Adaptive method and system for locating a sound source, electronic device and storage medium

Abstract

The embodiment of the present application relates to the technical field of audio processing, and discloses a positioning sound source adaptive method, system, electronic equipment and storage medium.The method comprises the following steps: determining whether a speaking condition is met based on a preset noise threshold and a data fluctuation level correlation; in the case that the speaking condition is met, triggering automatic adjustment of a microphone boom angle, in the automatic adjustment process, a third energy value is acquired in real time, and the angle at which the third energy value is maximum is taken as an optimal angle, and the angle of the microphone boom is adjusted to the optimal angle. By implementing the embodiment of the present application, interference caused by adjusting the angle of a conference microphone can be reduced, conference efficiency can be improved, the attention of participants can be maintained, the use experience of a conference microphone can be improved, compared with passively increasing the volume of a microphone, adaptive adjustment can improve the signal-to-noise ratio of a link without increasing the gain of the microphone, to a certain extent, the generation of howling is inhibited, and the sound transmission gain is improved.

Description

Technical Field

[0001] This invention relates to the field of audio processing technology, specifically to a method, system, electronic device, and storage medium for adaptive sound source localization. Background Technology

[0002] Conference microphones are the basic speaking units in a conference system. Currently, conference microphones are mainly operated by manually adjusting the microphone boom to ensure that the microphone is aimed at the speaker, capture the speaker's voice, and transmit the signal to the speakers to achieve system amplification. However, in real-world applications, users have different heights and do not always maintain a rigid posture (leaning forward, leaning back, standing). Individual differences in users, coupled with changes in posture, can cause the sound source position to change to varying degrees. Changes in the sound source position affect the microphone's sound pickup effect, reduce meeting efficiency and information transmission, and static, fixed-angle conference microphones are difficult to meet the real needs of users in a meeting. The distance between the current conference microphone and the person is generally kept at 80-100cm, and frequent adjustments affect the user experience. Summary of the Invention

[0003] To address the aforementioned shortcomings, this invention discloses a sound source localization adaptive method, system, electronic device, and storage medium. This solves the problem of sound source position changes caused by differences and variations in body posture among different users in actual meetings, leading to lost or inefficient sound acquisition. The goal is to achieve accurate and efficient dynamic sound acquisition and balance the voices of all speakers in the same meeting.

[0004] The first aspect of this invention discloses an adaptive method for locating a sound source, the method comprising:

[0005] Whether the speaking conditions are met is determined based on the correlation between the preset noise threshold and the data fluctuation level.

[0006] When the speaking conditions are met, the microphone boom elevation angle is automatically adjusted. During the automatic adjustment process, a third energy value is acquired in real time, and the elevation angle when the third energy value is at its maximum is taken as the optimal elevation angle. The microphone boom elevation angle is then adjusted to the optimal elevation angle.

[0007] As an optional implementation, in the first aspect of the present invention, determining whether the speaking conditions are met based on a preset noise threshold and the correlation between data fluctuation levels includes:

[0008] Select multiple samples and obtain the first energy value of the multiple samples;

[0009] The noise threshold and data fluctuation level are determined based on the first energy value;

[0010] Establish the correlation between the noise threshold and the data fluctuation level;

[0011] Determine whether the speaking conditions are met based on the aforementioned relationship.

[0012] As an optional implementation, in a first aspect of the present invention, establishing the correlation between the noise threshold and the data fluctuation level includes:

[0013] The first audio data of each sample is acquired in real time, and the first audio data is converted into a first digital signal at a preset sampling rate;

[0014] Obtain the root mean square of the first digital signal for each frame and record it as the first energy value of the audio data for each frame.

[0015] Determine the average and variance of multiple first energy values ​​for each sample within the collection period, and use the average of the multiple first energy values ​​as the noise threshold and the variance of the multiple first energy values ​​as the data fluctuation level;

[0016] The correlation between noise threshold and data fluctuation level is formed by the correlation between noise threshold and data fluctuation level for each sample.

[0017] As an optional implementation, in the first aspect of the present invention, determining whether the speaking conditions are met based on a preset noise threshold and the correlation between data fluctuation levels includes:

[0018] The system acquires second audio data from the target environment in real time and converts the second audio data into a second digital signal at a preset sampling rate.

[0019] Obtain the root mean square of the second digital signal for each frame, and record it as the second energy value of the second audio data for each frame;

[0020] Determine the average and variance of multiple second energy values ​​within the acquisition period, and use the average of the multiple second energy values ​​as the noise value and the variance of the multiple second energy values ​​as the data fluctuation value;

[0021] A target noise threshold corresponding to the noise value is determined. When the data fluctuation value meets the target data fluctuation level, the target environment meets the speaking conditions. The target data fluctuation level is the data fluctuation level corresponding to the target noise threshold determined based on the preset noise threshold and the correlation between the data fluctuation level and the data fluctuation level.

[0022] As an optional implementation, in the first aspect of the present invention, the step of determining whether the speaking condition is met based on a preset noise threshold and the correlation between data fluctuation levels further includes:

[0023] Liveness detection is performed on the target environment, and the speech condition is triggered when a human body is present in the target environment;

[0024] Or / and, the time interval for triggering automatic adjustment of the microphone boom elevation angle shall not be less than the preset duration.

[0025] As an optional implementation, in the first aspect of the present invention, during the automatic adjustment process, a third energy value is acquired in real time, and the elevation angle at which the third energy value is at its maximum is taken as the optimal elevation angle. The elevation angle of the microphone boom is then adjusted to the optimal elevation angle, comprising:

[0026] When the microphone pole elevation angle is automatically adjusted from the first elevation angle to the second elevation angle, the third audio data during the adjustment process is collected in real time, and the third audio data is converted into a third digital signal at a preset sampling rate;

[0027] Obtain the root mean square of all third digital signals and record it as the third energy value corresponding to the second elevation angle;

[0028] Determine the third energy value for all elevation angle positions when the microphone pole elevation angle is adjusted, and record the elevation angle position with the maximum third energy value as the optimal elevation angle;

[0029] Control the microphone boom to adjust to the optimal elevation angle.

[0030] A second aspect of this invention discloses a localization sound source adaptive system, comprising: a processor, a codec, a driving mechanism, and an angle detection mechanism, wherein:

[0031] The codec is used to acquire audio data and convert the audio data into digital signals before sending them to the processor;

[0032] The angle detection mechanism is used to collect the elevation angle of the microphone pole and send the collected elevation angle of the microphone pole to the processing unit.

[0033] The processor is used to determine whether the speaking conditions are met based on a preset noise threshold and the correlation between data fluctuation levels; and if the speaking conditions are met, it triggers the drive mechanism to adjust the elevation angle of the microphone pole. During the process of the drive mechanism adjusting the microphone pole, the processor also acquires a third energy value in real time, takes the elevation angle when the third energy value is at its maximum as the optimal elevation angle, and controls the elevation angle of the microphone pole to be adjusted to the optimal elevation angle through the drive mechanism.

[0034] As an optional implementation, in a second aspect of the present invention, the location sound source adaptive system further includes a liveness detection mechanism, which is used to perform liveness detection on the target environment and send the liveness detection signal to a processor. The processor triggers the discrimination of the speaking condition based on the liveness detection signal. The liveness detection mechanism is an infrared detection mechanism.

[0035] or / and,

[0036] The driving mechanism is a stepper motor;

[0037] or / and,

[0038] The angle detection mechanism is a six-axis gyroscope.

[0039] A third aspect of the present invention discloses an electronic device, comprising: a memory storing executable program code; a processor coupled to the memory; the processor calling the executable program code stored in the memory to execute a localization sound source adaptive method disclosed in the first aspect of the present invention.

[0040] A fourth aspect of the present invention discloses a computer-readable storage medium storing a computer program, wherein the computer program causes a computer to execute a sound source localization adaptive method disclosed in the first aspect of the present invention.

[0041] The fifth aspect of this invention discloses a computer program product that, when run on a computer, causes the computer to execute a sound source localization adaptive method disclosed in the first aspect of this invention.

[0042] The sixth aspect of this invention discloses an application publishing platform for publishing computer program products, wherein when the computer program products are run on a computer, the computer executes a sound source localization adaptive method disclosed in the first aspect of this invention.

[0043] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0044] In this embodiment of the invention, the present invention determines whether someone is speaking by establishing a correlation between a preset noise threshold and data fluctuation levels. Then, it automatically adjusts the microphone boom based on a third energy value during the speaking process. This reduces interference caused by adjusting the microphone angle, improves meeting efficiency, maintains the attention of participants, and enhances the microphone user experience. Compared to passively increasing the microphone volume, adaptive adjustment can improve the link signal-to-noise ratio without increasing the microphone gain, suppresses feedback to a certain extent, and improves the transmission gain. Attached Figure Description

[0045] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0046] Figure 1 This is a flowchart illustrating an adaptive method for locating a sound source disclosed in an embodiment of the present invention;

[0047] Figure 2 This is a flowchart illustrating whether the speaking conditions are met, as disclosed in an embodiment of the present invention.

[0048] Figure 3 This is a structural block diagram of a localization sound source adaptive system disclosed in an embodiment of the present invention;

[0049] Figure 4 This is a structural block diagram of the electronic device disclosed in an embodiment of the present invention. Detailed Implementation

[0050] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0051] It should be noted that the terms "first," "second," "third," "fourth," etc., in the specification and claims of this invention are used to distinguish different objects, not to describe a specific order. The terms "comprising" and "having," and any variations thereof, in the embodiments of this invention are intended to cover non-exclusive inclusion. Exemplarily, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to these processes, methods, products, or devices.

[0052] This invention determines whether someone is speaking by establishing a correlation between a preset noise threshold and data fluctuation levels. Then, it automatically adjusts the microphone boom based on a third energy value during the speaking process. This reduces interference caused by adjusting the microphone angle, improves meeting efficiency, maintains participant focus, and enhances the microphone user experience. Compared to passively increasing microphone volume, adaptive adjustment improves the link signal-to-noise ratio without increasing microphone gain, suppresses feedback to some extent, and increases transmission gain. The following is a detailed description in conjunction with the accompanying drawings.

[0053] Example 1

[0054] Please see Figure 1 , Figure 1 This is a flowchart illustrating a sound source localization adaptive method disclosed in an embodiment of the present invention. The method described in this embodiment is applicable to the Linux ALSA sound card architecture. Figure 2As shown, the adaptive method for locating sound sources includes the following steps:

[0055] S110. Perform liveness detection on the target environment. If a human body is present in the target environment, trigger the judgment of the speaking condition.

[0056] The target environment refers to the location where the conference microphone is located, such as a conference room. Liveness detection is used to determine if someone is near the conference microphone. If someone is near, the processor is activated to perform sound acquisition and adaptive adjustment of the microphone angle.

[0057] Adding liveness detection can save power by putting the microphone into sleep mode when no one is near it, and it can also enable unmanned operation throughout the meeting, improving the user experience.

[0058] There are various methods for liveness detection. For example, infrared sensors, such as infrared thermometers and infrared beam tubes, can be used. Other detection mechanisms, such as millimeter-wave radar, can also be employed. The liveness detection mechanism is mounted on the conference microphone, with a preferred detection distance of approximately 1 meter. A human body present at approximately 1 meter can be detected. Understandably, the closer the human body is to the conference microphone, the smaller the detection error.

[0059] When a human body is detected, it indicates that the person may use the conference microphone. Then, it is determined whether someone is speaking, which is the operation of step S120. If someone is speaking, the microphone angle is automatically adjusted, which is the operation of step S130.

[0060] S120. Determine whether the speaking conditions are met based on the preset noise threshold and the correlation between data fluctuation levels.

[0061] If someone speaks, step S130 will be executed; otherwise, step S120 will be performed in real time until the speaking condition is met or the liveness detection mechanism cannot detect the presence of a human body within the detection range.

[0062] Someone speaking means there is a certain amount of sound. Under normal circumstances, the sound fluctuation before and after the same person speaks is within a certain range. Based on this, we can determine whether someone has spoken by using noise threshold and data fluctuation level.

[0063] The correlation between the preset noise threshold and data fluctuation level was obtained through multiple trials with different samples, all starting from the initial microphone angle. These different samples included multiple samples from multiple individuals. The participants could be a number of men and a number of women, with each person at different distances from the microphone forming multiple samples, thus creating a sample set. For example, assuming a group of 10 men and 10 women, with distances to the microphone of 0.1m, 0.2m, 0.3m, 0.5m, and 1.0m respectively, the sample set would include 20 × 5 = 100 samples. In reality, to ensure the accuracy of the speaking condition discrimination, a larger sample size could be used.

[0064] A correlation is established between the noise threshold and the data fluctuation level determined for each sample, forming a relationship between the noise threshold and the data fluctuation level. This correlation can be implemented using a two-dimensional table; for example, when the noise threshold is within a certain range, it corresponds to a certain range of data fluctuation levels.

[0065] In some other embodiments, a functional relationship model between the noise threshold and the data fluctuation level can also be established. For example, the relationship curve between the noise threshold and the data fluctuation level can be linearly fitted using the least squares method.

[0066] It should be noted that, after fitting the curve, in order to avoid filtering out most of the cases that meet the speaking conditions, in a preferred embodiment of the present invention, a certain range can be set for one of the parameters, namely the noise threshold and the data fluctuation level. For example, assuming that the noise threshold is 50dB and the data fluctuation level obtained by the relationship curve is 5dB, a range can be set for the data fluctuation level, such as 5dB ± 2.

[0067] For the specific process of establishing the correlation between the noise threshold and the data fluctuation level, please refer to [link / reference needed]. Figure 2 As shown, it includes the following steps:

[0068] S121. Collect the first audio data of each sample in real time, and convert the first audio data into a first digital signal at a preset sampling rate.

[0069] That is, each person speaks once in a different position, and then the sound signal of the speech is recorded. This sound signal is then converted into a digital signal and recorded as the first digital signal.

[0070] After the conference microphone collects the sound signal, it obtains the first audio data of each sample. Then, the first audio data is converted by AD through an audio codec to obtain the first data signal. The sampling rate of the audio codec can be set as needed. In a preferred embodiment of the present invention, the sampling rate can be 48000Hz. The length of each frame sample collected is 480 samples, that is, the audio codec processes the signal according to the length of one frame every 10ms to obtain the first digital signal in 16-bit short format.

[0071] S122. Obtain the root mean square of the first digital signal of each frame and record it as the first energy value of the audio data of each frame.

[0072] Since the first digital signal is in 16-bit short format, the root mean square (RMS) X of the first digital signal can be calculated:

[0073]

[0074] Where, x i Let be the value of the i-th bit in the first digital signal.

[0075] The root mean square X of the first digital signal of each frame is used as the first energy value of the audio data of each frame.

[0076] S123. Determine the average value and variance of multiple first energy values ​​for each sample within the collection period, use the average value of the multiple first energy values ​​as the noise threshold, and use the variance of the multiple first energy values ​​as the data fluctuation level.

[0077] The calculation formulas are as follows:

[0078]

[0079]

[0080] Where S1 is the noise threshold of the sample, S2 is the data fluctuation level of the sample, and X j Let be the first energy value of the audio data in the j-th frame.

[0081] The acquisition period refers to the duration of picking up each sample sound signal. In a preferred embodiment of the present invention, the acquisition period can be set to 30s. It can be understood that each sample corresponds to 3000 frames (groups) of first energy values, i.e., n=3000. In order to reduce the error caused by distortion or other factors, in a preferred embodiment of the present invention, some first energy values ​​can be removed. For example, 10 groups of the maximum value of the first energy value and 10 groups of the minimum value of the first energy value can be removed, and the remainder is used to calculate the noise threshold and data fluctuation level.

[0082] S124. The correlation between noise threshold and data fluctuation level is formed by the correlation between noise threshold and data fluctuation level for each sample.

[0083] As described in step S110, the relationship between noise threshold and data fluctuation level can be represented by a two-dimensional table formed by interval relationships, or by a function model.

[0084] S125. Determine whether the speaking conditions are met based on the aforementioned relationship.

[0085] In practical applications, second audio data from the target environment is acquired in real time, and the energy value of the second audio data is determined using the same method as steps S121-S123, and recorded as the second energy value of the second audio data. The process also involves first converting the second audio data into a digital signal (referred to as the second digital signal for distinction) using an audio codec at a preset sampling rate, and then obtaining the root mean square of each frame of the second digital signal, which is recorded as the second energy value of each frame of the second audio data.

[0086] Calculate the average and variance of multiple second energy values ​​within the acquisition period, and use the average of the multiple second energy values ​​as the noise value and the variance of the multiple second energy values ​​as the data fluctuation value;

[0087] A target noise threshold corresponding to the noise value is determined. When the data fluctuation value meets the target data fluctuation level, the target environment meets the speaking conditions. The target data fluctuation level is the data fluctuation level corresponding to the target noise threshold determined based on the preset noise threshold and the correlation between the data fluctuation level and the data fluctuation level.

[0088] It is understandable that the target data fluctuation level corresponding to the data fluctuation value can be determined, and then the relationship between the noise value and the target noise threshold can be used to determine whether the speaking conditions are met.

[0089] S130. When the speaking conditions are met, the microphone pole elevation angle is automatically adjusted. During the automatic adjustment process, the third energy value is acquired in real time, and the elevation angle when the third energy value is at its maximum is taken as the optimal elevation angle. The elevation angle of the microphone pole is then adjusted to the optimal elevation angle.

[0090] The conference microphone is equipped with a drive mechanism, which includes a stepper motor and related gears. In a preferred embodiment of the invention, the drive mechanism can be mounted on the movable part connecting the microphone boom to the panel, thereby enabling the microphone boom to be adjusted vertically within a 60-degree range.

[0091] To avoid frequent automatic adjustments to the microphone boom elevation angle, in a preferred embodiment of the present invention, the automatic adjustment frequency can be limited by setting a frequency. For example, the frequency of automatic adjustment of the microphone boom elevation angle can be set to 5 minutes, that is, the time interval between the end of the current adjustment and the start of the next adjustment is 5 minutes. During these 5 minutes, the conference microphone will no longer perform adaptive adjustments.

[0092] Of course, it can also be achieved based on other parameters, such as setting a preset threshold for the change (absolute value) of the third energy value. When the change of the third energy value is greater than the preset threshold, it is very likely that someone else will speak or the distance of the speaker has changed significantly. At this time, the microphone pole tilt angle adjustment can be triggered.

[0093] During microphone boom elevation adjustment, the optimal value within the elevation adjustment range (0-60°) can be found by measuring the magnitude of the third energy value. This improves the microphone user experience without the speaker noticing. This method reduces interference caused by adjusting the microphone angle, improves meeting efficiency, and maintains the attention of participants. Compared to passively increasing the microphone volume, adaptive adjustment can improve the link signal-to-noise ratio, suppress feedback to some extent, and improve transmission gain without increasing the microphone gain.

[0094] Specifically, based on the X, Y, and Z values ​​obtained from the six-axis gyroscope, the microphone boom elevation angle is calculated. Within the range of 0 to 60 degrees, the microphone boom elevation angle is moved by 1 degree every 100ms. During the movement, 10 sets of digital signals can be obtained according to the above sampling rate, which are recorded as the third digital signals. Then, the root mean square of these 10 sets of digital signals is used as the third energy value, thus obtaining the third energy value at each angle, for a total of 60 third energy values. The microphone boom elevation angle corresponding to the maximum value among these third energy values ​​is recorded as the optimal elevation angle, thereby determining that the transmission gain is the highest at the optimal elevation angle. The control drive mechanism is then used to adjust the microphone boom elevation angle to the optimal elevation angle.

[0095] For example, when the microphone boom elevation angle is automatically adjusted from the first elevation angle to the second elevation angle (the third digital signal of the microphone boom at its initial position can be the 10 most recent sets of data before the drive mechanism is started), the third audio data during the adjustment process is collected in real time, and the third audio data is converted into a third digital signal at a preset sampling rate; the root mean square of all third digital signals is obtained and recorded as the third energy value corresponding to the second elevation angle; the third energy value of the microphone boom elevation angle is determined in this way when it is adjusted to all elevation angles, and the elevation angle at which the third energy value is the largest is recorded as the optimal elevation angle, and the microphone boom is controlled to adjust to the optimal elevation angle.

[0096] In summary, the embodiments of the present invention can intelligently adjust for the differences in height and posture of different users, solving the problem of sound source position changes caused by differences and changes in posture of different users in actual meetings, resulting in sound acquisition loss or low efficiency, so as to achieve accurate and efficient dynamic sound acquisition and a setting method to achieve equalization of the various speaking voices in the meeting.

[0097] The embodiments of the present invention can solve the problems of traditional conference systems, where speakers unconsciously and frequently adjust the microphone angle to ensure optimal sound gain during the meeting, thereby affecting the speaking process and experience, as well as abnormal noise and feedback caused by manual touch.

[0098] The benefits of this are as follows:

[0099] 1. It can reduce interference caused by adjusting the angle of the conference microphone, improve meeting efficiency, and keep the participants focused.

[0100] 2. Improve the user experience of conference microphones and make them more intelligent through algorithmic learning.

[0101] 3. Compared to passively increasing microphone volume, adaptive angle adjustment improves the link signal-to-noise ratio without increasing microphone gain, suppresses feedback to some extent, and increases transmission gain.

[0102] 4. It can effectively improve meeting quality and customer experience.

[0103] Example 2

[0104] Please see Figure 3 , Figure 3 This is a schematic diagram of the structure of a localization sound source adaptive system disclosed in an embodiment of the present invention. Figure 3 As shown, the adaptive positioning sound source system may include: a processor 210, a codec 220, a drive mechanism 230, an angle detection mechanism 240, and a liveness detection mechanism 250, wherein:

[0105] The liveness detection mechanism is used to perform liveness detection on the target environment and send the liveness detection signal to the processor; the codec is used to collect audio data and convert the audio data into a digital signal and send it to the processor; the angle detection mechanism is used to collect the elevation angle of the microphone pole and send the collected elevation angle of the microphone pole to the processor.

[0106] The processor is used to trigger the judgment of the speaking conditions based on the liveness detection signal. The method by which the processor judges the speaking conditions is based on the correlation between a preset noise threshold and data fluctuation level to determine whether the speaking conditions are met. If the speaking conditions are met, the processor triggers the drive mechanism to adjust the elevation angle of the microphone boom. During the process of the drive mechanism adjusting the microphone boom, the processor also acquires a third energy value in real time, takes the elevation angle when the third energy value is at its maximum as the optimal elevation angle, and controls the elevation angle of the microphone boom to be adjusted to the optimal elevation angle through the drive mechanism.

[0107] In a preferred embodiment of the present invention, the liveness detection mechanism can employ an infrared sensor, such as an infrared thermometer or an infrared beam transmitter, or a millimeter-wave radar. The driving mechanism can consist of a PWM stepper motor and related gears. The PWM stepper motor drives the microphone boom to adjust its elevation angle via the related gears. The angle detection mechanism can employ a six-axis gyroscope.

[0108] Example 3

[0109] Please see Figure 4 , Figure 4 This is a schematic diagram of the structure of an electronic device disclosed in an embodiment of the present invention. For example... Figure 4 As shown, the electronic device may include:

[0110] Memory 310 storing executable program code;

[0111] Processor 320 coupled to memory 310;

[0112] The processor 320 calls the executable program code stored in the memory 310 to execute some or all of the steps in the adaptive method for locating sound sources in Embodiment 1.

[0113] This invention discloses a computer-readable storage medium storing a computer program that causes a computer to perform some or all of the steps in a sound source localization adaptive method according to Embodiment 1.

[0114] This invention also discloses a computer program product, wherein when the computer program product is run on a computer, the computer performs some or all of the steps in the adaptive method for locating a sound source in Embodiment 1.

[0115] This invention also discloses an application publishing platform, which is used to publish computer program products. When the computer program products are run on a computer, the computer performs some or all of the steps in the adaptive method for locating sound sources in Embodiment 1.

[0116] In various embodiments of the present invention, it should be understood that the sequence number of each process does not necessarily imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0117] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they can be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0118] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0119] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-accessible memory. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several requests to cause a computer device (which can be a personal computer, server, or network device, specifically a processor in the computer device) to execute some or all of the steps of the methods described in the various embodiments of the present invention.

[0120] In the embodiments provided by this invention, it should be understood that "B corresponding to A" means that B is associated with A, and B can be determined based on A. However, it should also be understood that determining B based on A does not mean determining B solely based on A; B can also be determined based on A and / or other information.

[0121] Those skilled in the art will understand that some or all of the steps in the various methods of the embodiments described can be implemented by a program instructing related hardware. This program can be stored in a computer-readable storage medium, including read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically-erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, disk storage, magnetic tape storage, or any other computer-readable medium capable of carrying or storing data.

[0122] The above provides a detailed description of a sound source localization adaptive method, system, electronic device, and storage medium disclosed in the embodiments of the present invention. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. An adaptive method for locating a sound source, characterized in that, include: Whether the speaking conditions are met is determined based on the correlation between the preset noise threshold and the data fluctuation level. If the speaking conditions are met, the microphone boom tilt angle will automatically adjust. When the microphone pole elevation angle is automatically adjusted from the first elevation angle to the second elevation angle, the third audio data during the adjustment process is collected in real time, and the third audio data is converted into a third digital signal at a preset sampling rate; Obtain the root mean square of all third digital signals and record it as the third energy value corresponding to the second elevation angle; Determine the third energy value for all elevation angle positions when the microphone pole elevation angle is adjusted, and record the elevation angle position with the maximum third energy value as the optimal elevation angle; Control the microphone boom to adjust to the optimal elevation angle; The determination of whether the speaking conditions are met, based on the correlation between a preset noise threshold and data fluctuation levels, includes: Select multiple samples and obtain the first energy value of the multiple samples; The noise threshold and data fluctuation level are determined based on the first energy value; The first audio data of each sample is acquired in real time, and the first audio data is converted into a first digital signal at a preset sampling rate; Obtain the root mean square of the first digital signal for each frame and record it as the first energy value of the audio data for each frame. Determine the average and variance of multiple first energy values ​​for each sample within the collection period, and use the average of the multiple first energy values ​​as the noise threshold and the variance of the multiple first energy values ​​as the data fluctuation level; The correlation between noise threshold and data fluctuation level is formed by the correlation between noise threshold and data fluctuation level for each sample; Determine whether the speaking conditions are met based on the aforementioned relationship.

2. The adaptive method for locating sound sources according to claim 1, characterized in that, Whether the speaking conditions are met is determined based on the correlation between preset noise thresholds and data fluctuation levels, including: The system acquires second audio data from the target environment in real time and converts the second audio data into a second digital signal at a preset sampling rate. Obtain the root mean square of the second digital signal for each frame, and record it as the second energy value of the second audio data for each frame; Determine the average and variance of multiple second energy values ​​within the acquisition period, and use the average of the multiple second energy values ​​as the noise value and the variance of the multiple second energy values ​​as the data fluctuation value; A target noise threshold corresponding to the noise value is determined. When the data fluctuation value meets the target data fluctuation level, the target environment meets the speaking conditions. The target data fluctuation level is the data fluctuation level corresponding to the target noise threshold determined based on the preset noise threshold and the correlation between the data fluctuation level and the data fluctuation level.

3. The adaptive method for locating sound sources according to claim 1, characterized in that, The process of determining whether the speaking conditions are met based on the correlation between a preset noise threshold and data fluctuation levels also includes: Liveness detection is performed on the target environment, and the speech condition is triggered when a human body is present in the target environment; Or / and, the time interval for triggering automatic adjustment of the microphone boom elevation angle shall not be less than the preset duration.

4. A sound source localization adaptive system, characterized in that, It includes: The processor, codec, drive mechanism, and angle detection mechanism include: The codec is used to acquire audio data and convert the audio data into digital signals before sending them to the processor; The angle detection mechanism is used to collect the elevation angle of the microphone pole and send the collected elevation angle of the microphone pole to the processor; The processor is used to determine whether the speaking conditions are met based on a preset noise threshold and the correlation between data fluctuation levels; and if the speaking conditions are met, it triggers the drive mechanism to adjust the elevation angle of the microphone boom. When the microphone pole elevation angle is automatically adjusted from the first elevation angle to the second elevation angle, the third audio data during the adjustment process is collected in real time, and the third audio data is converted into a third digital signal at a preset sampling rate; Obtain the root mean square of all third digital signals and record it as the third energy value corresponding to the second elevation angle; Determine the third energy value for all elevation angle positions when the microphone pole elevation angle is adjusted, and record the elevation angle position with the maximum third energy value as the optimal elevation angle; Control the microphone boom to adjust to the optimal elevation angle; The determination of whether the speaking conditions are met, based on the correlation between a preset noise threshold and data fluctuation levels, includes: Select multiple samples and obtain the first energy value of the multiple samples; The noise threshold and data fluctuation level are determined based on the first energy value; The first audio data of each sample is acquired in real time, and the first audio data is converted into a first digital signal at a preset sampling rate; Obtain the root mean square of the first digital signal for each frame and record it as the first energy value of the audio data for each frame. Determine the average and variance of multiple first energy values ​​for each sample within the collection period, and use the average of the multiple first energy values ​​as the noise threshold and the variance of the multiple first energy values ​​as the data fluctuation level; The correlation between noise threshold and data fluctuation level is formed by the correlation between noise threshold and data fluctuation level for each sample; Determine whether the speaking conditions are met based on the aforementioned relationship.

5. The adaptive sound source localization system according to claim 4, characterized in that, The adaptive positioning sound source system also includes a liveness detection mechanism, which is used to perform liveness detection on the target environment and send the liveness detection signal to the processor. The processor triggers the discrimination of the speaking conditions based on the liveness detection signal. The liveness detection mechanism is an infrared detection mechanism. or / and, The driving mechanism is a stepper motor; or / and, The angle detection mechanism is a six-axis gyroscope.

6. An electronic device, characterized in that, include: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the adaptive method for locating a sound source as described in any one of claims 1 to 3.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, wherein the computer program causes a computer to perform an adaptive method for locating a sound source as described in any one of claims 1 to 3.

Images