Method and apparatus for automatic detection of conference room audio equipment

Abstract

The application provides an automatic detection method and device for conference room audio equipment, relates to the field of audio equipment operation and maintenance, and solves the technical problem of insufficient audio quality detection capability of the audio equipment in the prior art. The method comprises the following steps: controlling a loudspeaker to send different test signals in different time periods, collecting the test signals in each time period by using a second microphone to obtain a plurality of test response signals; collecting environmental noise by using a first microphone, and performing denoising processing on each test response signal based on the environmental noise, and extracting three-dimensional features of the denoised signals; calculating the comprehensive values of the same features of the plurality of test response signals, and performing weighted summation on the comprehensive spectral entropy, the comprehensive echo coherence and the comprehensive nonlinear distortion energy ratio to obtain the comprehensive feature values of different nodes; and when the comprehensive feature value of a node is greater than a corresponding preset threshold value, it is determined that the node is abnormal. The application is used in the operation and maintenance process of the audio equipment in the conference scene.

Description

Technical Field

[0001] This invention belongs to the field of audio equipment operation and maintenance, specifically an automatic detection method and device for conference room audio equipment. Background Technology

[0002] Current audio equipment testing technologies primarily rely on hardware circuits and signal interaction mechanisms to diagnose audio equipment faults by detecting physical parameters such as electrical connection status and bias voltage. However, because hardware circuits can only detect physical parameters like electrical connection status and bias voltage, they cannot detect soft faults caused by component aging or performance degradation. They have extremely low accuracy or are completely unable to identify performance degradation issues that do not affect basic electrical connections, such as dust accumulation on microphone diaphragms or aging of speaker voice coils. Furthermore, these testing methods are susceptible to interference from environmental noise, room reverberation / resonance, and changes in the relative positions of microphones and speakers. When there is air conditioning noise, wall resonance, or changes in equipment placement in a meeting room, the detection signal may be submerged by noise or experience reflection distortion, leading to misjudgments or missed detections. This makes stable operation difficult in complex environments. Summary of the Invention

[0003] This application provides an automatic detection method and apparatus for conference room audio equipment, which solves the technical problem of insufficient detection accuracy of audio equipment in the prior art.

[0004] To achieve the above objectives, this application adopts the following technical solution: Firstly, an automatic detection method for conference room audio equipment is provided, including: The speaker is controlled to send different test signals in different time periods, and the second microphone is used to collect the test signals in each time period to obtain multiple test response signals; Ambient noise is collected using a first microphone. Each test response signal is denoised based on the ambient noise, and three-dimensional features of the denoised signal are extracted. The three-dimensional features include spectral entropy, echo coherence, and nonlinear distortion energy ratio. Calculate the combined value of the same feature of multiple test response signals to obtain the combined spectral entropy, combined echo coherence, and combined nonlinear distortion energy ratio; The comprehensive spectral entropy, comprehensive echo coherence, and comprehensive nonlinear distortion energy ratio are weighted and summed to obtain the comprehensive characteristic values ​​of different nodes. The nodes include speakers, microphones, power amplifiers, and lines. Each node corresponds to a weighted summation weight value combination and a preset threshold. When the comprehensive feature value of a node is greater than the corresponding preset threshold, the node is determined to be abnormal.

[0005] Based on the above technical solutions, the automatic detection method for conference room audio equipment provided in this application comprehensively covers the working status of each link in the audio link by sending different test signals in different time periods and combining them with a dual-microphone acquisition mechanism. Utilizing environmental noise for noise reduction processing improves the purity of the test response signal and ensures the accuracy of feature extraction. Joint analysis of three-dimensional features constructs an evaluation system for equipment status based on frequency distribution complexity, signal consistency, and distortion. Weighted summation of different devices using feature composite values ​​enhances the reliability of the results through statistical regularity and enables precise location of faults in different devices such as speakers and microphones. This achieves automated detection of audio equipment and improves the efficiency and accuracy of audio system maintenance.

[0006] In conjunction with the first aspect above, in one possible implementation, the control speaker sends different test signals in time periods, including: The first signal is transmitted in the first time period, and the frequency range of the first signal is a phase-modulated signal of 30-80Hz; The second signal is sent during the second time period. The frequency range of the second signal is a linear sweep frequency signal of 17-19Hz. The third signal is transmitted in the third time period. The third signal is a broadband noise signal that has been spectrum encrypted.

[0007] In conjunction with the first aspect above, in one possible implementation, the denoising process for each test response signal based on environmental noise includes: Load the pre-stored conference room baseline acoustic fingerprint H ref The reference acoustic fingerprint represents the conference room impulse response measured when the audio equipment is functioning normally and the environment is quiet; Based on the Fast Fourier Transform (FFT) algorithm, each test signal x(t) and each test response signal y(t) are processed to obtain the transferred response value H. curr =FFT(y(t)) / FFT(x(t)); Calculate the noise energy E based on the environmental noise. noise Based on noise energy and the historical maximum noise energy E max Calculate the noise pollution factor γ = 0.8 + 0.4 × E noise / E max ; An adaptive kernel is calculated based on the baseline acoustic fingerprint, transfer function, and noise contamination factor. An adaptive kernel is used to construct a time-frequency mask, which is then applied to the short-time Fourier transform result of the test response signal. Finally, the denoised signal is obtained through the inverse short-time Fourier transform algorithm.

[0008] In conjunction with the first aspect above, in one possible implementation, the formula for calculating the adaptive kernel is: Where K represents the adaptive kernel and ⊙ represents the element-wise product.

[0009] In conjunction with the first aspect above, in one possible implementation, the construction of the time-frequency mask using an adaptive kernel includes: The environmental noise is subjected to nonnegative matrix decomposition to obtain the noise spectrum feature noise_profile; A time-frequency mask is constructed based on noise spectrum characteristics and an adaptive kernel: mask = K / (K + μ × noise_profile); where μ represents the balance factor and has a value of 0.5.

[0010] In conjunction with the first aspect above, in one possible implementation, the spectral entropy H is used to characterize the dispersion of frequency band energy distribution in the test response signal, and is calculated as follows: Where P(f) represents the normalized power spectral density at frequency f, obtained by squared-normalizing the amplitude of the fast Fourier transform result of the denoised signal. min ,f max () indicates the frequency range of the analysis.

[0011] In conjunction with the first aspect above, in one possible implementation, the echo coherence ρ is used to characterize the peak value of the cross-correlation between the multiple test signals and the test response signal, and is calculated as follows: ;in, The cross-correlation function representing the multiple test signals x(t) and the test response signal y(t) is calculated as follows: E X The energy of multiple test signals is represented by the following formula: E y The energy of the test response signal is expressed by the following formula: .

[0012] In conjunction with the first aspect above, in one possible implementation, the nonlinear distortion energy ratio δ is used to characterize the harmonic distortion rate of the test response signal, and is calculated as follows: ; where h k This represents the k-th harmonic component in the test response signal, FFT() represents the Fast Fourier Transform algorithm, || represents the modulus operation, and S(ω) represents the result of the Fourier transform on the denoised signal. min ,ω max ) indicates the signal integration range.

[0013] In conjunction with the first aspect above, in one possible implementation, calculating the comprehensive value of the same feature of multiple test response signals includes: For the spectral entropy, according to the formula H sum The comprehensive spectral entropy H is calculated as 0.2H1 + 0.3H2 + 0.5H3. sum Where H1 represents the spectral entropy of the first test response signal, H2 represents the spectral entropy of the second test response signal, and H3 represents the spectral entropy of the third test response signal; For echo coherence, the maximum value of the echo coherence of the second test response signal and the echo coherence of the third test response signal is taken as the comprehensive echo coherence. For the nonlinear distortion energy ratio, the nonlinear distortion energy ratio of the first test response signal is taken as the comprehensive nonlinear distortion energy ratio; Wherein, the first test response signal is the test response signal of the first signal, the second test response signal is the test response signal of the second signal, and the third test response signal is the test response signal of the third signal.

[0014] Secondly, an automatic detection device for conference room audio equipment is provided, comprising: a communication unit and a processing unit; The communication unit is used to establish a data connection with the first microphone, the second microphone and the speaker, collect the environmental noise signal output by the first microphone and the test response signal output by the second microphone, and send control commands for time-segmented test signals to the speaker, so as to realize the time-segmented transmission of test signals and the real-time acquisition of response signals. The processing unit is used to denoise each test response signal based on the environmental noise and extract three-dimensional features of the denoised signal, including spectral entropy, echo coherence and nonlinear distortion energy ratio. Calculate the combined value of the same feature of multiple test response signals to obtain the combined spectral entropy, combined echo coherence, and combined nonlinear distortion energy ratio; The comprehensive spectral entropy, comprehensive echo coherence, and comprehensive nonlinear distortion energy ratio are weighted and summed to obtain the comprehensive characteristic values ​​of different nodes. The nodes include speakers, microphones, power amplifiers, and lines. Each node corresponds to a weighted summation weight value combination and a preset threshold. When the comprehensive feature value of a node is greater than the corresponding preset threshold, the node is determined to be abnormal.

[0015] Thirdly, this application provides an automatic detection device for conference room audio equipment, comprising: a processor and a storage medium; the storage medium includes instructions, and the processor is configured to execute the instructions to implement the method described in the first aspect and any possible implementation thereof. The automatic detection device may be an electronic device or a chip within an electronic device.

[0016] Fourthly, this application provides a computer-readable storage medium storing instructions that, when executed on an automatic detection device for conference room audio equipment, cause the automatic detection device for conference room audio equipment to perform the methods described in the first aspect and any possible implementation thereof.

[0017] Fifthly, this application provides a computer program product containing instructions that, when run on an automatic detection device for conference room audio equipment, causes the automatic detection device for conference room audio equipment to perform the methods described in the first aspect and any possible implementation thereof.

[0018] This application provides an automatic detection method and apparatus for conference room audio equipment, which can comprehensively cover the working status of each link in the audio link through time-segmented multi-signal testing and a dual-microphone collaborative mechanism. The method utilizes a first microphone to collect ambient noise and a second microphone to collect time-segmented test response signals. Combined with a benchmark acoustic fingerprint and an adaptive denoising algorithm, it effectively eliminates environmental interference and ensures the purity of feature extraction. The joint analysis of three-dimensional features constructs a multi-dimensional evaluation system based on frequency distribution complexity, signal consistency, and distortion levels. This system can not only capture hardware circuit anomalies but also accurately identify problems that are difficult to detect using traditional methods, such as sound quality degradation and soft faults.

[0019] Meanwhile, this application employs a comprehensive feature value weighted summation mechanism to set specific weights and thresholds for different nodes such as speakers, microphones, and power amplifiers, achieving precise fault location. This detection scheme is unaffected by environmental noise, equipment position changes, or other factors, improving the automation level of audio system maintenance and fault location efficiency. It is particularly suitable for scenarios with high sound quality requirements, such as conference rooms, and can effectively ensure the stable operation of audio equipment.

[0020] It should be understood that the descriptions of technical features, technical solutions, beneficial effects, or similar language in this application do not imply that all features and advantages can be achieved in any single embodiment. Rather, it is understood that the description of a feature or beneficial effect means that a specific technical feature, technical solution, or beneficial effect is included in at least one embodiment. Therefore, the descriptions of technical features, technical solutions, or beneficial effects in this specification do not necessarily refer to the same embodiment. Furthermore, the technical features, technical solutions, and beneficial effects described in this embodiment can be combined in any suitable manner. Those skilled in the art will understand that embodiments can be implemented without one or more specific technical features, technical solutions, or beneficial effects of a particular embodiment. In other embodiments, additional technical features and beneficial effects may be identified in specific embodiments that do not embody all embodiments. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art 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.

[0022] Figure 1 A system architecture diagram of an automatic detection system for conference room audio equipment provided in this application embodiment; Figure 2 A flowchart illustrating an automatic detection method for conference room audio equipment provided in an embodiment of this application; Figure 3 A flowchart illustrating another automatic detection method for conference room audio equipment provided in this application embodiment; Figure 4 A flowchart illustrating another automatic detection method for conference room audio equipment provided in this application embodiment; Figure 5 A schematic diagram of an automatic detection device for conference room audio equipment provided in this application embodiment; Figure 6 This is a schematic diagram of the hardware structure of an automatic detection device for conference room audio equipment provided in an embodiment of this application. Detailed Implementation

[0023] In the description of this application, unless otherwise stated, " / " means "or," for example, A / B can mean A or B. The "and / or" in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, and B alone. Furthermore, "at least one" means one or more, and "multiple" means two or more. The terms "first," "second," etc., do not limit the quantity or order of execution, and "first," "second," etc., do not necessarily imply differences.

[0024] It should be noted that, in this application, the terms "exemplary" or "for example" are used to indicate that something is being described as an example, illustration, or illustration. Any embodiment or design described as "exemplary" or "for example" in this application should not be construed as being more preferred or advantageous than other embodiments or design solutions. Specifically, the use of terms such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner.

[0025] The automatic detection method for conference room audio equipment provided in this application embodiment can be applied to, for example... Figure 1 In an automatic detection system for conference room audio equipment, as shown,Figure 1 As shown, the communication system includes: a data acquisition module, a feature extraction module, and an anomaly detection module.

[0026] The data acquisition module is used to control the loudspeaker to send different test signals at different times, and to simultaneously collect environmental noise and test response signals.

[0027] The feature extraction module is used to denoise the acquired signals and extract three-dimensional feature values.

[0028] The anomaly detection module is used to calculate a comprehensive feature value based on three-dimensional feature values ​​and compare it with a preset threshold to detect anomalies in nodes such as speakers, microphones, power amplifiers, and lines.

[0029] To address the technical problem that existing hardware circuit testing methods are insufficient in detecting soft faults, performance degradation, and sound quality issues, this application provides an automatic detection method and apparatus for conference room audio equipment. The method includes: Use the first microphone to collect ambient noise; The speaker is controlled to send different test signals in different time periods, and the second microphone is used to collect the test signals in each time period to obtain multiple test response signals; The environmental noise is used to denoise each test response signal, and the three-dimensional features of the denoised signal are extracted. The three-dimensional features include spectral entropy, echo coherence, and nonlinear distortion energy ratio. Calculate the combined value of the same feature of multiple test response signals to obtain the combined spectral entropy, combined echo coherence, and combined nonlinear distortion energy ratio; The comprehensive spectral entropy, comprehensive echo coherence, and comprehensive nonlinear distortion energy ratio are weighted and summed to obtain the comprehensive characteristic values ​​of different nodes. The nodes include speakers, microphones, power amplifiers, and lines. Each node corresponds to a weighted summation weight value combination and a preset threshold. When the comprehensive feature value of a node is greater than the corresponding preset threshold, the node is determined to be abnormal.

[0030] Based on this, by employing time-segmented multi-signal testing and a dual-microphone collaborative mechanism, combined with adaptive denoising algorithms and 3D feature joint analysis, this approach can accurately detect circuit faults in hardware such as speakers and microphones, and quantitatively assess soft issues that are difficult to identify using traditional methods, such as sound quality degradation and nonlinear distortion. By setting specific weights and thresholds for different nodes, precise fault location is achieved, effectively solving the problems of incomplete fault coverage and lack of sound quality assessment in existing technologies. This significantly improves the automation level and detection efficiency of conference room audio equipment operation and maintenance, and is particularly suitable for real-time monitoring and fault diagnosis of audio systems in complex environments.

[0031] like Figure 2As shown in the embodiment of this application, an automatic detection method for conference room audio equipment includes: S1. Control the speaker to send different test signals in different time periods, and use the second microphone to collect the test signals in each time period to obtain multiple test response signals.

[0032] Among them, different test signals are used to evaluate the device's response capability across the entire frequency band, and the device is usually automatically tested half an hour before the meeting begins; the second microphone is a device microphone deployed in the sound field coverage area of ​​the conference room, used to record the on-site sound, and transmit it to the speaker end through the line for sound broadcasting and amplification.

[0033] In some implementations, the time-segmented test signals may include low-frequency vortex signals (30-80Hz), high-frequency sweep signals (17-19kHz), and pseudo-random noise, which are used to detect bass response, high-frequency extension, and full-band distortion, respectively.

[0034] It should be noted that traditional hardware circuit testing methods can detect whether the basic electrical connections of a device are normal, but they cannot identify soft faults such as sound quality degradation and nonlinear distortion. Therefore, using acoustic feature detection methods based on multi-time period test signals can cover the blind spots of hardware circuit testing and achieve quantitative evaluation of sound quality.

[0035] For example, when a 30Hz low-frequency signal is sent in the first time period, if the low-frequency attenuation is caused by the aging of the speaker voice coil, the spectral entropy of the response signal collected by the second microphone will be significantly reduced.

[0036] S2. Use the first microphone to collect ambient noise.

[0037] The first microphone is one or more high-precision, calibrated reference microphones installed at fixed locations in the conference room, used to collect ambient noise when no one is using the audio equipment in the conference room.

[0038] In some implementations, the first microphone can be deployed in unobstructed locations such as the four corners of the conference room or the ceiling, and the accuracy of the noise baseline can be improved by synchronously collecting data from multiple points.

[0039] It should be noted that environmental noise sampling should be conducted when the equipment is not turned on to avoid interference from the audio equipment's own signal to the noise reference.

[0040] For example, the noise collection program can be started 10 minutes before the meeting begins. At this time, the ambient noise comes from human voices, the operating sounds of various devices, etc. The ambient noise is collected for 5 minutes and stored as a reference noise file.

[0041] S3. Denoise each test response signal based on environmental noise and extract the three-dimensional features of the denoised signal. The three-dimensional features include spectral entropy, echo coherence, and nonlinear distortion energy ratio.

[0042] Among them, spectral entropy is used to characterize the dispersion of frequency band energy distribution in the test response signal, echo coherence is used to characterize the cross-correlation peak value between multiple test signals and the test response signal, and nonlinear distortion energy ratio is used to characterize the harmonic distortion rate of the test response signal.

[0043] In some implementations, the spectral entropy is calculated based on the fast Fourier transform of the denoised signal, the echo coherence is related to the time-domain correlation between the test signal and the response signal, and the calculation of the nonlinear distortion energy ratio requires the extraction of 2nd to 5th order harmonic components.

[0044] It should be noted that three-dimensional feature joint analysis can construct a device status assessment system from three dimensions: frequency distribution, signal consistency, and distortion level, avoiding misjudgments based on a single feature. For example, when dust accumulates on the microphone diaphragm, the nonlinear distortion energy ratio of the acquired response signal will increase, while the spectral entropy will decrease due to the loss of high-frequency components. If only one feature is used for device status assessment, the following misjudgments may occur: If the equipment malfunction is judged solely based on the increase in the nonlinear distortion energy ratio, the distortion caused by environmental noise interference may be misjudged as a microphone malfunction. Judging equipment malfunction solely by a decrease in spectral entropy might lead to misinterpreting high-frequency attenuation caused by reverberation in the conference room as a decrease in microphone sensitivity.

[0045] Similarly, as a loudspeaker voice coil ages, echo coherence decreases due to increased signal delay, while the nonlinear distortion energy ratio increases due to increased harmonic distortion. Relying solely on echo coherence for evaluation may overlook distortion issues; focusing only on the nonlinear distortion energy ratio may fail to pinpoint the root cause of signal transmission delay.

[0046] S4. Calculate the comprehensive value of the same feature of multiple test response signals, and provide anomaly warning for nodes based on the weighted summation results of different nodes and the preset threshold.

[0047] Among them, the three-dimensional feature characterizes the local state of the device under a single test signal, while the weighted summation of the comprehensive values ​​of different test response signals under the same feature can enhance the reliability of the results through statistical regularity.

[0048] It should be noted that the feature composite value is the combined value of multiple test response signals on the same feature. The composite feature value is the weighted sum of the composite values ​​of different features at the same node. The nodes include speakers, microphones, power amplifiers, and circuitry, and each node corresponds to a weighted summation combination and a preset threshold.

[0049] For example, when calculating the comprehensive characteristic value of a loudspeaker node, the weights are set as follows: spectral entropy 0.3, echo coherence 0.2, and nonlinear distortion energy ratio 0.5. If the comprehensive characteristic value is greater than 0.8, it is considered abnormal.

[0050] Furthermore, the preset thresholds differ for different nodes, and are all determined based on test results under normal conditions. For example, the preset threshold can be determined by collecting feature values ​​from 100 sets of normal devices and using three times the standard deviation as the boundary for anomaly detection.

[0051] Based on the above technical solution, the automatic detection method for conference room audio equipment provided in this application achieves comprehensive detection of hardware and software faults of audio equipment by using time-segmented multi-signal testing and dual-microphone collaborative mechanism, combined with three-dimensional feature joint analysis and weighted summation algorithm, and in particular solves the problem that traditional methods cannot assess sound quality degradation.

[0052] In one possible implementation of this application embodiment, the above-mentioned S1 can be specifically implemented by the following S101, S102 and S103, which are described in detail below: S101. Send the first signal, namely the low-frequency vortex signal, during the first time period, and use the second microphone to collect the test response signal during this time period to obtain the first test response signal.

[0053] The first signal is a phase-modulated signal with a frequency range of 30-80Hz, used to detect the device's response capability and resonance in the low-frequency band.

[0054] In some implementations, phase spiral modulation technology can be used to generate the low-frequency test signal s(t), as shown in the formula: Where A is the amplitude, set according to the maximum allowable volume of the device, and β is the attenuation factor, which is adaptively adjusted according to the ambient noise level. That is, the greater the ambient noise, the greater β, so as to reduce the impact of signal amplitude on the conference. The ambient noise level can be judged by the ambient noise energy. α represents the phase modulation parameter, which can be set to α = 0.02 × t 2 That is, as the square of time increases, the aroma rotates faster and faster, disrupting the coherence of the noise; f c This indicates a center frequency of 50Hz.

[0055] It should be noted that traditional detection methods have difficulty quantifying low-frequency distortion, while the 30-80Hz frequency band covers common low-frequency resonance points in conference rooms (such as the coupling frequency of table and chair vibrations). Therefore, this signal can effectively identify low-frequency attenuation caused by the aging of speaker suspension edges.

[0056] For example, when the speaker suspension is hardened, the spectral entropy of the acquired response signal at 50 Hz will be significantly reduced, and the echo coherence will decrease due to the increased phase delay.

[0057] S102. In the second time period, a second signal, namely a high-frequency sweep signal, is sent. The test response signal of this time period is collected using the second microphone to obtain the second test response signal.

[0058] The second signal is a linear sweep signal with a frequency range of 17-19kHz, used to evaluate the device's high-frequency extension and the sensitivity of the microphone diaphragm.

[0059] In some implementations, the high-frequency sweep signal s H The formula for generating (t) is: Among them, T s Indicates the frequency sweep duration.

[0060] The sweep rate can be set to 0.5kHz / ms, covering the high-frequency limit audible to the ultrasonic transition region, and testing the microphone's ability to capture high-frequency details.

[0061] It should be noted that the 17-19kHz frequency band is susceptible to environmental electromagnetic interference (such as fluorescent lamp noise), so it is necessary to combine it with an adaptive denoising algorithm to eliminate interference and avoid misjudgment.

[0062] S103. The third signal, namely pseudo-random noise, is sent during the third time period. The test response signal of this time period is collected by the second microphone to obtain the third test response signal.

[0063] The third signal is a broadband noise signal that has been spectrum encrypted, covering the entire frequency band from 20Hz to 20kHz, and is used to detect the overall distortion of the equipment under complex signals.

[0064] In some implementations, pseudo-random noise s N The expression generated by (t) is: Where IDFT{·} denotes the inverse discrete Fourier transform, random(f) denotes the frequency domain amplitude modulation factor, and f is a frequency variable, indicating that at different frequency points f, the signal is modulated with randomly generated amplitude values, thus randomizing the intensity of each frequency component. Represents the frequency domain phase modulation factor. The phase function corresponding to frequency f can be expressed in negative form. The phase of each frequency component is modulated to increase the randomness and encryption characteristics of the spectrum.

[0065] It should be noted that the bandwidth and energy distribution of pseudo-random noise can simulate the superposition of multiple sound sources in a real meeting scenario, which is closer to the actual application scenario than a single frequency signal.

[0066] It should be noted that in step S3, if there are multiple speakers, the audio processor can control the power amplifier to send a sequence of test signals to each speaker independently via each channel at a certain time interval. The waveforms of the original test signals need to be recorded synchronously as a reference for subsequent analysis. Based on the above technical solution, the design of multi-type test signals in different time periods can fully cover the full-band response characteristics of audio equipment. Combined with the dual-microphone acquisition mechanism, it can capture hardware faults such as low-frequency resonance and high-frequency attenuation, as well as identify soft problems such as power amplifier distortion and microphone sensitivity reduction, thereby improving the accuracy and comprehensiveness of audio equipment testing.

[0067] In one possible implementation of the embodiments of this application, combined with Figure 2 ,like Figure 3 As shown, the above S3 can be implemented through the following S301, S302 and S303, which are explained in detail below: S301. Load the pre-stored conference room reference acoustic fingerprint and calculate the transmission response value based on the Fast Fourier Transform.

[0068] Among them, the reference acoustic fingerprint H ref This represents the impulse response of the conference room measured when the audio equipment is functioning normally and the environment is quiet. It is used to construct an ideal acoustic model, and its calculation formula is: H ref =FFT(y ref (t)) / FFT(x ref (t)); where y ref (t) represents the reference test response signal collected by the second microphone when the audio device is functioning normally and the environment is quiet (meaning there is no one around and all devices except the audio device are turned off). ref (t) represents the reference test signal sent by the speaker when the audio equipment is working properly and the environment is quiet. The formula for calculating the transferred response value is: H curr =FFT(y(t)) / FFT(x(t)); reflects the frequency domain relationship between the current test signal x(t) and the response signal y(t).

[0069] In some implementations, the reference acoustic fingerprint can be generated by averaging the impulse responses collected from more than 10 groups under normal and quiet conditions, and the number of FFT transform points can be set to 4096 to balance computational accuracy and efficiency.

[0070] It should be noted that the decoration materials of the conference room (such as sound-absorbing panels and glass) can affect the uniqueness of the baseline acoustic fingerprint, so the baseline data needs to be collected again after the decoration is changed.

[0071] S302. Calculate the noise pollution factor and generate an adaptive kernel based on the baseline acoustic fingerprint and transfer function.

[0072] Wherein, the noise pollution factor γ = 0.8 + 0.4 × E noise / E max Through the current noise energy (x) noise (n) represents the ambient noise collected by the first microphone, and n represents the sequence index. The maximum historical noise energy E is then compared with... max The ratio is dynamically adjusted; Adaptive kernel It is used to quantify the degree of environmental noise pollution on the current signal.

[0073] In some implementations, E max The peak noise energy from the past 30 days of detection can be used, and the γ range is limited to 0.8-1.2 to avoid algorithm failure caused by extreme noise.

[0074] S303. Use an adaptive kernel and noise spectrum features to construct a time-frequency mask to denoise the test response signal.

[0075] The ambient noise spectral profile (noise_profile) is extracted using nonnegative matrix factorization (NMF). A time-frequency mask (mask = K / (K + μ × noise_profile)) is constructed using an adaptive kernel K (μ is set to 0.5). The denoised signal y is then obtained through inverse short-time Fourier transform (iFFT). clean (t), the formula is: y clean (t)=iFFT[mask×FFT((t)))].

[0076] In some implementations, the rank of the NMF decomposition can be set to 8-12 to capture the main spectral components of ambient noise; the time-frequency mask update frequency is synchronized with the test signal frame rate (e.g., 50ms / time).

[0077] It should be noted that this masking mechanism is effective in suppressing periodic noise (such as power frequency interference), but requires additional processing in conjunction with time-domain energy detection for sudden impulse noise (such as the sound of a door opening).

[0078] For example, for 50Hz power frequency noise, NMF can extract its spectral characteristics, and the weight of the time-frequency mask at 50Hz and its harmonics is reduced to below 0.3, effectively eliminating current noise interference.

[0079] S304 extracts the three-dimensional features of the denoised signal.

[0080] Among them, the three-dimensional features include spectral entropy, echo coherence, and nonlinear distortion energy ratio.

[0081] Specifically, the spectral entropy H is used to characterize the dispersion of frequency band energy distribution in the test response signal, and its calculation formula is: Where P(f) represents the normalized power spectral density at frequency f, obtained by squared-normalizing the amplitude of the fast Fourier transform result of the denoised signal. min ,f max () indicates the frequency range of the analysis; Echo coherence ρ is used to characterize the peak value of the cross-correlation between multiple test signals and the test response signal, and is calculated as follows: ;in, The cross-correlation function representing the multiple test signals x(t) and the test response signal y(t) is calculated as follows: E X The energy of multiple test signals is represented by the following formula: E y The energy of the test response signal is expressed by the following formula: ; The nonlinear distortion energy ratio δ is used to quantify the 2nd to 5th order harmonic distortion in the test response signal. The calculation formula is: ; where h k This represents the k-th harmonic component in the test response signal, FFT() represents the Fast Fourier Transform algorithm, || represents the modulus operation, and S(ω) represents the result of the Fourier transform on the denoised signal. min ,ω max ) indicates the signal integration range.

[0082] Based on the above technical solution, this denoising process achieves precise suppression of environmental noise and room reverberation through dynamic matching of a reference acoustic fingerprint and an adaptive kernel. Compared with traditional bandpass filtering methods, it can retain more high-frequency details of the effective signal. Simultaneously, the combination of time-frequency masking and NMF allows the denoising process to adapt to the acoustic characteristics of different conference rooms, ensuring the accuracy of subsequently extracted spectral entropy, echo coherence, and other three-dimensional features. This avoids misjudgments of equipment status due to noise interference and improves the environmental robustness of the detection system.

[0083] In one possible implementation, combining Figure 2 ,like Figure 4 Specifically, S4 may include: S401, calculate the feature composite value of each feature.

[0084] (1) Calculate the comprehensive spectrum entropy: according to the formula H sum The comprehensive spectral entropy H is calculated as 0.2H1 + 0.3H2 + 0.5H3. sum Where H1 represents the spectral entropy of the first test response signal, H2 represents the spectral entropy of the second test response signal, and H3 represents the spectral entropy of the third test response signal; (2) Calculate the overall echo coherence: Take the maximum value of the echo coherence of the second test response signal and the echo coherence of the third test response signal as the overall echo coherence; (3) Calculate the comprehensive nonlinear distortion energy ratio: take the nonlinear distortion energy ratio of the first test response signal as the comprehensive nonlinear distortion energy ratio.

[0085] It should be noted that in the formula for calculating the comprehensive spectral entropy, the first test response signal has the lowest weight at 0.2 because low frequencies mainly affect the "thickness" of the sound, while the human ear is relatively less sensitive to low-frequency distortion. The second test response signal has a medium weight at 0.3 because high frequencies determine the "clarity" of the sound, but high frequencies are easily affected by environmental noise, so the weight should not be too high. The third test response signal has the highest weight at 0.5 because it covers the entire 20Hz-20kHz frequency band, comprehensively reflecting the integrity of the signal energy distribution and providing a more comprehensive assessment of sound quality. Weighted summation can enhance the fusion of full-frequency characteristics. For example, increasing the weight of high-frequency signals can more sensitively detect high-frequency attenuation faults caused by dust accumulation on the microphone diaphragm or aging of the speaker's high-frequency unit, avoiding blind spots in the detection of only low-frequency signals.

[0086] In the calculation of overall echo coherence, the first test response signal is easily affected by the resonance of the conference room structure, resulting in environmental reflection noise being mixed into the echo coherence calculation, which cannot accurately reflect the signal transmission consistency of the equipment itself. The second test response signal can directly test the equipment's response capability to high-frequency details. High-frequency signals have short wavelengths and are less affected by environmental reverberation, and their coherence can better reflect the time-domain synchronization of the microphone and speaker. The third test response signal covers the entire frequency band, and its broadband characteristics can average out local frequency band interference. Taking the maximum coherence value can ensure that at least one frequency band can truly reflect the equipment status and avoid misjudgment caused by low-frequency resonance.

[0087] In the calculation of the comprehensive nonlinear distortion energy ratio, the second test response signal has lower energy, and the harmonic components of the nonlinear distortion are easily affected by electromagnetic interference, leading to unstable detection results. The broadband characteristics of the third test response signal cause the distortion energy to be dispersed across various frequency bands, making it difficult to quantify the impact of a single hardware fault. In contrast, in the first test response signal, the nonlinear distortion energy is more concentrated in the low-frequency band, and the 2nd to 5th order harmonics caused by hardware faults are easier to detect. Therefore, the nonlinear distortion energy ratio of the first test response signal is selected as the final comprehensive value.

[0088] S402 provides early warnings of node anomalies based on different node weight combinations and preset thresholds.

[0089] The weight combinations and preset thresholds for different nodes are shown in Table 1: Table 1. Node Feature Sensitivity Weight Table

[0090] Based on the above weight combination and weighted summation formula: F=w H ×H+w ρ ×ρ+w δ ×δ is used to calculate the comprehensive feature value F of different nodes, and compare it with the feature threshold. If it exceeds the feature threshold, it indicates that the equipment corresponding to the node is abnormal and needs to be manually inspected.

[0091] Based on the above technical solution, different weight combinations and feature thresholds are defined according to the hardware characteristics and fault sensitivity dimensions of different nodes. This not only highlights the core fault characteristics of each node, but also balances the false detection rate and false negative rate through dynamic adjustment of weights, thus achieving accurate adaptation of the detection logic.

[0092] The above primarily describes the solutions of the embodiments of this application from the perspective of device implementation. It is understood that each device, such as an automatic detection device for conference room audio equipment, includes at least one of the hardware structure and software module corresponding to each function in order to achieve the above-mentioned functions. Those skilled in the art should readily recognize that, based on the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein, this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is implemented in hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0093] This application embodiment can divide the automatic detection device for conference room audio equipment into functional units based on the above method example. For example, each function can be divided into separate functional units, or two or more functions can be integrated into one processing unit. The integrated unit can be implemented in hardware or as a software functional unit. It should be noted that the unit division in this application embodiment is illustrative and only represents one logical functional division; other division methods may be used in actual implementation.

[0094] When using integrated units, Figure 5 A possible structural schematic diagram of the automatic detection device for conference room audio equipment (referred to as the automatic detection device 50 for conference room audio equipment) involved in the above embodiments is shown. The automatic detection device 50 for conference room audio equipment includes a processing unit 501 and a communication unit 502, and may also include a storage unit 503. Figure 5 The structural diagram shown can be used to illustrate the structure of the automatic detection device for conference room audio equipment involved in the above embodiments.

[0095] when Figure 5 The schematic diagram shown illustrates the structure of the automatic detection device for conference room audio equipment involved in the above embodiments. The processing unit 501 is used to control and manage the operation of the automatic detection device for conference room audio equipment, the communication unit 502 is used for the automatic detection device for conference room audio equipment to communicate with other devices, and the storage unit 503 is used to store the program code and data of the automatic detection device for conference room audio equipment.

[0096] For example, the communication unit 502 is used to establish data connections with the first microphone, the second microphone and the speaker, send low-frequency vortex signals, high-frequency sweep signals and pseudo-random noise in time periods according to control commands, and collect environmental noise signals and test response signals in real time; the processing unit 501 is used to perform noise reduction processing on the test response signals based on environmental noise, extract spectral entropy, echo coherence and nonlinear distortion energy ratio, calculate comprehensive feature values ​​and compare them with preset thresholds to determine whether the node is abnormal.

[0097] In one possible implementation, the processing unit 501 is also used to load a pre-stored conference room reference acoustic fingerprint, calculate the transmitted response value through FFT, generate an adaptive kernel by combining noise contamination factors, and construct a time-frequency mask to achieve denoising processing of the test response signal.

[0098] In one possible implementation, the communication unit 502 is also used to receive detection parameters (such as the test signal frequency range and detection period) input by the user, and the processing unit 501 is also used to dynamically update the reference acoustic fingerprint according to the changes in the conference room decoration materials, optimize the parameter configuration of the adaptive denoising algorithm, and improve the detection robustness in complex environments.

[0099] The processing unit 501 can be a processor or a controller, and the communication unit 502 can be a communication interface, transceiver, transceiver circuit, transceiver device, etc. The term "communication interface" is a general term and may include one or more interfaces. The storage unit 503 can be a memory. When the automatic detection device 50 for the conference room audio equipment is a chip, the processing unit 501 can be a processor or a controller, and the communication unit 502 can be an input interface and / or an output interface, pins, or circuits, etc. The storage unit 503 can be a storage unit within the chip (e.g., a register, cache, etc.) or a storage unit located outside the chip (e.g., read-only memory (ROM), random access memory (RAM, etc.)).

[0100] The communication unit can also be called a transceiver unit. The antenna and control circuit with transceiver functions in the automatic detection device 50 for conference room audio equipment can be considered as the communication unit 502 of the automatic detection device 50 for conference room audio equipment, and the processor with processing functions can be considered as the processing unit 501 of the automatic detection device 50 for conference room audio equipment. Optionally, the device in the communication unit 502 used to implement the receiving function can be considered as the communication unit, which is used to execute the receiving steps in the embodiments of this application. The communication unit can be a receiver, a receiver circuit, etc. The device in the communication unit 502 used to implement the transmitting function can be considered as the transmitting unit, which is used to execute the transmitting steps in the embodiments of this application. The transmitting unit can be a transmitter, a transmitter, a transmitting circuit, etc.

[0101] Figure 5 If the integrated units in the process are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of this application, in essence, or the parts that contribute to the prior art, or all or part of the technical solutions, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute all or part of the steps of the methods described in the various embodiments of this application. Storage media for storing computer software products include various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory, random access memory, magnetic disks, or optical disks.

[0102] Figure 5 The units in the process can also be called modules; for example, a processing unit can be called a processing module.

[0103] This application also provides an embodiment of an automatic detection method and apparatus for conference room audio equipment. A schematic diagram of the hardware structure of the apparatus (referred to as the automatic detection apparatus 60 for conference room audio equipment) is shown below. Figure 6 The automatic detection device 60 for the conference room audio equipment includes a processor 601, and optionally, a memory 602 connected to the processor 601.

[0104] In the first possible implementation, see Figure 6The automatic detection device 60 for conference room audio equipment also includes a transceiver 603. The processor 601, memory 602, and transceiver 603 are connected via a bus. The transceiver 603 is used to communicate with other devices or communication networks. Optionally, the transceiver 603 may include a transmitter and a receiver. The device in the transceiver 603 that implements the receiving function can be considered as a receiver, which is used to perform the receiving steps in the embodiments of this application. The device in the transceiver 603 that implements the transmitting function can be considered as a transmitter, which is used to perform the transmitting steps in the embodiments of this application.

[0105] Based on the first possible implementation method Figure 6 The structural diagram shown can be used to illustrate the structure of the automatic detection device for conference room audio equipment involved in the above embodiments.

[0106] in, Figure 6 This can also be illustrated by the system chip in the automatic detection device for conference room audio equipment. In this case, the actions performed by the aforementioned automatic detection device for conference room audio equipment can be implemented by this system chip. The specific actions performed can be found above and will not be repeated here.

[0107] In implementation, each step of the method provided in this embodiment can be completed by integrated logic circuits in the processor or by instructions in software form. The steps of the method disclosed in the embodiments of this application can be directly manifested as being executed by a hardware processor, or being executed by a combination of hardware and software modules in the processor.

[0108] The processor in this application may include, but is not limited to, at least one of the following: a central processing unit (CPU), a microprocessor, a digital signal processor (DSP), a microcontroller unit (MCU), or an artificial intelligence processor, etc., which are various computing devices that run software. Each computing device may include one or more cores for executing software instructions to perform calculations or processing. The processor may be a separate semiconductor chip or integrated with other circuits into a single semiconductor chip. For example, it may be integrated with other circuits (such as encoding / decoding circuits, hardware acceleration circuits, or various bus and interface circuits) to form a SoC (System-on-a-Chip), or it may be integrated as a built-in processor within an ASIC. The ASIC with the integrated processor may be packaged separately or together with other circuits. In addition to the cores for executing software instructions to perform calculations or processing, the processor may further include necessary hardware accelerators, such as field-programmable gate arrays (FPGAs), PLDs (programmable logic devices), or logic circuits that implement dedicated logic operations.

[0109] The memory in the embodiments of this application may include at least one of the following types: read-only memory (ROM) or other types of static storage devices capable of storing static information and instructions; random access memory (RAM) or other types of dynamic storage devices capable of storing information and instructions; or electrically erasable programmable-only memory (EEPROM). In some scenarios, the memory may also be a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media, or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures that can be accessed by a computer, but is not limited thereto.

[0110] This application also provides a computer-readable storage medium including instructions that, when run on a computer, cause the computer to perform any of the methods described above.

[0111] This application also provides a computer program product containing instructions that, when run on a computer, cause the computer to perform any of the methods described above.

[0112] This application also provides a chip including a processor and an interface circuit. The interface circuit is coupled to the processor. The processor is used to run computer programs or instructions to implement the above-described method. The interface circuit is used to communicate with other modules outside the chip.

[0113] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented using software programs, implementation can be, in whole or in part, in the form of a computer program product. This computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device containing one or more servers, data centers, etc., that can be integrated with the medium. The available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid-state disks (SSDs)).

[0114] Although this application has been described herein in conjunction with various embodiments, those skilled in the art, by reviewing the accompanying drawings, disclosure, and appended claims, will understand and implement other variations of the disclosed embodiments in carrying out the claimed application. In the claims, the word "comprising" does not exclude other components or steps, and "a" or "an" does not exclude multiple instances. A single processor or other unit can implement several functions listed in the claims. While different dependent claims may recite certain measures, this does not mean that these measures cannot be combined to produce good results.

[0115] Although this application has been described in conjunction with specific features and embodiments, it is obvious that various modifications and combinations can be made thereto without departing from the spirit and scope of this application. Accordingly, this specification and drawings are merely exemplary illustrations of this application as defined by the appended claims, and are considered to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from the spirit and scope of this application. Thus, if such modifications and modifications of this application fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and modifications.

Claims

1. An automatic detection method for conference room audio equipment, characterized in that, include: The speaker is controlled to send different test signals in different time periods, and the second microphone is used to collect the test signals in each time period to obtain multiple test response signals; Ambient noise is collected using a first microphone. Each test response signal is denoised based on the ambient noise, and three-dimensional features of the denoised signal are extracted. The three-dimensional features include spectral entropy, echo coherence, and nonlinear distortion energy ratio. Calculate the combined value of the same feature of multiple test response signals to obtain the combined spectral entropy, combined echo coherence, and combined nonlinear distortion energy ratio; The comprehensive spectral entropy, comprehensive echo coherence, and comprehensive nonlinear distortion energy ratio are weighted and summed to obtain the comprehensive characteristic values ​​of different nodes. The nodes include speakers, microphones, power amplifiers, and lines. Each node corresponds to a weighted summation weight value combination and a preset threshold. When the comprehensive feature value of a node is greater than the corresponding preset threshold, the node is determined to be abnormal.

2. The automatic detection method for conference room audio equipment according to claim 1, characterized in that, The noise reduction process for each test response signal based on environmental noise includes: Load the pre-stored conference room baseline acoustic fingerprint H ref The reference acoustic fingerprint represents the conference room impulse response measured when the audio equipment is functioning normally and the environment is quiet; Based on the Fast Fourier Transform (FFT) algorithm, each test signal x(t) and each test response signal y(t) are processed to obtain the transferred response value H. curr =FFT(y(t)) / FFT(x(t)); Calculate the noise energy E based on the environmental noise. noise Based on noise energy and the historical maximum noise energy E max Calculate the noise pollution factor γ = 0.8 + 0.4 × E noise / E max ; An adaptive kernel is calculated based on the baseline acoustic fingerprint, transfer function, and noise contamination factor. An adaptive kernel is used to construct a time-frequency mask, which is then applied to the short-time Fourier transform result of the test response signal. Finally, the denoised signal is obtained through the inverse short-time Fourier transform algorithm.

3. The automatic detection method for conference room audio equipment according to claim 2, characterized in that, The formula for calculating the adaptive kernel is: Where K represents the adaptive kernel and ⊙ represents the element-wise product.

4. The automatic detection method for conference room audio equipment according to claim 2, characterized in that, The method of constructing a time-frequency mask using an adaptive kernel includes: The environmental noise is subjected to nonnegative matrix decomposition to obtain the noise spectrum feature noise_profile; A time-frequency mask is constructed based on noise spectrum characteristics and an adaptive kernel: mask = K / (K + μ × noise_profile); where μ represents the balance factor and has a value of 0.

5.

5. The automatic detection method for conference room audio equipment according to claim 1, characterized in that, The spectral entropy H is used to characterize the dispersion of frequency band energy distribution in the test response signal, and is calculated as follows: Where P(f) represents the normalized power spectral density at frequency f, obtained by squared-normalizing the amplitude of the fast Fourier transform result of the denoised signal. min ,f max () indicates the frequency range of the analysis.

6. The automatic detection method for conference room audio equipment according to claim 1, characterized in that, The echo coherence ρ is used to characterize the peak value of the cross-correlation between the multiple test signals and the test response signal, and is calculated as follows: ;in, The cross-correlation function representing the multiple test signals x(t) and the test response signal y(t) is calculated as follows: E X The energy of multiple test signals is represented by the following formula: E y The energy of the test response signal is expressed by the following formula: .

7. The automatic detection method for conference room audio equipment according to claim 1, characterized in that, The nonlinear distortion energy ratio δ is used to characterize the harmonic distortion rate of the test response signal, and is calculated as follows: ; where h k This represents the k-th harmonic component in the test response signal, FFT() represents the Fast Fourier Transform algorithm, || represents the modulus operation, and S(ω) represents the result of the Fourier transform on the denoised signal. min ,ω max ) indicates the signal integration range.

8. The automatic detection method for conference room audio equipment according to claim 1, characterized in that, The control speaker sends different test signals at different times, including: The first signal is transmitted in the first time period, and the frequency range of the first signal is a phase-modulated signal of 30-80Hz; The second signal is sent during the second time period. The frequency range of the second signal is a linear sweep frequency signal of 17-19Hz. The third signal is transmitted in the third time period. The third signal is a broadband noise signal that has been spectrum encrypted.

9. The automatic detection method and apparatus for conference room audio equipment according to claim 8, characterized in that, The calculation of the combined value of the same feature of multiple test response signals includes: For the spectral entropy, according to the formula H sum The comprehensive spectral entropy H is calculated as 0.2H1 + 0.3H2 + 0.5H3. sum Where H1 represents the spectral entropy of the first test response signal, H2 represents the spectral entropy of the second test response signal, and H3 represents the spectral entropy of the third test response signal; For echo coherence, the maximum value of the echo coherence of the second test response signal and the echo coherence of the third test response signal is taken as the comprehensive echo coherence. For the nonlinear distortion energy ratio, the nonlinear distortion energy ratio of the first test response signal is taken as the comprehensive nonlinear distortion energy ratio; Wherein, the first test response signal is the test response signal of the first signal, the second test response signal is the test response signal of the second signal, and the third test response signal is the test response signal of the third signal.

10. An automatic detection device for conference room audio equipment, comprising: Communication unit and processing unit; The communication unit is used to establish a data connection with the first microphone, the second microphone and the speaker, collect the environmental noise signal output by the first microphone and the test response signal output by the second microphone, and send control commands for time-segmented test signals to the speaker, so as to realize the time-segmented transmission of test signals and the real-time acquisition of response signals. The processing unit is used to denoise each test response signal based on the environmental noise and extract three-dimensional features of the denoised signal, including spectral entropy, echo coherence and nonlinear distortion energy ratio. Calculate the combined value of the same feature of multiple test response signals to obtain the combined spectral entropy, combined echo coherence, and combined nonlinear distortion energy ratio; The comprehensive spectral entropy, comprehensive echo coherence, and comprehensive nonlinear distortion energy ratio are weighted and summed to obtain the comprehensive characteristic values ​​of different nodes. The nodes include speakers, microphones, power amplifiers, and lines. Each node corresponds to a weighted summation weight value combination and a preset threshold. When the comprehensive feature value of a node is greater than the corresponding preset threshold, the node is determined to be abnormal.

Images