Method for optimizing microphone array compensation and audio acquisition device

Abstract

The application relates to a microphone array compensation optimization method and an audio acquisition device, the optimization compensation method comprising the following steps: taking the overall signal strength value of the audio signal collected by the microphone array at each correction angle selected in advance as a measurement value; obtaining the inclination angle of the microphone array after installation; calculating the inclination correction value corresponding to the inclination angle by using a linear interpolation method, and performing gain compensation on the audio signal actually collected by the microphone array according to the inclination correction value. In the application, the measurement value is obtained by preselecting multiple correction angles, the inclination correction value corresponding to the inclination angle can be calculated by using the linear interpolation method according to the measurement value, the overall signal strength value of the microphone array is compensated according to the inclination correction value, and the influence of the inclination installation environment on the microphone array is greatly reduced.

Description

Technical Field

[0001] This invention belongs to the field of microphone array technology, and relates to an optimization method for microphone array compensation and an audio acquisition device. Background Technology

[0002] A microphone array is an array of multiple microphones arranged in a specific geometric pattern. It captures, processes, and recognizes audio signals by processing the signals from the microphone array. Compared to a single microphone, microphone array technology offers higher accuracy and efficiency in noise reduction, voice enhancement, direction recognition, and distance measurement. It is widely used in speech recognition, speech translation, smart homes, conferencing systems, and wireless communication.

[0003] When installing microphone arrays, for aesthetic reasons, the base is usually a square, the same size as the ceiling panel, allowing for ceiling mounting. Sometimes, it's also suspended from the ceiling using a boom. Regardless of the mounting method, the microphone array will generally have a slight tilt angle after installation, making it difficult to ensure a perfectly horizontal installation. Furthermore, boom mounting sometimes inherently results in an angled installation. Current microphone array calibration methods typically use a fixed horizontal mounting configuration, requiring extremely precise installation to ensure the array remains level and avoid interference from the installation angle; this also prevents the tilt angle from negatively impacting the microphone array's directional recognition. Therefore, it's necessary to improve the algorithm to reduce the impact of tilted microphone array installation. Summary of the Invention

[0004] To address the shortcomings of the prior art, the technical problem to be solved by the present invention is to provide an optimization method for microphone array compensation and an audio acquisition device.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] An optimization method for microphone array compensation includes the following steps:

[0007] S110. Select multiple sound source emission angles in advance as correction angles, and test the overall signal strength value of the audio signal collected by the microphone array at each correction angle as the measurement value.

[0008] S120. Obtain the tilt angle of the microphone array after installation;

[0009] S130. Based on the measured values ​​obtained from the pre-test, calculate the tilt correction value corresponding to the tilt angle using a linear interpolation method; and perform gain compensation on the overall signal strength value of the microphone array based on the tilt correction value.

[0010] Furthermore, in step S110, the method for selecting multiple sound source emission angles as correction angles is as follows:

[0011] The range of sound source emission angles in which the overall signal strength of the microphone array increases with the increase of the sound source emission angle is divided into a linearly increasing range, and the range of sound source emission angles in which the overall signal strength of the microphone array decreases with the increase of the sound source emission angle is divided into a linearly decreasing range; multiple angles are selected at equal intervals in the linearly increasing and linearly decreasing ranges as correction angles.

[0012] Furthermore, the sound source emission angle range of the linearly increasing interval is 0° to 180°, and the sound source emission angle range of the linearly decreasing interval is 180° to 360°; the correction angle includes multiple angles selected at equal intervals between 0°, 180°, 360°, and between 0° and 180°, as well as multiple angles selected at equal intervals between 180° and 360°.

[0013] Furthermore, in step S110, after obtaining the measured values ​​through testing, the overall signal strength value of the microphone array is obtained as a function curve of the sound source emission angle through data fitting, and each measured value is corrected according to the function curve.

[0014] Furthermore, the microphone array is equipped with an angle sensor to obtain the tilt angle of the microphone array after installation.

[0015] Furthermore, step S130 includes the following sub-steps:

[0016] S131. Select a correction angle smaller than the tilt angle as the first interpolation angle, and select a correction angle larger than the tilt angle as the second interpolation angle.

[0017] S132. The theoretical overall signal strength value when the tilt angle is θ is calculated using the following formula:

[0018]

[0019] Where θ represents the tilt angle of the microphone array after installation; signal(θ) represents the theoretical signal strength value when the tilt angle is θ; a represents the first interpolation angle; b represents the second interpolation angle; val(a) represents the measured value corresponding to the first interpolation angle; val(b) represents the measured value corresponding to the second interpolation angle.

[0020] S133. Calculate the tilt correction value corresponding to a tilt angle of θ using the following formula:

[0021] signal(correction)=signal(0)-signal(θ)

[0022] Where signal(correction) represents the tilt correction value; signal(0) represents the measurement value when the correction angle is 0°;

[0023] S134. Perform gain compensation on the overall signal strength of the microphone array based on the tilt correction value; the gain compensation formula is as follows:

[0024] signal(compensation)=signal(measured)+signal(correction)

[0025] Here, signal(compensation) represents the overall signal strength value of the microphone array after gain compensation, and signal(measured) represents the overall signal strength value of the microphone array before gain compensation.

[0026] Furthermore, in step S131, a correction angle that is smaller than the tilt angle and closest to the tilt angle is selected as the first interpolation angle, and a correction angle that is larger than the tilt angle and closest to the tilt angle is selected as the second interpolation angle.

[0027] An audio acquisition device, comprising

[0028] A microphone array, wherein the microphone array is an array of multiple microphones arranged geometrically on a base;

[0029] The audio processing unit is used for sound source localization and speech enhancement processing of the signals acquired by the microphone array; and

[0030] The compensation unit is used to compensate the gain of the microphone array using an optimized method for microphone array compensation.

[0031] Furthermore, the audio processing unit includes

[0032] The preprocessing module is used to preprocess the signals acquired by the microphone array;

[0033] The sound source localization module is used to identify the direction and distance of the sound source based on the signals collected by the microphone array;

[0034] The speech enhancement module is used to form a beam in a desired direction using beamforming technology, thereby extracting the sound source and suppressing noise; and to process the acquired speech signal to improve the quality and reliability of the speech; and

[0035] The output module is used to output the signal processed by the speech enhancement module.

[0036] Furthermore, the compensation unit includes

[0037] The compensation value storage module is used to store the measurement values ​​obtained after measuring the overall signal strength of the audio signals acquired by the microphone array at various correction angles;

[0038] The recognition module is used to identify the tilt angle of the microphone array after installation; and

[0039] The adaptive algorithm module is used to calculate the tilt correction value corresponding to the tilt angle using a linear interpolation method, and to perform gain compensation on the overall signal strength value of the microphone array based on the tilt correction value.

[0040] In this invention, multiple sound source emission angles are selected at pre-equal intervals as correction angles, and measured values ​​are obtained. These measured values ​​can be used to fit a function curve showing the overall signal strength of the microphone array as a function of the sound source emission angle. The measured values ​​can then be corrected according to this function curve, thereby eliminating measurement errors. A linear interpolation method is used to calculate the tilt correction value corresponding to the tilt angle, and gain compensation is applied to the overall signal strength of the microphone array based on this tilt correction value. This significantly reduces the impact of tilted installation on the microphone array. Only a few pre-measurements at correction angles are required to apply compensation for all tilt angles; the calculation method is simple and has good versatility. Attached Figure Description

[0041] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0042] Figure 1 The flowchart of an optimization method for microphone array compensation according to one embodiment of the present invention is shown below.

[0043] Figure 2 A schematic diagram of a microphone array mounted at an angle.

[0044] Figure 3 This is a flowchart illustrating gain compensation for a microphone array using a linear interpolation method.

[0045] Figure 4 This is a schematic diagram of a microphone array structure.

[0046] Figure 5 This is a schematic diagram of the microphone assembly.

[0047] Figure 6 This is a schematic diagram showing the positional relationship between the four sides of the base, the four microphone components, and the obstruction when one side is blocked.

[0048] Figure 7 This is a schematic diagram showing the positional relationship between the four sides of the base, the four microphone components, and the obstruction when two adjacent sides are blocked.

[0049] Figure 8 This is a schematic diagram showing the positional relationship between the four sides of the base, the four microphone components, and the obstruction when there is relative obstruction on two sides.

[0050] Figure 9 This is a schematic diagram showing the positional relationship between the four sides of the base, the four microphone components, and the obstruction when three sides are blocked.

[0051] Figure 10 This is a schematic diagram showing the distribution of each microphone component in the microphone array.

[0052] Figure 11 The present invention provides a structural block diagram of the audio acquisition device in one embodiment.

[0053] The meanings of the labels in the attached diagram are as follows:

[0054] Microphone array - 100; Microphones - 101, 102, 103, 104, 105, 106, 107, 108; Microphone assemblies - 110, 120, 130, 140; Base - 150; First side - 151; Second side - 152; Third side - 153; Fourth side - 154; Obstructions - 201, 202, 203. Detailed Implementation

[0055] The following specific examples illustrate the implementation of the present invention. The illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0056] Example 1

[0057] like Figure 1 As shown, a selected embodiment of the microphone array compensation optimization method of the present invention includes the following steps:

[0058] S110. Multiple sound source emission angles are pre-selected as correction angles, and the overall signal strength value of the audio signals collected by the microphone array 100 at each correction angle is measured as the measurement value. During the test, the sound source is placed directly below the microphone array 100. The overall signal strength value is the signal strength value of the signals collected by all microphones (32 microphones in this embodiment) of the microphone array 100 after fusion.

[0059] When selecting the sound source's emission angle as the correction angle, based on the variation of the microphone array's overall signal strength with the sound source's emission angle, the range where the microphone array's overall signal strength increases with the sound source's emission angle is divided into a linearly increasing range, and the range where the microphone array's overall signal strength decreases with the sound source's emission angle is divided into a linearly decreasing range. Analysis of the test data from microphone array 100 reveals that when the sound source's emission angle is between 0° and 180°, the overall signal strength of the audio signal acquired by microphone array 100 increases linearly with the sound source's emission angle; conversely, when the sound source's emission angle is between 180° and 360°, the overall signal strength of the audio signal acquired by microphone array 100 decreases linearly with the sound source's emission angle. Therefore, the range of 0° to 180° is designated as the linearly increasing range, and the range of 180° to 360° is designated as the linearly decreasing range. Among them, a correction angle of 0° means that the sound source is emitting sound directly at the center point of the microphone array 100; a correction angle of 180° means that the sound source is emitting sound directly away from the center point of the microphone array 100; and a correction angle of 360° means that the sound source is emitting sound one revolution and then returns to the direction that is emitting sound directly at the center point of the microphone array 100.

[0060] Next, multiple angles are selected at equal intervals within the linearly increasing and decreasing intervals, and these angles, along with 0°, 180°, and 360°, are used as correction angles. For example, when 30° is used as the interval between the linearly increasing and decreasing intervals, the correction angles include 0°, 30°, 60°, 90°, 120°, 150°, 180°, 210°, 240°, 270°, 300°, 330°, and 360°. Of course, to improve compensation accuracy, smaller angle values ​​can be used for the intervals.

[0061] After obtaining the measured values ​​through testing, the overall signal strength value of the microphone array 100 can be obtained as a function curve of the sound source emission angle through data fitting. Based on the function curve, each measured value is corrected so that the corrected measured values ​​are all located on the function curve, thereby eliminating measurement errors.

[0062] S120. Obtain the tilt angle of the microphone array 100 after installation, that is, the angle between the microphone array 100 and the horizontal direction after installation. An angle sensor can be installed on the microphone array 100, and the tilt angle of the microphone array 100 after installation can be measured by the angle sensor. Figure 2 As shown, the dashed line represents the horizontal direction, and θ represents the tilt angle of the microphone array 100 after installation.

[0063] S130. Based on the measured values ​​obtained from pre-tests, a tilt correction value corresponding to the tilt angle is calculated using a linear interpolation method, and gain compensation is performed on the audio signal actually acquired by the microphone array 100 based on the tilt correction value. For example... Figure 3 As shown, this step may include the following sub-steps:

[0064] S131. Select a correction angle smaller than the tilt angle as the first interpolation angle, and select a correction angle larger than the tilt angle as the second interpolation angle; the first interpolation angle and the second interpolation angle are in the same interval (i.e., both the first interpolation angle and the second interpolation angle are in a linearly increasing interval or both are in a linearly decreasing interval). Generally, the correction angle that is smaller than the tilt angle and closest to the tilt angle is selected as the first interpolation angle, and the correction angle that is larger than the tilt angle and closest to the tilt angle is selected as the second interpolation angle.

[0065] S132. Since the measured values ​​obtained in step S110 reflect that the overall signal strength of the audio signal acquired by the microphone array 100 is approximately linearly related to the sound source's emitting angle within both the linearly increasing and linearly decreasing intervals; and since the function curve of the overall signal strength value of the microphone array 100 changing with the sound source's emitting angle, obtained through fitting, is also approximately a straight line within both the linearly increasing and linearly decreasing intervals, the theoretical signal strength value at this tilt angle can be calculated using the following formula:

[0066]

[0067] Where θ represents the tilt angle of the microphone array 100 after installation, that is, the angle between the microphone array 100 and the horizontal direction after installation; signal(θ) represents the theoretical signal strength value when the tilt angle is θ; a represents the first interpolation angle; b represents the second interpolation angle; val(a) represents the measured value corresponding to the first interpolation angle; val(b) represents the measured value corresponding to the second interpolation angle.

[0068] S133. Calculate the tilt correction value corresponding to a tilt angle of θ using the following formula:

[0069] signal(correction)=signal(0)-signal(θ)

[0070] Where signal(correction) represents the tilt correction value; signal(0) represents the measured value when the correction angle is 0°, which is the expected overall signal strength value.

[0071] S134. Perform gain compensation on the overall signal strength value of the microphone array 100 based on the tilt correction value; the gain compensation formula is as follows:

[0072] signal(compensation)=signal(measured)+signal(correction)

[0073] Wherein, signal(compensation) represents the overall signal strength value of microphone array 100 after gain compensation, and signal(measured) represents the overall signal strength value of microphone array 100 before gain compensation.

[0074] For example, when the tilt angle θ = 50°, 30° can be taken as the first interpolation angle and 60° as the second interpolation angle. Substituting the measured values ​​corresponding to the correction angles of 30° and 60° into the above formula, the tilt correction value signal(correction) can be obtained. This increases the overall signal strength value of the microphone array 100 by signal(correction), thus completing the tilt angle compensation of the microphone array 100.

[0075] like Figure 3 As shown, the installation environment may also be affected by obstructions around the microphone array 100. Under different obstruction conditions, the gain of the microphone array 100 will change, thus affecting its performance and accuracy, and negatively impacting the identification of the sound source direction. Therefore, to compensate for the impact of obstructions on the microphone array 100, the installation environment compensation method may further include the following steps:

[0076] S210. The occlusion conditions of the microphone array 100 are classified in advance, and the occlusion compensation value of each microphone of the microphone array 100 under various occlusion conditions is obtained through testing.

[0077] The microphone array 100 includes a base 150 and multiple microphones disposed on the base 150. The multiple microphones are arranged in a geometric array. The base 150 is a regular polygon, and the geometric shape formed by the arrangement of the multiple microphones is a rotationally symmetric figure, wherein the rotation angle of the rotationally symmetric figure is equal to the exterior angle of the regular polygon. Figure 4as shown, in this embodiment, the base 150 is square. Since the exterior angle of a regular polygon is 90°, correspondingly, the rotation angle of the rotationally symmetric pattern formed by the arrangement of the plurality of microphones should be equal to 90°; therefore, in this embodiment, four microphone assemblies (110, 120, 130, 140) arranged at intervals of 90° in sequence are adopted. As Figure 5 as shown, each microphone assembly includes eight linearly arranged microphones (101, 102, 103, 104, 105, 106, 107, 108), so that 32 microphones are arranged to form an inclined "cross" shape. Certainly, the geometric pattern formed by the arrangement of the plurality of microphones of the microphone array 100 may also be a geometric pattern with a rotation angle of 90° such as a "star" shape, a square, a circle, etc. When the microphone array 100 is rotated 90° clockwise, the microphone assembly 110 is rotated to the previous position of the microphone assembly 120, the microphone assembly 120 is rotated to the previous position of the microphone assembly 130, the microphone assembly 130 is rotated to the previous position of the microphone assembly 140, and the microphone assembly 140 is rotated to the previous position of the microphone assembly 110, so that the geometric pattern formed after the 32 microphones are rotated by 90° is consistent with that before the rotation. Based on this feature, for the square microphone array 100 adopted in this embodiment, the occlusion situation classification may include five types: one-surface occlusion, adjacent two-surface occlusion, opposite two-surface occlusion, three-surface occlusion, and four-surface occlusion.

[0078] To facilitate numbering each microphone of the microphone array 100, the four sides of the microphone array 100 are sequentially defined as a first side 151, a second side 152, a third side 153, and a fourth side 154 in the clockwise direction, the microphone assembly aligned with the included angle of the first side 151 and the fourth side 154 is defined as the microphone assembly 110, and the other three microphone assemblies are sequentially defined as the microphone assembly 120, the microphone assembly 130, and the microphone assembly 140 in the clockwise direction. The microphones in each microphone assembly are sequentially defined as microphone 101, microphone 102, microphone 103, microphone 104, microphone 105, microphone 106, microphone 107, and microphone 108 in the order from the outer side of the base 150 to the center of the base 150.

[0079] In this step, obtaining the occlusion compensation value of each microphone of the microphone array 100 under various occlusion situations through testing includes the following sub-steps:

[0080] S211. Number each microphone in the test microphone array 100 according to its position. Since the microphone array 100 in this embodiment includes four microphone components, the microphone number can include both a group number and a number within the group, using "group number" + "number within the group" for microphone numbering. For example, the group number of microphone component 110 can be determined as 11, the group number of microphone component 120 as 12, the group number of microphone component 130 as 13, and the group number of microphone component 140 as 14. Set the number within the group of microphone 101 to 01, the number within the group of microphone 102 to 02, the number within the group of microphone 103 to 03, the number within the group of microphone 104 to 04, the number within the group of microphone 105 to 05, the number within the group of microphone 106 to 06, the number within the group of microphone 107 to 07, and the number within the group of microphone 108 to 08. That is, microphone 101 in microphone assembly 110 is numbered "1101", microphone 101 in microphone assembly 120 is numbered "1201", and so on, thereby assigning a unique number to each microphone in microphone array 100.

[0081] S212. Test the microphone array 100 under unobstructed conditions to obtain the signal strength of each numbered microphone under unobstructed conditions. By testing the signal strength of the microphones under unobstructed conditions, and using the signal strength of the microphones under unobstructed conditions as a benchmark, compensate for the microphone gain changes caused by various obstruction conditions.

[0082] S213. Record the relative positional relationship between each numbered microphone and the obstruction under various occlusion conditions, and test the microphone array 100 to obtain the signal strength of each numbered microphone under various occlusion conditions. Specifically, one side of the base 150 can be taken as the positioning side, and the relative positional relationship between the positioning side and the obstruction, as well as the relative positional relationship between each numbered microphone and the positioning side, can be recorded. The following uses the first side 151 as the positioning side as an example to illustrate the five types of occlusion conditions.

[0083] An obstruction 201 is placed at the first side 151 (i.e., the positioning side) of the microphone array 100, thus creating a one-sided obstruction. At this time, the positional relationships between the second side 152, the third side 153, the fourth side 154, the microphone assembly 110, the microphone assembly 120, the microphone assembly 130, the microphone assembly 140, the obstruction 201, and the first side 151 are as follows: Figure 6 As shown.

[0084] Obstacles 201 and 202 are placed on the first side 151 (i.e., the positioning side) and the second side 152 of the microphone array 100, respectively, thus creating a situation where adjacent sides are obscured. At this time, the positional relationships between the second side 152, the third side 153, the fourth side 154, the microphone assembly 110, the microphone assembly 120, the microphone assembly 130, the microphone assembly 140, the obstacles 201 and 202, and the first side 151 are as follows: Figure 7 As shown.

[0085] Obstacles 203 and 202 are placed on the fourth side 154 and the second side 152 of the microphone array 100, respectively, thus creating a situation of relative two-sided obstruction. At this time, the positional relationships between the second side 152, the third side 153, the fourth side 154, the microphone assembly 110, the microphone assembly 120, the microphone assembly 130, the microphone assembly 140, the obstacles 203 and 202, and the first side 151 are as follows: Figure 8 As shown.

[0086] Obstacles 203, 201, and 202 are placed on the fourth side 154, the first side 151, and the second side 152 of the microphone array 100, respectively, thus forming a three-sided obstruction. At this time, the positional relationship between the second side 152, the third side 153, the fourth side 154, the microphone assembly 110, the microphone assembly 120, the microphone assembly 130, the microphone assembly 140, the obstacles 203, 201, and 202, and the first side 151 is as follows: Figure 9 As shown.

[0087] Obstacles are placed on the first side 151, second side 152, third side 153, and fourth side 154 of the microphone array 100, thus creating a four-sided obstruction. Since the shape formed by the obstacles is also a rotationally symmetric shape with a rotation angle of 90°, there is only one positional relationship between the microphone assembly 110, microphone assembly 120, microphone assembly 130, microphone assembly 140, and the four obstacles.

[0088] S214. Subtract the signal strength of the microphone numbered under the unobstructed condition from the signal strength of the microphone numbered under each type of obstruction condition to obtain the obstruction compensation value of each microphone numbered under the five types of obstruction conditions: one-sided obstruction, two adjacent obstructions, two opposite obstructions, three-sided obstruction, and four-sided obstruction.

[0089] It should be noted that the microphone array used for testing is not necessarily the same as the microphone array installed later, meaning the occlusion compensation values ​​obtained from the test are universal. Furthermore, the numbering of each microphone in the microphone array during testing is only for recording the positional relationship between each microphone and the obstruction, and is not a fixed numbering system for the microphones. That is, the numbering indicates the positional relationship between the microphone and the obstruction; for the same microphone array, the microphone numbering will change accordingly when the obstruction is placed in different locations.

[0090] S220. After the microphone array 100 is installed, the installation environment of the microphone array 100 (i.e., the type of obstruction around the microphone array 100 and the location of the obstruction) is identified. Infrared sensors can be installed on the four sides of the base 150 to identify whether there are obstacles on the four sides. When determining the type of obstruction, if an infrared sensor on one side detects an obstruction, it is determined to be one-sided obstruction; if infrared sensors on two sides detect an obstruction, it is further determined whether the two sides that detected the obstruction are adjacent. If they are adjacent, it is determined to be two-sided obstruction; otherwise, it is determined to be two-sided obstruction; if infrared sensors on three sides detect an obstruction, it is determined to be three-sided obstruction. When determining the location of the obstruction, the four sides of the base 150 can be divided into four directions: front, back, left, and right. Based on the detection results of the infrared sensors, it can be determined whether there are obstructions in the front, back, left, and right directions respectively.

[0091] S230. Based on the type and location of occlusion, determine the occlusion compensation value for each microphone in the microphone array 100, and compensate the gain of each microphone according to the occlusion compensation value. This step may include the following sub-steps:

[0092] S231. Based on the identified occlusion type, occlusion location, and the recorded relative positional relationship between each microphone number and the occluder, determine the number of each microphone in the microphone array 100. At this time, first determine the positioning side of the base 150 of the microphone array 100 based on the occlusion type and occlusion location, and then determine the number of each microphone in the microphone array 100 based on the relative positional relationship between each microphone number and the positioning side. For example: when the identified occlusion type is one-sided occlusion, and the occluder is on the left side of the microphone array 100, then the left side of the base 150 is the first side 151 (i.e., the positioning side). Figure 5 (Will Figure 5Rotating counterclockwise by 90 degrees (meaning the obstruction is on the left side of microphone array 100) allows us to determine the positions of microphone components 110, 120, 130, and 140. At this point, the determined positions of the first side 151, second side 152, third side 153, fourth side 154, microphone components 110, 120, 130, 140, and the obstruction 201 are as follows: Figure 10 As shown, the group number of each microphone can be determined. Of course, in the case of four-sided obstruction, any side of the base 150 can be designated as the first side 151, thereby determining the positions of the second side 152, the third side 153, the fourth side 154, the microphone assembly 110, the microphone assembly 120, the microphone assembly 130, and the microphone assembly 140.

[0093] Since the four microphone components are arranged at 90° intervals, the group number of each microphone remains unchanged regardless of how the microphone array 100 is installed; only the group number changes depending on the direction of the obstruction. For example, when the obstruction is located behind the microphone array 100, the microphone component in the rear left corner is microphone component 110; when the obstruction is located on the left side of the microphone array 100, the microphone component in the rear left corner becomes microphone component 120. That is, due to the different positions of the obstruction, the group number of the microphone component in the rear left corner changes from "11" to "12". However, the group number of microphone 101 remains "01" regardless of the group number change. Therefore, once the group number of each microphone is determined, the number of each microphone can be determined.

[0094] S232. Based on the identified occlusion type, the occlusion compensation value corresponding to each microphone number is obtained. Since the occlusion compensation value corresponding to each microphone number has already been obtained in S104, the occlusion compensation value corresponding to that microphone can be retrieved after determining the number of each microphone.

[0095] S233. Adjust the parameters of the microphone array 100 according to the occlusion compensation value, and perform gain compensation on each numbered microphone to eliminate the influence of the occlusion on the microphone array 100.

[0096] In this embodiment, multiple calibration angles are pre-selected for testing to obtain measurement values. These values ​​can be used to calculate the tilt correction value corresponding to the tilt angle using a linear interpolation method. This tilt correction value is then used to compensate for the overall signal strength of the microphone array 100, significantly reducing the impact of tilted installation on the microphone array 100. When there are obstructions around the microphone array 100, by pre-testing the obstruction compensation value of each microphone under various obstruction conditions, the gain of each microphone can be compensated according to the installation environment. This greatly reduces the impact of obstructions on the microphone array 100 caused by different installation methods, avoiding negative impacts on the identification of the sound source direction. Furthermore, by not fixing the microphone numbers before installation, and identifying the location of the obstruction after installation, and then determining the microphone numbers based on the location of the obstruction, compared to fixing microphone numbers, any side of the microphone array 100 can be aligned with the obstruction during installation, eliminating the need to align a fixed side of the microphone array 100 with the obstruction during installation, making installation more flexible.

[0097] Example 2

[0098] like Figure 11 As shown, one preferred embodiment of the audio acquisition device of the present invention includes a microphone array 100, an audio processing unit, and a compensation unit. The microphone array 100 is an array of multiple microphones arranged geometrically on a base 150; in this embodiment, the microphone array 100 includes four microphone assemblies (110, 120, 130, 140) arranged at 90° intervals. Each microphone assembly includes eight microphones (101, 102, 103, 104, 105, 106, 107, 108) arranged linearly, thereby using 32 microphones to form an inclined "+" shape.

[0099] The audio processing unit is used to perform sound source localization and speech enhancement processing on the signals acquired by the microphone array 100. The audio processing unit may include a preprocessing module, a sound source localization module, a speech enhancement module, and an output module. The preprocessing module is used to preprocess the signals acquired by the microphone array 100, including filtering, gain control, and noise cancellation. The sound source localization module is used to identify the direction and distance of the sound source based on the signals acquired by the microphone array 100; algorithms such as beamforming or Gaussian mixture models can be used to identify the direction and distance of the sound source. The speech enhancement module is used to effectively form a beam in the desired direction using beamforming technology, picking up only the signal within the beam, thereby achieving the purpose of simultaneously extracting the sound source and suppressing noise; and to improve the quality and reliability of the speech by processing the acquired speech signal; typically, techniques including noise reduction, echo suppression, and sound enhancement are used to process the speech signal. The output module is used to output the signal processed by the speech enhancement module to subsequent audio processing devices or embedded systems.

[0100] The compensation unit is used to compensate the gain of the microphone array 100 using an optimization method for microphone array compensation according to Embodiment 1. The installation environment compensation unit may include a compensation value storage module, an identification module, and an adaptive algorithm module. The compensation value storage module stores the measurement values ​​obtained in step S110 after measuring the overall signal strength of the audio signals acquired by the microphone array 100 at various correction angles. When compensation is required based on occlusion conditions, the compensation value storage module is also used to store the occlusion compensation values ​​of each microphone in the microphone array 100 under various occlusion conditions, obtained in step S210.

[0101] The identification module is used to identify the tilt angle of the microphone array 100 after installation. An angle sensor can be used to collect the angle information of the microphone array 100 to identify the tilt angle of the microphone array 100, i.e., to execute step S120 in Embodiment 1. When compensation based on occlusion is required, the identification module is also used to identify the occlusion around the microphone array 100 and the location of the occluding objects. An infrared sensor can be used to collect the location information of the occluding objects around the microphone array 100, and the occlusion around the microphone array 100 and the location of the occluding objects can be identified based on the location information. Of course, other methods can also be used to identify the occlusion around the microphone array 100 and the location of the occluding objects, i.e., to execute step S220 in Embodiment 1.

[0102] The adaptive algorithm module is used to calculate the tilt correction value corresponding to the tilt angle using a linear interpolation method, and to perform gain compensation on the overall signal strength value of the microphone array 100 based on the tilt correction value; that is, to execute step S130 in Embodiment 1. When compensation is required based on occlusion, the adaptive algorithm module is also used to obtain the occlusion compensation value corresponding to each microphone of the microphone array 100 based on the occlusion situation and occlusion direction, and to make the audio processing unit compensate the gain of each microphone of the microphone array 100 according to the occlusion compensation value, that is, to execute step S230 in Embodiment 1, so as to adaptively adjust the signal processing algorithm parameters of the audio processing unit according to the state of the microphone array 100 and the environmental conditions.

[0103] The audio acquisition device of this embodiment addresses the impact of the installation environment on the microphone array 100. It can compensate for the overall signal strength value of the microphone array 100 based on the tilt installation, thereby greatly reducing the impact of the tilt angle during installation on the reception effect of the microphone array 100, avoiding negative impact on the sound source direction identification of the microphone array 100, and improving the reception effect and accuracy of the microphone array 100.

[0104] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. An optimization method for microphone array compensation, characterized in that, Includes the following steps: S110. Multiple sound source emission angles are pre-selected as correction angles, and the overall signal strength of the audio signals acquired by the microphone array at each correction angle is measured as the measurement value. The method for selecting multiple sound source emission angles as correction angles is as follows: The range of sound source angles where the overall signal strength of the microphone array increases with the increase of the sound source angle is divided into a linearly increasing range, and the range of sound source angles where the overall signal strength of the microphone array decreases with the increase of the sound source angle is divided into a linearly decreasing range. Multiple angles are selected at equal intervals within both the linearly increasing and linearly decreasing ranges as correction angles. S120. Obtain the tilt angle of the microphone array after installation; S130. Based on the measured values ​​obtained from the pre-test, calculate the tilt correction value corresponding to the tilt angle using a linear interpolation method; and perform gain compensation on the overall signal strength value of the microphone array based on the tilt correction value; this step includes the following sub-steps: S131. Select a correction angle smaller than the tilt angle as the first interpolation angle, and select a correction angle larger than the tilt angle as the second interpolation angle. S132. The theoretical overall signal strength value when the tilt angle is θ is calculated using the following formula: Where θ represents the tilt angle of the microphone array after installation; signal(θ) represents the theoretical signal strength value when the tilt angle is θ; a represents the first interpolation angle; b represents the second interpolation angle; val(a) represents the measured value corresponding to the first interpolation angle; val(b) represents the measured value corresponding to the second interpolation angle. S133. Calculate the tilt correction value corresponding to a tilt angle of θ using the following formula: Where signal(correction) represents the tilt correction value; signal(0) represents the measurement value when the correction angle is 0°; S134. Perform gain compensation on the overall signal strength of the microphone array based on the tilt correction value; the gain compensation formula is as follows: Here, signal(compensation) represents the overall signal strength value of the microphone array after gain compensation, and signal(measured) represents the overall signal strength value of the microphone array before gain compensation.

2. The optimization method for microphone array compensation according to claim 1, characterized in that: The sound source emission angle range of the linearly increasing interval is 0° to 180°, and the sound source emission angle range of the linearly decreasing interval is 180° to 360°; the correction angle includes multiple angles selected at equal intervals between 0°, 180°, 360°, and between 0° and 180°, as well as multiple angles selected at equal intervals between 180° and 360°.

3. The optimization method for microphone array compensation according to claim 1, characterized in that: In step S110, after obtaining the measured values ​​through testing, the overall signal strength value of the microphone array is obtained as a function curve of the sound source emission angle through data fitting, and each measured value is corrected according to the function curve.

4. The optimization method for microphone array compensation according to claim 1, characterized in that: An angle sensor is installed on the microphone array to obtain the tilt angle of the microphone array after installation.

5. The optimization method for microphone array compensation according to any one of claims 1 to 4, characterized in that: In step S131, a correction angle that is smaller than the tilt angle and closest to the tilt angle is selected as the first interpolation angle, and a correction angle that is larger than the tilt angle and closest to the tilt angle is selected as the second interpolation angle.

6. An audio acquisition device, characterized in that: include A microphone array, wherein the microphone array is an array of multiple microphones arranged geometrically on a base; The audio processing unit is used to perform sound source localization and speech enhancement processing on the signals acquired by the microphone array; as well as The compensation unit is used to compensate the gain of the microphone array using an optimization method for microphone array compensation as described in any one of claims 1 to 5.

7. The audio acquisition device according to claim 6, characterized in that: The audio processing unit includes The preprocessing module is used to preprocess the signals acquired by the microphone array; The sound source localization module is used to identify the direction and distance of the sound source based on the signals collected by the microphone array; The speech enhancement module is used to form a beam in the desired direction using beamforming technology, thereby extracting the sound source and suppressing noise; and to process the acquired speech signal to improve the quality and reliability of the speech. as well as The output module is used to output the signal processed by the speech enhancement module.

8. The audio acquisition device according to claim 6, characterized in that: The compensation unit includes The compensation value storage module is used to store the measurement values ​​obtained after measuring the overall signal strength of the audio signals acquired by the microphone array at various correction angles; The recognition module is used to identify the tilt angle of the microphone array after installation; and The adaptive algorithm module is used to calculate the tilt correction value corresponding to the tilt angle using a linear interpolation method, and to perform gain compensation on the overall signal strength value of the microphone array based on the tilt correction value.

Images