An omnidirectional microphone automatic directional sound pickup noise reduction method and device

Abstract

This application relates to an automatic directional sound pickup and noise reduction method and apparatus for an omnidirectional microphone. The automatic directional sound pickup and noise reduction method for an omnidirectional microphone includes the following steps: a microphone array composed of multi-diaphragm microphones performs multi-directional sound source pickup, locates the direction of the target sound source, constructs a beam focusing on the target sound source, and obtains several first sound source electrical signals; selects one of the first sound source electrical signals according to a user preset and marks it as a standard sound source; performs sound source unification processing on the other first sound source electrical signals according to the standard sound source; and performs noise reduction processing after mixing all sound sources with unified sound characteristics to form a stereo sound source. The automatic directional sound pickup and noise reduction method for an omnidirectional microphone of this application can quantitatively enhance the speech and noise reduction intensity, ensuring a uniform and clear speech output regardless of the speaker's location, and preventing over-noise reduction or no noise reduction in the subsequent noise reduction process.

Description

Technical Field

[0001] This application relates to the technical field of speech noise reduction, and more specifically to an automatic directional sound pickup and noise reduction method and apparatus for an omnidirectional microphone. Background Technology

[0002] Conference microphones specifically designed for multi-person meeting scenarios rely on multi-microphone arrays and speech processing algorithms for directional sound pickup and noise reduction to achieve clear interactive speech. Existing microphone arrays can pick up sound directionally, but the volume of the sound picked up by different microphones varies due to their relative position to the sound source, resulting in different levels of noise reduction. This is especially true for near-field and far-field sounds, where the output sound differs significantly, making it impossible to guarantee consistently clear speech and causing the sound to appear to come from different distances.

[0003] In addition, the sound noise reduction technology used in the existing technology is usually spectral subtraction. The mechanism of spectral subtraction is based on threshold judgment and energy reduction based on the noise baseline. The principle is to first estimate the spectral distribution of noise in the silent segment by detecting speech activity to form a "noise baseline", and then analyze the spectral components of each frame of signal. If its amplitude is lower than the baseline, it is regarded as a noise-dominant region and is zeroed or attenuated.

[0004] This noise reduction method is less than ideal and can easily lead to speech distortion (such as amplitude clipping causing consonant loss, phase distortion affecting spatial positioning), speech interruption (due to excessive suppression by median filtering or noise threshold), abnormal listening experience (such as "telephone voice," "metallic sound," "tunneling effect"), and decreased naturalness (such as excessive dynamic range compression causing speech to lose its layering, or even accidentally deleting weak sound components such as breathy sounds and consonants). For example, in a multi-person conference, when a person speaks at a distance, the sound picked up by the far-field microphone will have a spectrum similar to the surrounding noise. After using spectral subtraction for noise reduction, there may be speech loss (over-noise reduction) or no noise at a distance (no noise reduction), which can lead to sound masking phenomena, such as loud background noise masking soft speech, or near sound masking far sound. Summary of the Invention

[0005] The technical problem to be solved by this application is to provide an automatic directional sound pickup and noise reduction method for an omnidirectional microphone, which solves the problem of inconsistent output sound after directional sound pickup by microphone arrays in the prior art. After the microphone detects the speaker's location, it can quantitatively enhance the speech and noise reduction intensity, so that no matter where the person speaks from the microphone, there can be a uniform and clear speech output, and there will be no over-noise reduction or no noise reduction during noise reduction.

[0006] The technical solution of this application is to provide an automatic directional sound pickup and noise reduction method for an omnidirectional microphone, including the following steps: Step 1: Set up a microphone array consisting of multi-diaphragm microphones; Step 2: The microphone array performs multi-directional sound source pickup, locates the direction of the target sound source, constructs a beam focused on the target sound source, and each multi-diaphragm microphone picks up the target sound source to obtain several first sound source electrical signals. Step 3: Set the first preset parameter, which represents the user-preset sound characteristics, including sound intensity value and near-far field sensory level value; compare each first sound source electrical signal with the first preset parameter, select the first sound source electrical signal that is closest to the first preset parameter and mark it as the standard sound source, and the corresponding microphone is the standard microphone group; Step 4: Perform sound source unification processing on each first sound source electrical signal. Specifically, compare each other first sound source electrical signal with the standard sound source. By switching the diaphragms of each microphone with different sensitivities, ensure that each first sound source electrical signal conforms to the characteristics of the standard sound source, so that each microphone can output a sound source with uniform sound characteristics. Step 5: Mix all sound sources with uniform sound characteristics into a stereo sound source, and perform noise reduction processing on the stereo sound source.

[0007] Compared with existing technologies, the automatic directional sound pickup and noise reduction method for omnidirectional microphones in this application has the following advantages: The system selects a standard sound source based on the user's preset sound characteristics to meet the user's preferences. By standardizing the sound sources of each microphone according to the standard sound source, the voice and noise reduction intensity can be quantitatively enhanced, so that no matter where a person speaks from the microphone, there can be a consistent and clear voice output. In this way, the problem of inconsistent output sound after directional sound pickup by microphone arrays in the existing technology can be solved, and there will be no over-noise reduction or no noise reduction in the subsequent noise reduction process. The microphone array consists of multi-diaphragm microphones. By switching the diaphragms of each microphone with different sensitivities, the sound sources picked up by each microphone can be uniformly processed, so that the electrical signals of each first sound source are adjusted to conform to the characteristics of the standard sound source.

[0008] Preferably, step 4 includes the following steps: Step 4-1: Set a second preset parameter, which represents the unit threshold range of the difference between the first sound source electrical signal and the standard sound source, including the threshold range of sound intensity and the threshold range of sensory level of near and far fields; compare each of the other first sound source electrical signals with the standard sound source; if the difference between the first sound source electrical signal and the standard sound source is within the unit threshold range of the second preset parameter, then the first sound source electrical signal is marked as a first-class standard sound source; if the difference between the first sound source electrical signal and the standard sound source is greater than the maximum value of the unit threshold range of the second preset parameter, then the first sound source electrical signal is marked as a near-field sound source; if the difference between the first sound source electrical signal and the standard sound source is less than the minimum value of the unit threshold range of the second preset parameter, then the first sound source electrical signal is marked as a far-field sound source; Step 4-2: Switch the diaphragm of the microphone corresponding to the near-field sound source to a low-sensitivity diaphragm, convert the near-field sound source into a second sound source electrical signal, and repeat step 4 until the difference between the second sound source electrical signal and the standard sound source is within the unit threshold range of the second preset parameter, then the second sound source electrical signal is marked as the second type of standard sound source. Step 4-3: Switch the diaphragm of the microphone corresponding to the far-field sound source to a high-sensitivity diaphragm, convert the far-field sound source into a third sound source electrical signal, and repeat step 4 until the difference between the third sound source electrical signal and the standard sound source is within the unit threshold range of the second preset parameter, then the third sound source electrical signal is marked as a third type of standard sound source. Step 5 involves mixing the first type of standard sound source, the second type of standard sound source, and the third type of standard sound source to create a stereo sound source.

[0009] Using this method, the electrical signals of each first sound source can be classified according to the second preset parameters, and near-field sound sources and far-field sound sources can be screened out. Then, a low-sensitivity diaphragm is switched for the near-field sound source to reduce the pickup intensity of the near-field sound source, and a high-sensitivity diaphragm is switched for the far-field sound source to increase the pickup intensity of the far-field sound source. In this way, the electrical signals of each first sound source can be adjusted to conform to the characteristics of the standard sound source.

[0010] Preferably, step 5 involves noise reduction processing of the stereo sound source, including the following steps: Step 5-1: Detect noise points in the stereo sound source, cut out noise segments, and separate normal speech segments and noise segments from the stereo sound source. Step 5-2: Set the third preset parameter, which represents the noise reduction level preset by the user; amplify the signal of the far-field sound source and mark it as a backup sound source, and the amplification ratio is positively correlated with the third preset parameter; Step 5-3: Replace the noise segment with the backup sound source; Step 5-4: Combine the normal speech segment and the backup audio source into an audio file.

[0011] This method amplifies the signal of the far-field sound source according to the user's preset noise reduction level, which helps to amplify the speech and reduce noise interference. First, the normal speech segment and the noise segment are separated from the stereo sound source, and then the noise segment is replaced with the amplified signal of the far-field sound source. In this way, the speech can be restored without distortion, the speech has a sense of layering, and it sounds very natural. There will be no situation of over-noise reduction or no noise reduction, and there will be no sound masking phenomenon.

[0012] Preferably, in step 3, the first preset parameter is set based on the pickup characteristics of the group of microphones closest to the target sound source. The group of microphones closest to the target sound source is a standard microphone group, and the first sound source electrical signal obtained by the standard microphone group is marked as the standard sound source. Using this method, the first sound source electrical signal obtained by the group of microphones closest to the target sound source contains the strongest speech segment information, which helps to restore the speech to the greatest extent.

[0013] Preferably, in step 3, a fourth preset parameter is also set. The fourth preset parameter represents the unit threshold range of the difference between the first sound source electrical signal and the first preset parameter, including the threshold range of sound intensity and the threshold range of sensory level of near and far fields. After selecting the first sound source electrical signal that is closest to the first preset parameter, the selected first sound source electrical signal is compared with the first preset parameter. If the difference between the first sound source electrical signal and the first preset parameter is within the unit threshold range of the fourth preset parameter, the selected first sound source electrical signal is directly marked as a standard sound source. If the difference between the first sound source electrical signal and the first preset parameter is greater than the maximum value of the unit threshold range of the fourth preset parameter, or less than the minimum value of the unit threshold range of the fourth preset parameter, the diaphragm of the corresponding microphone is switched so that the difference between the first sound source electrical signal and the first preset parameter is within the unit threshold range of the second preset parameter, and then it is marked as a standard sound source.

[0014] Using this method, the standard sound source can be made closer to the sound characteristics represented by the first preset parameter, thus meeting the user's preferences.

[0015] Preferably, in step 1, the microphone array is a spherical array used for 360° omnidirectional sound source pickup. This method enables 360° omnidirectional sound source pickup, facilitating accurate positioning of the target sound source direction and the construction of a beam focused on the target sound source.

[0016] This application provides an automatic directional sound pickup and noise reduction device for an omnidirectional microphone. The technical solution includes a microphone array composed of multi-diaphragm microphones, a preamplifier, an ADC analog-to-digital converter, a DSP audio processor, and a feedback unit. It can execute the automatic directional sound pickup and noise reduction method for an omnidirectional microphone as described above. The microphone array is used to spatially sample sound waves and generate an analog sound signal. The preamplifier is used to amplify the analog sound signal. The ADC analog-to-digital converter is used to convert the amplified analog sound signal into a digital sound signal. The DSP audio processor is used to locate the direction of the target sound source based on the digital sound signal, construct a beam focusing on the target sound source, and reduce noise in the pickup signal. The feedback unit is used to feed back the output signal of the DSP audio processor to the microphone array and the preamplifier.

[0017] Compared with the prior art, the automatic directional sound pickup and noise reduction device for omnidirectional microphones of this application has the following advantages: The automatic directional sound pickup and noise reduction device for omnidirectional microphones of this application can perform the above-mentioned automatic directional sound pickup and noise reduction method for omnidirectional microphones, solving the problem of inconsistent output sound after directional sound pickup in the prior art, so that no matter where a person speaks from the microphone, there will be a uniform and clear voice output, and there will be no over-noise reduction or no noise reduction during noise reduction.

[0018] Preferably, the microphone array is formed by arranging multi-diaphragm silicon microphones in a three-dimensional spatial configuration according to a spherical geometry, for omnidirectional sound pickup. This structure facilitates the precise location of the target sound source and the construction of a beam focused on the target sound source.

[0019] Preferably, it includes a noise-reduction roller encoder for setting the DSP audio processor to dynamically adjust the noise reduction level. This structure facilitates dynamic adjustment of the noise reduction level. Attached Figure Description

[0020] Figure 1 This is a flowchart illustrating the automatic directional sound pickup and noise reduction method for the omnidirectional microphone of this application.

[0021] Figure 2 This is a structural block diagram of the automatic directional sound pickup and noise reduction device for the omnidirectional microphone of this application. Detailed Implementation

[0022] To better understand this application, various aspects of this application will be described in more detail with reference to the accompanying drawings. It should be understood that these detailed descriptions are merely illustrative of exemplary embodiments of this application and are not intended to limit the scope of this application in any way. Throughout the specification, the same reference numerals refer to the same elements.

[0023] In the accompanying drawings, the thickness, size, and shape of the objects have been slightly exaggerated for illustrative purposes. The drawings are for illustrative purposes only and are not drawn to scale.

[0024] It should also be understood that the terms "comprising," "having," "including," and "containing," when used in this specification, indicate the presence of the stated features, integrals, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components, and / or combinations thereof. Furthermore, when expressions such as "...at least one" appear after a list of listed features, they modify the entire listed feature, not individual elements in the list.

[0025] like Figure 2 As shown, the automatic directional sound pickup and noise reduction device for an omnidirectional microphone of this application includes a microphone array composed of multi-diaphragm microphones, a preamplifier, an ADC analog-to-digital converter, a DSP audio processor, a feedback unit, and a noise reduction roller encoder. The microphone array is formed by arranging multi-diaphragm silicon microphones in a three-dimensional spatial arrangement according to a spherical geometry, used for 360° omnidirectional sound pickup, spatially sampling sound waves and generating analog sound signals. The preamplifier amplifies the received analog sound signals. The ADC analog-to-digital converter converts the amplified analog sound signals into digital sound signals. The DSP audio processor locates the direction of the target sound source based on the digital sound signals, constructs a beam focusing on the target sound source, and reduces noise in the pickup signal. The feedback unit feeds back the output signal of the DSP audio processor to the microphone array and the preamplifier. The noise reduction roller encoder sets the DSP audio processor to adjust the noise reduction level. All of the above components are existing technologies, but after circuit integration, they can perform the automatic directional sound pickup and noise reduction method for an omnidirectional microphone as described below, solving the problem of inconsistent output sound after directional sound pickup by microphone arrays in the prior art.

[0026] like Figure 1 As shown, the automatic directional sound pickup and noise reduction method for an omnidirectional microphone of this application includes the following steps: Step 1: Set up a microphone array consisting of multi-diaphragm microphones; Step 2: The microphone array performs multi-directional sound source pickup, locates the direction of the target sound source, constructs a beam focused on the target sound source, and each multi-diaphragm microphone picks up the target sound source to obtain several first sound source electrical signals. Step 3: Set the first preset parameter, which represents the user-preset sound characteristics, including sound intensity value and near-far field sensory level value; compare each first sound source electrical signal with the first preset parameter, select the first sound source electrical signal that is closest to the first preset parameter and mark it as the standard sound source, and the corresponding microphone is the standard microphone group; Step 4: Perform sound source unification processing on each first sound source electrical signal. Specifically, compare each other first sound source electrical signal with the standard sound source. By switching the diaphragms of each microphone with different sensitivities, ensure that each first sound source electrical signal conforms to the characteristics of the standard sound source, so that each microphone can output a sound source with uniform sound characteristics. Step 5: Mix all sound sources with uniform sound characteristics into a stereo sound source, and perform noise reduction processing on the stereo sound source.

[0027] The automatic directional sound pickup and noise reduction method for omnidirectional microphones in this application first selects a standard sound source based on user-preset sound characteristics, and then performs unified processing on the sound sources of each microphone based on the standard sound source. This can quantitatively enhance the speech and noise reduction intensity, so that no matter where a person speaks from the microphone, there can be a uniform and clear speech output. This solves the problem of inconsistent output sound after directional sound pickup by microphone arrays in the prior art, and there will be no over-noise reduction or no noise reduction in the subsequent noise reduction process.

[0028] Step 4 involves unifying the audio signals from each first audio source, specifically including the following steps: Step 4-1: Set a second preset parameter, which represents the unit threshold range of the difference between the first sound source electrical signal and the standard sound source, including the threshold range of sound intensity and the threshold range of sensory level of near and far fields; compare each of the other first sound source electrical signals with the standard sound source; if the difference between the first sound source electrical signal and the standard sound source is within the unit threshold range of the second preset parameter, then the first sound source electrical signal is marked as a first-class standard sound source; if the difference between the first sound source electrical signal and the standard sound source is greater than the maximum value of the unit threshold range of the second preset parameter, then the first sound source electrical signal is marked as a near-field sound source; if the difference between the first sound source electrical signal and the standard sound source is less than the minimum value of the unit threshold range of the second preset parameter, then the first sound source electrical signal is marked as a far-field sound source; Step 4-2: Switch the diaphragm of the microphone corresponding to the near-field sound source to a low-sensitivity diaphragm, convert the near-field sound source into a second sound source electrical signal, and repeat step 4 until the difference between the second sound source electrical signal and the standard sound source is within the unit threshold range of the second preset parameter, then the second sound source electrical signal is marked as the second type of standard sound source. Step 4-3: Switch the diaphragm of the microphone corresponding to the far-field sound source to a high-sensitivity diaphragm, convert the far-field sound source into a third sound source electrical signal, and repeat step 4 until the difference between the third sound source electrical signal and the standard sound source is within the unit threshold range of the second preset parameter. Then the third sound source electrical signal is marked as a third type of standard sound source.

[0029] According to the second preset parameters, the electrical signals of each first sound source are classified and near-field sound sources and far-field sound sources are selected. The microphone array is composed of multi-diaphragm microphones. By switching the diaphragms of each microphone with different sensitivities, the sound sources picked up by each microphone can be uniformly processed. For example, switching to a low-sensitivity diaphragm for near-field sound sources reduces the sound pickup intensity of near-field sound sources, while switching to a high-sensitivity diaphragm for far-field sound sources increases the sound pickup intensity of far-field sound sources. In this way, the electrical signals of each first sound source can be adjusted to conform to the characteristics of a standard sound source.

[0030] Step 5 involves mixing the first, second, and third type of standard sound sources to create a stereo sound source, followed by noise reduction. The noise reduction process for the stereo sound source specifically includes the following steps: Step 5-1: Detect noise points in the stereo sound source, cut out noise segments, and separate normal speech segments and noise segments from the stereo sound source. Step 5-2: Set the third preset parameter, which represents the noise reduction level preset by the user; amplify the signal of the far-field sound source and mark it as a backup sound source, and the amplification ratio is positively correlated with the third preset parameter; Step 5-3: Replace the noise segment with the backup sound source; Step 5-4: Combine the normal speech segment and the backup audio source into an audio file.

[0031] Step 5 involves mixing the first type of standard sound source, the second type of standard sound source, and the third type of standard sound source to create a stereo sound source.

[0032] Amplifying the signal from the far-field sound source according to the user's preset noise reduction level helps to amplify speech and reduce noise interference. First, the normal speech segment and the noise segment are separated from the stereo sound source, and then the noise segment is replaced with the amplified signal from the far-field sound source. In this way, the speech can be restored without distortion, the speech has a sense of layering, and it sounds very natural. There will be no situation of over-noise reduction or no noise reduction, and there will be no sound masking phenomenon.

[0033] In other embodiments, the standard sound source can be made closer to the sound characteristics represented by the first preset parameter to meet the user's preferences, specifically including the following steps: A fourth preset parameter is set to characterize the unit threshold range of the difference between the first sound source electrical signal and the first preset parameter, including the threshold range of sound intensity and the threshold range of sensory level of near and far fields. In step 3, after selecting the first sound source electrical signal that is closest to the first preset parameter, the selected first sound source electrical signal is compared with the first preset parameter. If the difference between the first sound source electrical signal and the first preset parameter is within the unit threshold range of the fourth preset parameter, the selected first sound source electrical signal is directly marked as a standard sound source. If the difference between the first sound source electrical signal and the first preset parameter is greater than the maximum value of the unit threshold range of the fourth preset parameter, or less than the minimum value of the unit threshold range of the fourth preset parameter, the diaphragm of the corresponding microphone is switched so that the difference between the first sound source electrical signal and the first preset parameter is within the unit threshold range of the second preset parameter, and then it is marked as a standard sound source.

[0034] In other embodiments, the focus can be solely on maximizing the preservation of speech information and simplifying the process of marking standard sound sources. For example, a first preset parameter can be set based on the pickup characteristics of the group of microphones closest to the target sound source. After the microphone array locates the direction of the target sound source, the group of microphones closest to the target sound source is marked as the standard microphone group. The first sound source electrical signal obtained by the standard microphone group is then marked as the standard sound source. The first sound source electrical signal obtained by the group of microphones closest to the target sound source contains the strongest speech segment information, which helps to restore speech to the greatest extent possible.

[0035] The above are merely specific embodiments of this application and are not intended to limit the scope of this application. Any modifications or equivalent substitutions made to this application without departing from the spirit and scope thereof shall be covered within the protection scope of the claims of this application.

Claims

1. An automatic directional sound pickup and noise reduction method for an omnidirectional microphone, characterized in that, Includes the following steps: Step 1: Set up a microphone array consisting of multi-diaphragm microphones; Step 2: The microphone array performs multi-directional sound source pickup, locates the direction of the target sound source, constructs a beam focused on the target sound source, and each multi-diaphragm microphone picks up the target sound source to obtain several first sound source electrical signals. Step 3: Set the first preset parameter, which represents the user-preset sound characteristics, including sound intensity value and near-far field sensory level value; Each first sound source electrical signal is compared with the first preset parameter, and the first sound source electrical signal that is closest to the first preset parameter is selected and marked as the standard sound source, and the corresponding microphone is the standard microphone group; Step 4: Perform sound source unification processing on each first sound source electrical signal. Specifically, compare each other first sound source electrical signal with the standard sound source. By switching the diaphragms of each microphone with different sensitivities, ensure that each first sound source electrical signal conforms to the characteristics of the standard sound source, so that each microphone can output a sound source with uniform sound characteristics. Step 5: Mix all sound sources with uniform sound characteristics into a stereo sound source, and perform noise reduction processing on the stereo sound source.

2. The automatic directional sound pickup and noise reduction method for an omnidirectional microphone according to claim 1, characterized in that, Step 4 includes the following steps: Step 4-1: Set a second preset parameter, which represents the unit threshold range of the difference between the first sound source electrical signal and the standard sound source, including the threshold range of sound intensity and the threshold range of sensory level of near and far fields; compare each of the other first sound source electrical signals with the standard sound source; if the difference between the first sound source electrical signal and the standard sound source is within the unit threshold range of the second preset parameter, then the first sound source electrical signal is marked as a first-class standard sound source; if the difference between the first sound source electrical signal and the standard sound source is greater than the maximum value of the unit threshold range of the second preset parameter, then the first sound source electrical signal is marked as a near-field sound source; if the difference between the first sound source electrical signal and the standard sound source is less than the minimum value of the unit threshold range of the second preset parameter, then the first sound source electrical signal is marked as a far-field sound source; Step 4-2: Switch the diaphragm of the microphone corresponding to the near-field sound source to a low-sensitivity diaphragm, convert the near-field sound source into a second sound source electrical signal, and repeat step 4 until the difference between the second sound source electrical signal and the standard sound source is within the unit threshold range of the second preset parameter, then the second sound source electrical signal is marked as the second type of standard sound source. Step 4-3: Switch the diaphragm of the microphone corresponding to the far-field sound source to a high-sensitivity diaphragm, convert the far-field sound source into a third sound source electrical signal, and repeat step 4 until the difference between the third sound source electrical signal and the standard sound source is within the unit threshold range of the second preset parameter, then the third sound source electrical signal is marked as a third type of standard sound source. Step 5 involves mixing the first type of standard sound source, the second type of standard sound source, and the third type of standard sound source to create a stereo sound source.

3. The automatic directional sound pickup and noise reduction method for an omnidirectional microphone according to claim 1 or 2, characterized in that, Step 5 involves noise reduction processing of the stereo sound source, including the following steps: Step 5-1: Detect noise points in the stereo sound source, cut out noise segments, and separate normal speech segments and noise segments from the stereo sound source. Step 5-2: Set the third preset parameter, which represents the noise reduction level preset by the user; amplify the signal of the far-field sound source and mark it as a backup sound source, and the amplification ratio is positively correlated with the third preset parameter; Step 5-3: Replace the noise segment with the backup sound source; Step 5-4: Combine the normal speech segment and the backup audio source into an audio file.

4. The automatic directional sound pickup and noise reduction method for an omnidirectional microphone according to claim 1, characterized in that, In step 3, the first preset parameter is set based on the sound pickup characteristics of the group of microphones closest to the target sound source. The group of microphones closest to the target sound source is the standard microphone group, and the first sound source electrical signal obtained by the standard microphone group is marked as the standard sound source.

5. The automatic directional sound pickup and noise reduction method for an omnidirectional microphone according to claim 1 or 4, characterized in that, In step 3, a fourth preset parameter is also set. The fourth preset parameter represents the unit threshold range of the difference between the first sound source electrical signal and the first preset parameter, including the threshold range of sound intensity and the threshold range of sensory level of near and far fields. After selecting the first sound source electrical signal that is closest to the first preset parameter, the selected first sound source electrical signal is compared with the first preset parameter. If the difference between the first sound source electrical signal and the first preset parameter is within the unit threshold range of the fourth preset parameter, the selected first sound source electrical signal is directly marked as the standard sound source. If the difference between the first sound source electrical signal and the first preset parameter is greater than the maximum value of the unit threshold interval of the fourth preset parameter, or less than the minimum value of the unit threshold interval of the fourth preset parameter, then the corresponding microphone will switch its diaphragm so that the difference between the first sound source electrical signal and the first preset parameter is within the unit threshold interval of the second preset parameter, and then it will be marked as a standard sound source.

6. The automatic directional sound pickup and noise reduction method for an omnidirectional microphone according to claim 1, characterized in that, In step 1, the microphone array is a spherical array used for 360° omnidirectional sound source pickup.

7. An automatic directional sound pickup and noise reduction device for an omnidirectional microphone, characterized in that, The device includes a microphone array composed of multi-diaphragm microphones, a preamplifier, an ADC analog-to-digital converter, a DSP audio processor, and a feedback unit, and is capable of executing the automatic directional sound pickup and noise reduction method for an omnidirectional microphone as described in any one of claims 1 to 6. The microphone array is used to spatially sample sound waves and generate an analog sound signal. The preamplifier is used to amplify the analog sound signal. The ADC analog-to-digital converter is used to convert the amplified analog sound signal into a digital sound signal. The DSP audio processor is used to locate the direction of the target sound source based on the digital sound signal, construct a beam focusing on the target sound source, and reduce noise in the pickup signal. The feedback unit is used to feed back the output signal of the DSP audio processor to the microphone array and the preamplifier.

8. The automatic directional sound pickup and noise reduction device for an omnidirectional microphone according to claim 5, characterized in that, The microphone array is formed by arranging multi-diaphragm silicon microphones in three-dimensional space according to a spherical geometric structure for omnidirectional sound pickup.

9. The automatic directional sound pickup and noise reduction device for an omnidirectional microphone according to claim 5, characterized in that, It includes a noise-reduction roller encoder for setting the DSP audio processor to dynamically adjust the noise reduction level.

Images