[0003] Currently, no effective solution has been proposed for the problems in the related technologies. Summary of the invention

[0004] In view of the problems in the related art, the present invention proposes a method for microphone multi-microphone core noise reduction to overcome the above-mentioned technical problems existing in the existing related art.

[0005] To this end, the specific technical solution adopted by the present invention is as follows:

[0006] A method for noise reduction using multiple microphone cores of a microphone, wherein the microphone used in the method has two or more microphone cores, and the method comprises obtaining a multi-channel sound signal through the multiple microphone cores of the microphone, and transmitting the sound signal to a beamforming system; the beamforming system can output the processed multi-channel sound signal and transmit it to a filtering control system and an adaptive noise cancellation system; the filtering control system can output the processed multi-channel sound signal and transmit it to an adaptive noise cancellation system and a single-channel noise reduction system; the adaptive noise cancellation system can convert the multi-channel sound signal into a single-channel sound signal and transmit it to the single-channel noise reduction system.

[0007] As a further solution of the present invention, a multi-channel sound signal is received from a microphone, wherein the sound signal of one channel is transmitted to a DC blocking filter, and after processing, is transmitted to a de-emphasis filter, and finally a main channel sound signal is output and transmitted to the other systems; the sound signals of other channels are transmitted to a DC blocking filter, and after processing, are transmitted to a gain and phase adjustment filter, and after processing again, are transmitted to a de-emphasis filter, and finally a reference channel sound signal is output and transmitted to the other systems.

[0008] As a further solution of the present invention, multi-channel sound signals are received, transmitted to the voice activity detection unit, and transmitted to the suppression logic unit after processing; the main channel sound signal also needs to be transmitted to the main channel detector, and transmitted to the suppression logic unit after processing; the reference channel sound signal also needs to be transmitted to the reference channel detector, and transmitted to the suppression logic unit after processing; the suppression logic unit outputs a control signal and transmits it to the other systems.

[0009] As a further solution of the present invention, multi-channel sound signals and control signals are received, wherein the main channel sound signal is transmitted to the delay unit and transmitted to the adder after processing; the reference channel sound signal and control signal are transmitted to the adaptive filter and transmitted to the adder after processing; the adder outputs a single-channel sound signal and transmits it to the other systems.

[0010] As a further solution of the present invention, a single-channel sound signal and a control signal are received, transmitted to a linear filter, and a sound signal having a significant noise reduction effect is output.

[0011] As a further solution of the present invention, the microphone structure includes an outer shell and a circuit board assembly fixed in the outer shell; the outer shell is made of high-strength and lightweight material and has good durability and protection performance; the circuit board assembly includes multiple circuits and components, which can support the noise reduction function and other related functions of the microphone.

[0012] As a further solution of the present invention, the outer shell has sound receiving holes, each of which has a diameter of 2 mm or less and is designed to effectively collect sound signals; the arrangement positions of the sound receiving holes are precisely calculated and optimized to ensure the collection effect of sound signals; the outside is covered with a mesh, which is made of highly sound-transparent material, which can prevent dust and impurities from entering while not affecting the transmission quality of sound.

[0013] As a further solution of the present invention, the circuit board assembly includes two or more microphone core arrays, one of which is defined as the microphone core closest to the speech source and is used to output the main channel sound signal; the other microphone cores are used to receive ambient noise and reference signals to assist in noise reduction processing; the circuit board assembly also includes an adaptive filter, a control unit and a signal processing unit, etc., which can complete the noise reduction method.

[0014] The beneficial effects of the present invention are: through technical means such as beamforming, filter control and adaptive noise cancellation, the background noise can be effectively suppressed, the clarity and quality of the voice signal can be improved, and the needs of modern communication and audio processing equipment can be met. The microphone used in this method has two or more microphone cores, and a multi-channel sound signal is obtained through multiple microphone cores and transmitted to the beamforming system. The beamforming system outputs the processed multi-channel sound signal, which is transmitted to the filter control system and the adaptive noise cancellation system, and finally outputs a sound signal with a significant noise reduction effect. The microphone structure that implements this method includes an outer shell and a circuit board assembly fixed therein, including multiple circuits and components, supporting the noise reduction function and other related functions of the microphone. This patent is applicable to smart devices, audio equipment, public place communication equipment, voice recognition systems, etc., to provide a high-quality user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0016] Figure 1 This is the overall appearance of the microphone.

[0017] Figure 2 This is a schematic diagram of the relationship between the microphone core, human voice and noise.

[0018] Figure 3 is a flow chart of the noise reduction method.

[0019] Figure 4 It is a schematic diagram of the overall system structure of the noise reduction method.

[0020] Figure 5 It is a schematic diagram of the beamforming system structure of the noise reduction method.

[0021] Figure 6 It is a schematic diagram of the filter control system structure of the noise reduction method.

[0022] Figure 7 It is a schematic diagram of the structure of an adaptive noise cancellation system of the noise reduction method.

[0023] Figure 8 Schematic diagram of the structure of a single-channel noise reduction system of the noise reduction method. DETAILED DESCRIPTION

[0024] To further illustrate each embodiment, the present invention provides drawings, which are part of the disclosure of the present invention and are mainly used to illustrate the embodiments and can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. With reference to these contents, ordinary technicians in the field should be able to understand other possible implementations and advantages of the present invention. The components in the figures are not drawn to scale, and similar component symbols are generally used to represent similar components.

[0025] According to an embodiment of the present invention, a method for noise reduction using multiple microphone cores is provided.

[0026] Please refer to the instruction manual Figure 1 and Figure 2 A sound receiving device with two or more microphone cores, one of which is confirmed to be closer to the expected voice source, can utilize the noise reduction method through wired or wireless transmission.

[0027] For the noise reduction method proposed in this patent, its workflow is as follows: Figure 3 As shown, the system structure is as Figure 4 The detailed structure of each part is shown in Figure 5 , Figure 6 , Figure 7 , Figure 8 shown.

[0028] Figure 1 The overall appearance of the microphone is shown, where 11 is the outer shell, 12 is the microphone core close to the expected voice source, 13 is the microphone core slightly farther from the expected voice source, and 14 is an optional microphone core slightly farther from the expected voice source. The positions of 12, 13, and 14 are not fixed and do not need to be adjusted. Figure 1 In fact, the noise reduction method proposed in this patent can also be applied to many other types of microphones, but they must have two or more microphone cores.

[0029] Figure 2 The figure shows the relationship between the microphone core and the reception of human voice and noise during use, wherein 21 is the human voice source (i.e. the desired voice source), 22 is the noise source (i.e. the undesired voice source), 23 is the microphone, 24 and 25 are the microphone cores on the microphone, and 26 is an optional microphone core.

[0030] Figure 3 The flowchart of the noise reduction method is shown. 31 indicates that the microphone receives the sound signal, corresponding to Figure 4 401, 402 and 403 in. 32 corresponds to 33 Figure 4 34 corresponds to the operation of the filter control system and the adaptive noise cancellation system, and 35 corresponds to the operation of the filter control system and the single channel noise reduction system.

[0031] Figure 4 The overall system architecture of the noise reduction method is shown, including a beamforming system 41, a filter control system 42, an adaptive noise cancellation system 43, and a single-channel noise reduction system 44. The system input includes sound signals received by a first microphone core 401, a second microphone core 402, and an optional third microphone core 403, which may contain desired audio (such as human voice) and undesired audio (such as background noise); the system output is a sound signal 410 that achieves a significant noise reduction effect.

[0032] After receiving the sound signal, the beam forming system 41 (will Figure 5 The beamforming system creates main channels 404a / 404b and reference channels 405a / 405b (and optionally 406a / 406b) by adjusting the gain and phase of the signal. The main channel signal usually contains a higher signal-to-noise ratio, while the reference channel signal is mainly used to capture environmental noise. In this process, the beamforming system focuses on the sound in a specific direction by controlling the gain and phase of the signal, thereby enhancing the target sound and suppressing noise in other directions.

[0033] The filter control system 42 plays an important role in the whole process. It analyzes the signal characteristics from the beamformers 404a and 405a (and optionally 406a), and generates control signals 407 and 408 to manage the filtering process of the adaptive noise cancellation system 43 and the mono noise reduction system 44. In this way, the system can dynamically adapt to the noise changes in the environment, ensuring high-quality audio output in different acoustic environments.

[0034] The main channel 404b and reference channel 405b (and optionally 406b) signals from the beamformer are then input to the adaptive noise cancellation system 43 (which will be Figure 7 The system performs a first stage of noise reduction processing on the signal. The adaptive filter dynamically adjusts the filter parameters to reduce the noise in the main channel and outputs a preliminarily processed signal 409. This signal continues to the mono noise reduction system 44 (described in detail in Figure 8 ), the signal is further processed using linear filtering techniques to further reduce the remaining noise and generate a final output signal 410, which will mainly contain the desired audio with significantly reduced background noise.

[0035] Figure 5 The beamforming system architecture is presented, i.e. Figure 4 The 41 part in the system realizes beamforming in a specific direction by performing gain and phase adjustment on multiple input signals. The system input is the first microphone core signal 501 (corresponding to Figure 4401 in), the second microphone signal 502 (corresponding to Figure 4 402) and the optional third microphone core signal 503 (corresponding to Figure 4 403 in the system); the system outputs are main channel signals 509a / 509b, reference channel signals 510a / 510b and optional 511a / 511b.

[0036] The first sound channel 501 first passes through the DC blocking filter 51 to remove the DC component and very low-frequency components, and then enters the de-emphasis filter 52 to adjust the frequency response of the signal to make the frequency response of the sound flatter, and finally forms the main channel output signal 509a / 509b, of which 509a will be input into the adaptive noise cancellation system and 509b will be input into the filter control system.

[0037] The second sound channel 502 first passes through a DC blocking filter 53 to remove the DC component, and then enters a gain and phase adjustment filter 54, which adjusts the gain and phase of the signal to ensure the desired match with the first microphone signal. The signal then passes through a de-emphasis filter 55 to adjust the frequency response to make it flatter, and finally forms a reference channel output signal 510a / 510b, which are also input into the adaptive noise cancellation system and the filter control system respectively.

[0038] In addition, the optional sound channel 503 can also be processed in the same manner as 502.

[0039] Figure 6 The filter control system architecture is shown, i.e. Figure 4 The signal output by the system will control the adaptive noise cancellation system 43 and the mono noise reduction system 44, helping to dynamically adjust the filter parameters and improve the filtering efficiency, and dynamically adapt to the changes in environmental noise. The system input is the main channel signal 601a / 601b (corresponding to Figure 5 509b) and reference channel signals 602a / 602b (corresponding to Figure 5 510b), and optional 603a / 603b (corresponding to Figure 5 511b); the system output is the control signal 607.

[0040] The main channel input signal 601a and the reference channel input signal 602a (and optionally 603a) are passed to the voice activity detection unit 61, which identifies when the desired audio is present by analyzing the signals of the main channel and the reference channel, and generates a voice activity detection signal 604, which is passed to the suppression logic unit 64. At the same time, the input signals 601b and 602b are passed to the main channel activity detector 62 and the reference channel activity detector 63, respectively. These detectors determine the activity state of the signal by detecting the power or energy level of the signal, and generate a main channel activity flag 605 and a reference channel activity flag 606, which are passed to the suppression logic unit 64.

[0041] The suppression logic unit 64 outputs a signal 607 to determine whether to suppress or enable the operation of the filter according to the input signal. When voice activity is detected and both the main channel and the reference channel are active, the suppression logic unit will allow the filter operation, thereby enabling the filter when the desired voice signal is present to improve the signal quality; if no voice activity is detected or the signal channel is inactive, the filter operation is suppressed to avoid unnecessary processing.

[0042] Figure 7 The adaptive noise cancellation system architecture is presented, i.e. Figure 4 The system integrates multi-channel sound signals into a single channel to achieve noise elimination in a complex acoustic environment. The system input is the main channel signal 701 (corresponding to Figure 5 509a in, reference channel signal 702 (corresponding to Figure 5 510a) and control signal 703 (corresponding to Figure 6 607); the system output is a single-channel sound signal 706.

[0043] In order to synchronously process multi-channel signals, the main channel input signal 701 passes through the delay element 71 to ensure that it is aligned in time with the reference channel signal, and the delayed signal 704 then enters the adder 72. The reference channel input signal 702 enters the adaptive filter 73, which is controlled by 703. When the system detects voice activity and both the main channel and the reference channel are active, the suppression logic allows the adaptive filter to be dynamically adjusted to ensure that the filter parameters are always in the best state to adapt to the real-time changing environmental noise; when no voice activity is detected, the suppression logic structure can output a signal to freeze the parameters of the adaptive filter to avoid unnecessary filtering operations in the absence of voice signals, reduce the consumption of system resources, and prevent the introduction of new noise by misoperation. The filtered signal 705 is output to the adder 610.

[0044] The adder subtracts 704 from 705 , and through this operation, the noise component in the main channel signal is effectively removed, and the clean desired audio signal 706 is retained.

[0045] Figure 8 The mono noise reduction system architecture is presented, i.e. Figure 4 The system processes the signal containing the desired audio and noise through a linear filter to reduce the unwanted audio components. The system input is a single-channel sound signal 801 (corresponding to Figure 7 706 in) and control signal 802 (corresponding to Figure 6 607); the system output is the processed sound signal 803.

[0046] The input signal 801 is passed to a linear filter 81 which processes the signal using a specific algorithm, typically a Wiener filter or a minimum mean square error filter (MMSE). The linear filter works based on the estimation of the statistical properties of the signal and the noise. In the absence of speech activity, the system estimates the noise component from the input signal, and this noise estimate is frozen when speech activity is detected and remains unchanged. In this way, the filter can use this frozen noise estimate to effectively suppress noise when a speech signal is present. Specifically, the filter applies a specific mathematical formula:

[0047]

[0048] The denominator represents the total power when there is desired audio activity, including the sum of the power of the desired audio and the noise, and the numerator represents the total power when there is desired audio activity minus the estimated value of the noise power when there is no desired audio activity, which actually reflects the power of the desired audio. When there is no desired audio, an estimate of the noise is obtained from the input signal 801, which is determined by the suppression logic (corresponding to Figure 6 The noise estimate is updated when the desired audio is not present on signal 801. When the desired audio is present, the noise estimate is frozen and the filtering process is performed using the noise estimate established during the last time the desired audio was not present.

[0049] Finally, the output signal 803 after the linear filter processing mainly contains the desired audio, and the noise component is significantly reduced. This output signal can be used for further processing or directly transmitted to the user to provide a high-quality audio experience.

[0050] When in use, multiple advanced technologies such as multi-microphone core technology, beamforming, filter control and adaptive noise cancellation are used to effectively suppress background noise and improve the clarity and quality of voice signals. Compared with the existing technology, this patent has significant advantages in dealing with complex noise environments and can provide better quality audio input in various noisy environments.

[0051] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for reducing noise using multiple microphone cores, characterized in that: The microphone used in the method has two or more microphone cores. The method includes obtaining a multi-channel sound signal through the multiple microphone cores of the microphone, and transmitting the sound signal to a beamforming system; the beamforming system can output the processed multi-channel sound signal and transmit it to a filtering control system and an adaptive noise cancellation system; the filtering control system can output the processed multi-channel sound signal and transmit it to an adaptive noise cancellation system and a single-channel noise reduction system; the adaptive noise cancellation system can convert the multi-channel sound signal into a single-channel sound signal and transmit it to the single-channel noise reduction system.

2. The method for noise reduction using multiple microphone cores according to claim 1, characterized in that: A multi-channel sound signal is received from a microphone, wherein the sound signal of one channel is passed to a DC blocking filter, and after processing, is passed to a de-emphasis filter, and finally a main channel sound signal is output and transmitted to the other systems; the sound signals of other channels are passed to a DC blocking filter, and after processing, are passed to a gain and phase adjustment filter, and after processing again, are passed to a de-emphasis filter, and finally a reference channel sound signal is output and transmitted to the other systems.

3. The method for noise reduction using multiple microphone cores according to claim 1, characterized in that: Receive multi-channel sound signals, transmit them to the voice activity detection unit, and transmit them to the suppression logic unit after processing; the main channel sound signal also needs to be transmitted to the main channel detector, and transmitted to the suppression logic unit after processing; the reference channel sound signal also needs to be transmitted to the reference channel detector, and transmitted to the suppression logic unit after processing; the suppression logic unit outputs a control signal and transmits it to the other systems.

4. The method for noise reduction using multiple microphone cores according to claim 1, characterized in that: The system receives multi-channel sound signals and control signals, wherein the main channel sound signal is transmitted to the delay unit and then transmitted to the adder after processing; the reference channel sound signal and control signal are transmitted to the adaptive filter and then transmitted to the adder after processing; the adder outputs a single channel sound signal and transmits it to the other systems.

5. The method for noise reduction using multiple microphone cores according to claim 1, characterized in that: It receives a single-channel audio signal and control signal, passes them to a linear filter, and outputs an audio signal that achieves a significant noise reduction effect.

6. The method for noise reduction using multiple microphone cores according to claim 1, characterized in that: The microphone structure includes an outer shell and a circuit board assembly fixed in the outer shell; the outer shell is made of high-strength and lightweight material and has good durability and protection performance; the circuit board assembly includes multiple circuits and components, which can support the noise reduction function and other related functions of the microphone.

7. The method for noise reduction using multiple microphone cores according to claim 6, characterized in that: The outer shell has sound receiving holes, each of which has a diameter of 2 mm or less and is designed to effectively collect sound signals; the layout of the sound receiving holes has been precisely calculated and optimized to ensure the collection effect of sound signals; the outside is covered with a mesh made of highly sound-transparent material, which can prevent dust and impurities from entering without affecting the transmission quality of sound.

8. The method for reducing noise using multiple microphone cores according to claim 6, characterized in that: The circuit board assembly includes two or more microphone core arrays, one of which is defined as the microphone core closest to the speech source and is used to output the main channel sound signal; the other microphone cores are used to receive ambient noise and reference signals to assist in noise reduction processing; the circuit board assembly also includes an adaptive filter, a control unit and a signal processing unit, etc., which can complete the noise reduction method.

Images