Audio system inspection method, device and equipment for video conference, medium and product

Abstract

The application discloses a kind of video conference audio system inspection method, device, equipment, medium and product, the method includes determining the audio analysis reference information of audio system according to inspection configuration information;According to audio analysis reference information, time domain analysis and frequency domain transformation are carried out to audio signal data, and time-frequency data is obtained;According to audio analysis reference information, audio quality analysis is carried out to time-frequency data, and audio quality analysis result is obtained;According to audio quality analysis result and audio analysis reference information, audio quality deviation value is determined;If audio quality deviation value is not in the quality threshold range of pre-set, then trigger equipment abnormal positioning instruction;According to equipment abnormal positioning instruction and the real-time audio feature data of each audio channel, the inspection report of audio system is generated. It can improve the accuracy and efficiency of fault detection, solve the problem of low efficiency of existing audio system inspection method debugging, so as to meet the maintenance needs of modern video conference audio system.

Description

Technical Field

[0001] This invention relates to the field of audio system technology, and in particular to a method, apparatus, equipment, medium and product for inspecting audio systems in video conferencing. Background Technology

[0002] With the rapid development of remote work and online education, video conferencing systems have become an important communication tool for modern enterprises and educational institutions. Modern video conferencing systems typically include various audio devices, such as microphones, speakers, and audio processors. The increasingly complex connections between these devices pose significant challenges to system maintenance and troubleshooting.

[0003] Currently, the main method for inspecting audio systems in video conferencing relies on manual inspection of each audio device, checking their operation by manually testing audio inputs and outputs and adjusting device parameters. This method has the following problems: First, it is extremely labor-intensive and time-consuming, making it difficult to meet the efficient maintenance needs of complex systems; second, it cannot accurately identify the specific type and location of connection anomalies, resulting in low debugging efficiency; and third, it is difficult to quickly locate the source of the fault in multi-device audio systems, affecting the stability of the audio system. Therefore, the existing manual inspection method cannot meet the maintenance needs of modern video conferencing audio systems. Summary of the Invention

[0004] This invention provides a method, apparatus, equipment, medium, and product for inspecting the audio system of a video conferencing system. By performing specialized analysis on the collected audio signals and comparing the analysis results with an established audio analysis benchmark, faulty equipment identification parameter information is obtained, thereby generating an audio system inspection report. This solves the problem of low debugging efficiency in existing audio system inspection methods and meets the maintenance needs of modern video conferencing audio systems.

[0005] To achieve the above objectives, embodiments of the present invention provide a method for inspecting the audio system of a video conferencing system, comprising: Acquire the inspection configuration information and audio signal data of the audio system during video conferencing; Based on the inspection configuration information, determine the audio analysis reference information of the audio system; perform time-domain analysis and frequency-domain transformation on the audio signal data based on the audio analysis reference information to obtain the time-frequency data of the audio signal data; Based on the audio analysis benchmark information, audio quality analysis is performed on the time-frequency data to obtain the audio quality analysis results of the audio signal data; Based on the audio quality analysis results and the audio analysis benchmark information, the audio quality deviation value of the audio signal data is determined; if the audio quality deviation value is not within the preset quality threshold range, the device abnormality location command of the audio system is triggered. Based on the device anomaly location command and the real-time audio characteristic data of each audio channel in the audio system, an inspection report of the audio system is generated.

[0006] As an improvement to the above scheme, the step of performing audio quality analysis on the time-frequency data based on the audio analysis benchmark information to obtain the audio quality analysis result of the audio signal data includes: Based on the signal-to-noise ratio (SNR) benchmark value in the audio analysis benchmark information, a preset SNR separation algorithm is used to perform SNR separation processing on the time-frequency data to obtain the SNR separation result of the audio signal data; Based on the frequency domain transformation parameters in the audio analysis reference information, a preset howling recognition algorithm is used to perform howling recognition processing on the time-frequency data to obtain the howling recognition result of the audio signal data; Based on the echo suppression benchmark value in the audio analysis benchmark information, a preset echo perception algorithm is used to perform echo perception processing on the time-frequency data to obtain the echo detection result of the audio signal data; The signal-to-noise separation result, the howling recognition result, and the echo detection result are combined to obtain the audio quality analysis result of the audio signal data.

[0007] As an improvement to the above solution, the step of generating an inspection report for the audio system based on the device anomaly location command and the real-time audio characteristic data of each audio channel in the audio system includes: According to the device anomaly location command, real-time audio feature data of each audio channel in the audio system is obtained; based on the real-time audio feature data and the audio analysis benchmark information, audio feature deviation information of each audio channel is determined. Based on the audio feature deviation information, each audio channel is compared and analyzed to obtain the channel comparison analysis result information of the audio system; Based on the channel comparison analysis results and the audio feature deviation information, the current abnormal state information of the audio system is obtained; Based on the current abnormal state information and the real-time audio feature data, obtain the fault device identification parameter information of the audio system; An inspection report for the audio system is generated based on the faulty equipment identification parameter information.

[0008] As an improvement to the above solution, the step of obtaining the current abnormal state information of the audio system based on the channel comparison analysis results and the audio feature deviation information includes: Based on the channel comparison analysis results, the timestamp sequence information of each abnormal channel in the audio system is obtained; Based on the timestamp sequence information, the abnormal propagation path of each abnormal channel is traced to obtain the path tracing result information of each abnormal channel; Based on the path tracing results and the audio feature deviation information, the primary fault channel identification information of the audio system is obtained; Based on the original fault channel identification information, the fault source is determined from the abnormal channels of the audio system to obtain the current abnormal state information of the audio system.

[0009] As an improvement to the above solution, the step of determining the audio analysis benchmark information of the audio system based on the inspection configuration information includes: Based on the inspection mode of the inspection configuration information, determine the time-domain analysis window parameters and frequency-domain transformation parameters of the audio system; Based on the audio quality level of the inspection configuration information, determine the signal-to-noise ratio reference value and echo suppression reference value of the audio system; By correlating the time-domain analysis window parameters, the frequency-domain transformation parameters, the signal-to-noise ratio reference value, and the echo suppression reference value, the audio analysis reference information of the audio system is obtained.

[0010] As an improvement to the above solution, the step of determining the time-domain analysis window parameters and frequency-domain transformation parameters of the audio system based on the inspection mode of the inspection configuration information includes: Based on the inspection mode of the inspection configuration information and the preset mode parameter mapping table, the window length and window overlap rate of the audio system are obtained; The temporal analysis window parameters of the audio system are determined based on the window length and window overlap rate. Based on the window length and preset frequency domain analysis rules, the FFT transform length and frequency resolution of the audio system are obtained; The frequency domain transformation parameters of the audio system are determined based on the FFT transform length and frequency resolution.

[0011] To achieve the above objectives, embodiments of the present invention provide an audio system inspection device for video conferencing, comprising: The inspection data acquisition module is used to acquire the inspection configuration information and audio signal data of the audio system in video conferencing; The time-frequency data acquisition module is used to determine the audio analysis reference information of the audio system based on the inspection configuration information; and to perform time-domain analysis and frequency-domain transformation on the audio signal data based on the audio analysis reference information to obtain the time-frequency data of the audio signal data. The audio quality analysis module is used to perform audio quality analysis on the time-frequency data based on the audio analysis benchmark information to obtain the audio quality analysis results of the audio signal data. An abnormal instruction triggering module is used to determine the audio quality deviation value of the audio signal data based on the audio quality analysis results and the audio analysis benchmark information; if the audio quality deviation value is not within the preset quality threshold range, the device abnormality location instruction of the audio system is triggered. The inspection report generation module is used to generate an inspection report for the audio system based on the device anomaly location command and the real-time audio characteristic data of each audio channel in the audio system.

[0012] To achieve the above objectives, this invention provides an audio system inspection device for video conferencing, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the aforementioned audio system inspection method for video conferencing.

[0013] To achieve the above objectives, embodiments of the present invention also provide a computer-readable storage medium, the computer-readable storage medium including a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to perform the above-mentioned audio system inspection method for video conferencing.

[0014] To achieve the above objectives, embodiments of the present invention also provide a computer program product, which is stored in a storage medium and executed by at least one processor to implement the steps of the above-described audio system inspection method for video conferencing.

[0015] Compared with existing technologies, the present invention discloses a method, apparatus, device, medium, and product for inspecting the audio system of a video conferencing system. This involves: acquiring inspection configuration information and audio signal data of the audio system in the video conferencing; determining audio analysis benchmark information of the audio system based on the inspection configuration information; performing time-domain analysis and frequency-domain transformation on the audio signal data based on the audio analysis benchmark information to obtain time-frequency data of the audio signal data; performing audio quality analysis on the time-frequency data based on the audio analysis benchmark information to obtain audio quality analysis results of the audio signal data; determining the audio quality deviation value of the audio signal data based on the audio quality analysis results and the audio analysis benchmark information; triggering a device anomaly location command for the audio system if the audio quality deviation value is not within a preset quality threshold range; and generating an inspection report for the audio system based on the device anomaly location command and real-time audio characteristic data of each audio channel in the audio system. By comparing the analysis results of audio signal data with the established audio analysis benchmark, faulty equipment identification parameter information can be obtained to generate an audio system inspection report. This reduces the manual workload of audio system inspection, quickly locates the source of the fault, and improves the accuracy and efficiency of fault detection, thereby enhancing the stability of the video conferencing system and solving the problem of low debugging efficiency in existing audio system inspection methods. This meets the maintenance needs of modern video conferencing audio systems. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating a video conferencing audio system inspection method provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the structure of an audio system inspection device for video conferencing provided in an embodiment of the present invention; Figure 3 This is a structural block diagram of an audio system inspection device for video conferencing provided in an embodiment of the present invention. Detailed Implementation

[0017] 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.

[0018] It should be noted that the terms "comprising" and "specific" in this invention, and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or units that are not explicitly listed or that are inherent to such process, method, product, or device.

[0019] Please see Figure 1 , Figure 1 This is a flowchart illustrating a video conferencing audio system inspection method provided by an embodiment of the present invention. The video conferencing audio system inspection method includes: S1, obtain the inspection configuration information and audio signal data of the audio system in the video conference; S2, determine the audio analysis reference information of the audio system according to the inspection configuration information; perform time-domain analysis and frequency-domain transformation on the audio signal data according to the audio analysis reference information to obtain the time-frequency data of the audio signal data; S3, Perform audio quality analysis on the time-frequency data based on the audio analysis reference information to obtain the audio quality analysis result of the audio signal data; S4. Based on the audio quality analysis results and the audio analysis benchmark information, determine the audio quality deviation value of the audio signal data; if the audio quality deviation value is not within the preset quality threshold range, trigger the device abnormality location command of the audio system. S5. Based on the device anomaly location command and the real-time audio characteristic data of each audio channel in the audio system, generate an inspection report for the audio system.

[0020] For example, the audio system inspection method for video conferencing described in this embodiment of the invention can be implemented by an inspection server, which is capable of interacting with the target user and with the management terminal. The inspection server acquires the inspection configuration information (which refers to pre-set inspection task parameter configurations, such as inspection time interval, inspection mode type, analysis accuracy level, and target device list) and audio signal data for each audio channel of the audio system in the video conference; determines the audio analysis benchmark information (which refers to a set of audio feature reference data collected and stored under normal working conditions, used as a comparison benchmark for subsequent anomaly detection) based on the inspection configuration information; and performs time-domain analysis and frequency-domain transformation on the audio signal data based on the audio analysis benchmark information to obtain the time-frequency data of the audio signal data; wherein, the time-frequency data includes the time-domain feature parameters of the audio signal data. And spectral distribution data; time-domain characteristic parameters refer to the statistical characteristics of audio signals in the time dimension, including signal amplitude, peak factor, dynamic range, and waveform distortion; spectral distribution data refers to the energy distribution information of audio signals in the frequency domain obtained by Fast Fourier Transform, reflecting the intensity and phase relationship of each frequency component (for example, using a preset time-domain analysis parameter mapping table to determine the analysis window length range and overlap rate range; time-domain analysis obtains time-domain characteristic parameters by calculating the root mean square value, peak detection, and zero-crossing rate statistics of the signal; frequency-domain transformation uses a pre-configured FFT parameter table to determine the transform length range and frequency resolution range, performs signal preprocessing through the Hanning window function, and then performs Fast Fourier Transform).

[0021] Audio quality analysis is performed on the time-frequency data based on audio analysis benchmark information to obtain the audio quality analysis results of the audio signal data. Based on the audio quality analysis results and the audio analysis benchmark information, the audio quality deviation value of the audio signal data is determined (the audio quality deviation value refers to the quantitative index of the difference between the actual detection result and the benchmark value, including signal-to-noise ratio deviation, howling intensity deviation, and echo delay deviation). If the audio quality deviation value is not within the preset quality threshold range (the preset quality threshold range refers to the boundary value of the preset acceptable audio quality range, used to determine whether the audio quality meets normal usage requirements), then the device anomaly location command of the audio system is triggered. Based on the device anomaly location command and the real-time audio characteristic data of each audio channel in the audio system, an inspection report of the audio system is generated (the inspection report refers to structured data containing warning information including the location of the faulty device, fault type, severity, and handling suggestions; the pre-established warning information template library is queried based on the faulty device identification parameter information, and the warning information template library includes device identification code, fault level code, descriptive text template, and processing flow guidance).

[0022] The embodiments of the present invention can significantly reduce the workload of manual inspection, improve the accuracy and efficiency of fault detection, especially in complex multi-device audio systems, it can quickly and accurately locate the source of faults and improve the stability of video conferencing systems.

[0023] It is worth noting that the preset time-domain analysis parameter mapping table is an important reference for the system when performing time-domain analysis of audio signals. It stores the correspondence between different inspection modes and corresponding time-domain analysis window parameters (such as window length and overlap rate). Through the preset time-domain analysis parameter mapping table, appropriate time-domain analysis parameters can be automatically selected according to different inspection scenarios, optimizing inspection efficiency while ensuring detection accuracy, and achieving efficient and accurate inspection of the audio system.

[0024] Specifically, in step S2, determining the audio analysis benchmark information of the audio system based on the inspection configuration information includes: S21, determine the time-domain analysis window parameters and frequency-domain transformation parameters of the audio system according to the inspection mode of the inspection configuration information; S22, Based on the audio quality level of the inspection configuration information, determine the signal-to-noise ratio reference value and echo suppression reference value of the audio system; S23, the time-domain analysis window parameters, the frequency-domain transformation parameters, the signal-to-noise ratio reference value, and the echo suppression reference value are correlated to obtain the audio analysis reference information of the audio system.

[0025] For example, the system receives user-defined inspection configuration information, which includes an inspection mode and an audio quality level. The inspection mode defines the depth and range of detection, while the audio quality level specifies the expected standard for audio performance. Based on the specific settings of the inspection mode, the system determines the corresponding time-domain analysis window parameters and frequency-domain transformation parameters. The time-domain analysis window parameters control the accuracy and range of time-domain feature extraction, while the frequency-domain transformation parameters determine the resolution and bandwidth of the spectrum analysis. Based on the requirements of the audio quality level, the system calculates suitable signal-to-noise ratio (SNR) and echo suppression (ESR) benchmark values. The system then integrates and correlates the time-domain analysis window parameters, frequency-domain transformation parameters, SNR benchmark values, and ESR benchmark values ​​to form complete audio analysis benchmark information. For example, based on the currently set audio quality level identifier, the system queries the corresponding calculation parameters and performs benchmark value calculation, while considering device type correction coefficients and usage environment compensation factors to generate benchmark value data adapted to the current configuration. This embodiment of the invention can dynamically generate suitable analysis benchmarks according to the needs of different application scenarios, improving the accuracy and applicability of audio quality assessment.

[0026] It's worth noting that the inspection mode refers to the preset audio detection strategy type based on different usage scenarios, while the audio quality level refers to the audio quality standard classification set for different business needs, used to determine the detection accuracy and the strictness of the evaluation criteria. The audio quality level benchmark value calculation table contains the benchmark value calculation formula and parameter range corresponding to each quality level. The time domain analysis window parameters refer to the data segmentation settings when performing time domain processing on audio signals, including window length, overlap rate, and window function type; the frequency domain transform parameters refer to the calculation configuration when performing Fast Fourier Transform, including the FFT point count, sampling frequency, and frequency resolution settings.

[0027] The signal-to-noise ratio (SNR) benchmark value refers to the lowest acceptable standard for the ratio of audio signal to background noise under specific audio quality requirements. The SNR benchmark value is calculated using a graded incremental method and dynamically adjusted through quality level weighting coefficients and environmental noise correction factors. The echo suppression benchmark value refers to the minimum degree of echo suppression required by an audio system. It is used to evaluate whether the performance of audio processing equipment meets the standards. The echo suppression benchmark value is determined through a preset suppression degree grading table.

[0028] More specifically, step S21 includes: S211, based on the inspection mode of the inspection configuration information and the preset mode parameter mapping table, obtain the window length and window overlap rate of the audio system; S212, determine the temporal analysis window parameters of the audio system based on the window length and window overlap rate; S213, Based on the window length and preset frequency domain analysis rules, obtain the FFT transform length and frequency resolution of the audio system; S214, determine the frequency domain transformation parameters of the audio system based on the FFT transform length and frequency resolution.

[0029] For example, the corresponding window length and window overlap rate are obtained according to the inspection mode and the preset mode parameter mapping table. The inspection mode includes three types: pre-meeting rapid inspection mode, meeting interval standard inspection mode, and non-working time deep inspection mode. Each mode corresponds to different time constraints and accuracy requirements. The time domain analysis window parameters are determined according to the window length and window overlap rate. The window length determines the amount of data analyzed in a single operation, while the window overlap rate affects the data continuity between adjacent analysis windows and the detection time resolution. The FFT transform length and frequency resolution are obtained according to the window length and the preset frequency domain analysis rules. The FFT transform length is usually related to the window length but computational efficiency needs to be considered, while the frequency resolution directly affects the accuracy of frequency domain feature extraction. The frequency domain transform parameters are determined according to the FFT transform length and frequency resolution to form a complete frequency domain analysis configuration. This embodiment of the invention can automatically adjust the analysis parameters according to the characteristics of different inspection scenarios, optimizing inspection efficiency while ensuring detection accuracy.

[0030] It's worth noting that the inspection modes include a pre-meeting rapid inspection mode, a meeting-interval standard inspection mode, and a non-working-hour deep inspection mode. A key-value pair structure is used to store the parameter ranges corresponding to each mode, establishing a preset mode parameter mapping table. For example, the pre-meeting rapid inspection mode sets the window length range to 8ms to 20ms and the window overlap rate range to 25% to 40% to achieve rapid scanning and detection. The meeting-interval standard inspection mode sets the window length range to 20ms to 40ms and the window overlap rate range to 50% to 65%, striking a balance between detection accuracy and time cost. The non-working-hour deep inspection mode sets the window length range to 40ms to 80ms and the window overlap rate range to 70% to 85%, ensuring the highest detection accuracy. Based on the input inspection mode identifier, a hash lookup algorithm quickly retrieves the corresponding window length and overlap rate parameters from the mapping table and dynamically adjusts them within the parameter range according to the current system load.

[0031] Pre-meeting rapid inspection mode refers to a quick equipment status check performed before the meeting begins, focusing on detecting basic connectivity and major faults; Meeting break standard inspection mode refers to a routine quality inspection performed during meeting breaks, balancing inspection accuracy and time efficiency; Off-peak time in-depth inspection mode refers to a comprehensive and in-depth inspection performed during system idle periods, pursuing the highest inspection accuracy and coverage. The preset mode parameter mapping table is a pre-established data structure that maps inspection modes to analysis parameters.

[0032] The time-domain analysis window parameters refer to the complete set of parameters used for time-domain signal processing, including the number of effective window samples, window step length, window function type, and boundary handling method. In optional embodiments, the window function type is determined by a pre-established function selection table: fast mode uses a rectangular window or Hanning window, standard mode uses a Hanning window or Hamming window, and deep mode uses a Blackman window or Kaiser window. The preset frequency-domain analysis rules are preset criteria used to determine frequency-domain transform parameters (such as FFT transform length and frequency resolution) based on the time-domain analysis window length.

[0033] Specifically, step S3 includes: S31, based on the signal-to-noise ratio reference value in the audio analysis reference information, a preset signal-to-noise separation algorithm is used to perform signal-to-noise separation processing on the time-frequency data to obtain the signal-to-noise separation result of the audio signal data; S32, based on the frequency domain transformation parameters in the audio analysis reference information, a preset howling recognition algorithm is used to perform howling recognition processing on the time-frequency data to obtain the howling recognition result of the audio signal data; S33, based on the echo suppression reference value in the audio analysis reference information, a preset echo perception algorithm is used to perform echo perception processing on the time-frequency data to obtain the echo detection result of the audio signal data; S34, combine the signal-to-noise separation result, the howling recognition result, and the echo detection result to obtain the audio quality analysis result of the audio signal data.

[0034] For example, based on the signal-to-noise ratio (SNR) benchmark value in the audio analysis benchmark information, a preset SNR separation algorithm is used to perform SNR separation processing on the time-frequency data to obtain the SNR separation result of the audio signal data. For example, by matching a preset noise feature library (including spectral templates such as fan noise and current noise), useful signals and noise are separated to obtain the SNR separation result (actual SNR value). Based on the frequency domain transformation parameters in the audio analysis benchmark information, a preset howling recognition algorithm is used to perform howling recognition processing on the time-frequency data to obtain the howling recognition result of the audio signal data. For example, monitoring whether there is a single frequency component in the spectral distribution data, when the duration (100ms to 2s) and amplitude gain (exceeding the benchmark value by 10dB to 30dB) of the component meet preset conditions, it is determined to be howling, and howling recognition result (howling intensity and frequency point) is generated. Based on the echo suppression benchmark value in the audio analysis benchmark information, a preset echo perception algorithm is used to perform echo perception processing on the time-frequency data to obtain the echo detection result of the audio signal data. For example, the correlation between the signal delay characteristics and spectral distribution data in the time-domain feature parameters is analyzed. When the cross-correlation coefficient (0.3 to 0.8) and delay range (10ms to 300ms) of the original signal and the delayed signal meet the set standard, it is identified as an echo, and the echo detection result (echo delay time and intensity attenuation) is obtained. The signal-to-noise separation result, the howling recognition result, and the echo detection result are combined to obtain the audio quality analysis result of the audio signal data.

[0035] Understandably, signal-to-noise separation uses a pre-established noise feature library for matching and identification. This library contains spectral templates for fan noise, current noise, and ambient noise. Feedback detection employs a pre-defined frequency monitoring interval table, dividing the audio spectrum into three monitoring intervals: low-frequency, mid-frequency, and high-frequency. Feedback is identified when a single frequency component is detected for a duration exceeding 100ms to 2s and its amplitude gain exceeds a reference value by 10dB to 30dB. Echo detection establishes a delay correlation analysis matrix, calculating the cross-correlation coefficient between the original signal and the delayed signal. Echoes are identified when the correlation coefficient exceeds 0.3 to 0.8 and the delay range is 10ms to 300ms. The echo intensity level is determined by combining this with amplitude attenuation rate analysis.

[0036] It is worth noting that the preset signal-to-noise separation algorithm refers to a noise detection method based on spectral subtraction and adaptive filtering, the preset howling recognition algorithm refers to a feedback sound detection method based on frequency domain peak detection and duration analysis, and the preset echo perception algorithm refers to an echo detection method based on time delay estimation and correlation analysis.

[0037] Specifically, in step S4, determining the audio quality deviation value of the audio signal data based on the audio quality analysis results and the audio analysis benchmark information includes: S41, compare the signal-to-noise separation result with the signal-to-noise ratio reference value, and calculate the signal-to-noise ratio deviation value of the audio signal data; S42, compare the feedback recognition result with the preset feedback threshold, and calculate the feedback deviation value of the audio signal data; S43, compare the echo detection result with the echo suppression reference value, and calculate the echo deviation value of the audio signal data; S44. Based on the preset quality index weighting coefficient, the signal-to-noise ratio deviation value, the howling deviation value, and the echo deviation value are weighted and calculated to obtain the audio quality deviation value of the audio signal data.

[0038] For example, the signal-to-noise separation result is compared with the signal-to-noise ratio (SNR) benchmark value in the audio analysis benchmark information to obtain the SNR deviation value; the feedback recognition result is compared with a preset feedback threshold (the preset feedback threshold can be set according to actual conditions based on expert experience) to obtain the feedback deviation value; the echo detection result is compared with the echo suppression benchmark value in the audio analysis benchmark information to obtain the echo deviation value; the SNR deviation value, feedback deviation value, and echo deviation value are weighted according to preset quality index weighting coefficients (the preset quality index weighting coefficients can be set according to actual conditions by combining subjective weighting methods and objective weighting methods). Through reasonable weight allocation, different types of quality problems are unified to the same evaluation scale to obtain an audio quality deviation value that can comprehensively reflect the overall quality status of the audio system. This embodiment of the invention comprehensively considers the superposition effect of multiple audio quality problems, improving the comprehensiveness and accuracy of fault detection.

[0039] It is understandable that the signal-to-noise ratio (SNR) result refers to the actual SNR value obtained after processing by the SNR algorithm, expressed in dB as the power ratio of the useful audio signal to the background noise. The SNR reference value refers to the preset SNR standard reference value in the audio analysis reference information. The SNR deviation value refers to the difference between the measured SNR and the reference value, used to assess the degree of deviation of the current audio quality from the standard requirements.

[0040] Feedback recognition results refer to the feedback intensity and duration parameters output by the feedback recognition algorithm, including the amplitude gain of the feedback frequency point and the duration of the feedback phenomenon. The preset feedback threshold refers to the maximum acceptable value of the feedback phenomenon. Exceeding this threshold is judged as an audio quality problem. Feedback deviation value refers to the quantitative index of the deviation between the actual detected feedback intensity and the threshold.

[0041] Echo detection results refer to the echo delay time and echo intensity attenuation output by the echo perception algorithm, reflecting the severity of echo phenomena in the audio system. Echo suppression benchmark value refers to the minimum echo suppression requirement specified in the audio analysis benchmark information. Echo deviation value refers to the quantified result of the deviation between the actual echo suppression effect and the benchmark requirement.

[0042] The preset quality index weighting coefficients refer to the importance weighting allocation of each audio quality parameter in the comprehensive evaluation, reflecting the relative importance of different audio problems to the overall quality. The audio quality deviation value is a quantitative indicator of the overall audio system quality deviation obtained through weighted calculation, used to uniformly evaluate the overall performance status of the audio system. For example, a dynamic weighting allocation strategy table is established, setting corresponding weighting coefficient combinations according to different application scenarios and audio quality levels. Application scenarios include voice communication, music playback, remote conferencing, sound reinforcement, and recording / broadcasting. The signal-to-noise ratio deviation weighting coefficient ranges from 0.4 to 0.6, with a higher weight in voice communication scenarios and a moderate weight in music playback scenarios; the feedback deviation weighting coefficient ranges from 0.2 to 0.4, with a higher weight in sound reinforcement systems and a lower weight in recording / broadcasting systems; the echo deviation weighting coefficient ranges from 0.2 to 0.4, with a higher weight in remote conferencing scenarios and a lower weight in local playback scenarios. The total weighting coefficients are always kept at 1.0 to ensure the consistency of the calculation results.

[0043] Specifically, step S5 includes: S51, according to the device anomaly location instruction, obtain real-time audio feature data of each audio channel in the audio system; determine the audio feature deviation information of each audio channel according to the real-time audio feature data and the audio analysis benchmark information; S52, perform comparative analysis on each audio channel based on the audio feature deviation information to obtain the channel comparison analysis result information of the audio system; S53, based on the channel comparison analysis results and the audio feature deviation information, obtain the current abnormal state information of the audio system; S54, Based on the current abnormal state information and the real-time audio feature data, obtain the fault device identification parameter information of the audio system; S55, Generate an inspection report for the audio system based on the faulty equipment identification parameter information.

[0044] For example, upon responding to a device anomaly location command, real-time audio feature data for each audio channel is collected, including audio amplitude features (such as dynamic range and gain characteristics), frequency response features (such as transmission characteristic curves of signals at different frequencies), and phase features (such as phase shift and group delay characteristics). The deviation of these features relative to the audio analysis benchmark information is calculated to form audio feature deviation information for each channel. This deviation information includes the degree of deviation of each channel's audio amplitude, frequency response, and phase features relative to the benchmark state. Based on the audio feature deviation information for each channel, a channel comparison analysis command is triggered, initiating a multi-channel horizontal comparison analysis process to obtain channel comparison analysis results. The channel comparison analysis results refer to the feature difference between the abnormal channel and the normal channel. Based on the channel comparison analysis results and the audio feature deviation information for each channel, the current abnormal state information is obtained. Based on the current abnormal state information and the real-time audio feature data for each audio channel, faulty device identification parameter information is obtained through feature pattern matching and anomaly degree assessment. An inspection report for the audio system is generated based on the faulty device identification parameter information. An early warning is issued to the user terminal based on the content of the inspection report. This embodiment of the invention effectively eliminates environmental interference through multi-channel feature comparison, achieving precise location of faulty devices.

[0045] It is worth noting that the device anomaly location instruction refers to the fault location task instruction automatically generated by the system after detecting that the audio quality deviation exceeds the threshold. It includes the anomaly type identifier, severity level, and preliminary location range. The audio feature deviation information of each channel refers to the deviation of the feature parameters of each audio channel relative to the normal operating state, which is used to identify the specific abnormal channel and deviation mode.

[0046] Channel comparison analysis commands are horizontal comparison analysis tasks automatically initiated based on the differences in characteristic deviations between channels. They are used to identify characteristic difference patterns between abnormal and normal channels, improving the accuracy of fault location through correlation analysis of multi-channel data. For example, a channel comparison analysis trigger condition judgment table is established, defining the threshold conditions and priority rules for initiating comparison analysis. When the comprehensive deviation index of a single channel exceeds a preset threshold range, a single-channel deep analysis command is automatically triggered; when multiple channels simultaneously exhibit deviations and the differences in deviations between channels exceed a preset difference threshold range, a multi-channel correlation analysis command is triggered. The system determines the scope and depth of the comparison analysis through a preset analysis strategy selection algorithm, including three analysis modes: adjacent channel comparison, similar channel comparison, and full-channel comprehensive comparison. Analysis commands include a comparison channel list, analysis parameter configuration, output format requirements, and analysis priority settings. Priorities are divided into three levels: urgent, important, and general, corresponding to different resource allocations and processing sequences. The generated comparison analysis command is sent to the analysis processing module to initiate the corresponding channel feature comparison calculation process.

[0047] Faulty equipment identification parameters refer to a set of key parameters used to locate specific faulty equipment. These parameters include equipment type identifier, fault type classification, severity rating, and repair suggestion code. Accurate identification of faulty equipment is achieved through comprehensive analysis of abnormal states and channel characteristics. For example, a faulty equipment identification parameter mapping database is established, storing characteristic parameter templates for different equipment types under various fault modes. Equipment type identification is determined through a channel function mapping table: microphone channels correspond to pickup devices, speaker channels to playback devices, processor channels to audio processing devices, and transmission channels to network devices. Fault type classification uses a decision tree algorithm: based on amplitude characteristic anomaly patterns, it determines power amplifier faults or connection faults; based on frequency response anomaly patterns, it determines filter faults or speaker faults; and based on phase characteristic anomaly patterns, it determines delay faults or synchronization faults.

[0048] More specifically, step S53 includes: S531, Based on the channel comparison analysis results, obtain the timestamp sequence information of each abnormal channel in the audio system; S532, based on the timestamp sequence information, perform abnormal propagation path tracing for each abnormal channel to obtain path tracing result information for each abnormal channel; S533, Based on the path tracing result information and the audio feature deviation information, obtain the primary fault channel identification information of the audio system; S534, Based on the original fault channel identification information, determine the fault source from the abnormal channel of the audio system to obtain the current abnormal state information of the audio system.

[0049] For example, since anomalies in audio systems often propagate and spread across multiple channels through signal transmission paths, multiple channels may simultaneously exhibit abnormal characteristics. Simple anomaly detection cannot distinguish between primary and secondary anomalies. For instance, when the main microphone experiences feedback, the feedback signal propagates through the audio processor to the speaker channel, affecting other microphone channels and forming a complex anomaly propagation chain. If this propagation process cannot be accurately tracked, secondary anomaly channels may be misidentified as the source of the fault, causing maintenance personnel to waste time on the wrong equipment. The solution involves obtaining the timestamp sequence information of each anomalous channel based on the channel comparison analysis results; triggering an anomaly propagation path tracing command based on the timestamp sequence information to initiate dynamic analysis of the anomaly propagation process across multiple channels and obtaining the anomaly propagation path tracing results; reconstructing the anomaly propagation process by analyzing the order of appearance of anomaly features on the time axis and the spatial propagation path; obtaining the primary fault channel identification information based on the anomaly propagation path tracing results and the audio feature deviation information of each channel to determine the initial source of the anomaly; and accurately identifying the fault source from multiple anomalous channels based on the primary fault channel identification information to obtain current anomaly status information reflecting the true fault condition. This invention accurately identifies the primary fault source through anomaly propagation path analysis, avoiding interference from secondary anomalies in fault location.

[0050] It is worth noting that timestamp sequence information refers to the precise time record sequence of the first occurrence of abnormal signs in each abnormal channel, including the start time of the abnormality, the time point of change in the intensity of the abnormality, and the time of stability of the abnormality. It is used to analyze the temporal development pattern of abnormal phenomena and the temporal correlation between channels.

[0051] Anomaly propagation path tracing commands refer to propagation path analysis tasks automatically initiated based on the temporal relationships of channel anomalies. These tasks identify the propagation direction and impact diffusion patterns of anomalies across different channels, determining the origin channel and propagation mechanism of the anomaly through temporal analysis. For example, an anomaly propagation path tracing trigger strategy library is established, including identification conditions for three basic propagation modes: single-source propagation mode, multi-source concurrent mode, and cascading propagation mode. When multiple anomaly channels are detected with timestamp intervals within a preset propagation time window of 100 milliseconds to 5 seconds, a propagation path tracing command is triggered. The system determines the physical connections and logical relationships between channels through a pre-established channel topology graph, which includes three connection types: direct connection, indirect influence, and independent isolation.

[0052] Anomaly propagation path tracing results refer to the direction and scope of anomaly propagation determined based on the timing of anomaly occurrence and the correlation between channels.

[0053] In this embodiment, a propagation path analysis model is established. The propagation direction is determined by calculating the order and time interval of abnormal timestamps between channels. The time interval ranges from a few milliseconds to tens of seconds, reflecting different propagation speeds. The correlation between channels is determined by a preset correlation calculation formula, which comprehensively considers the range of physical distance coefficients, the range of signal flow weights, and the range of frequency response similarity. The scope of influence is determined by analyzing the maximum diffusion distance of abnormal propagation and the number of affected channels. The diffusion distance is calculated based on the number of channel hops, ranging from 1 to 5 hops, and the number of affected channels ranges from 2 to 50% of the total number of channels in the system. The identification of the primary fault channel adopts a multi-dimensional comprehensive evaluation method, including the time priority scoring range, the feature deviation severity scoring range, and the propagation centrality scoring range. The primary probability score of each channel is calculated by weighted summation, and the channel with the highest score is identified as the primary fault channel. The identification results are matched and verified with a historical fault pattern database to improve the identification accuracy.

[0054] This invention discloses a method for inspecting the audio system of a video conferencing system. The method involves: acquiring inspection configuration information and audio signal data of the audio system in the video conference; determining audio analysis benchmark information of the audio system based on the inspection configuration information; performing time-domain analysis and frequency-domain transformation on the audio signal data based on the audio analysis benchmark information to obtain time-frequency data of the audio signal data; performing audio quality analysis on the time-frequency data based on the audio analysis benchmark information to obtain audio quality analysis results of the audio signal data; determining the audio quality deviation value of the audio signal data based on the audio quality analysis results and the audio analysis benchmark information; if the audio quality deviation value is not within a preset quality threshold range, triggering a device anomaly location command for the audio system; and generating an inspection report for the audio system based on the device anomaly location command and real-time audio feature data of each audio channel in the audio system. By comparing the analysis results of audio signal data with the established audio analysis benchmark, faulty equipment identification parameter information can be obtained to generate an audio system inspection report. This reduces the manual workload of audio system inspection, quickly locates the source of the fault, and improves the accuracy and efficiency of fault detection, thereby enhancing the stability of the video conferencing system and solving the problem of low debugging efficiency in existing audio system inspection methods. This meets the maintenance needs of modern video conferencing audio systems.

[0055] See Figure 2 , Figure 2 This is a schematic diagram of the structure of an audio system inspection device 10 for video conferencing provided in an embodiment of the present invention. The audio system inspection device 10 for video conferencing includes: The inspection data acquisition module 11 is used to acquire the inspection configuration information and audio signal data of the audio system in the video conference. The time-frequency data acquisition module 12 is used to determine the audio analysis reference information of the audio system according to the inspection configuration information; and to perform time-domain analysis and frequency-domain transformation on the audio signal data according to the audio analysis reference information to obtain the time-frequency data of the audio signal data. The audio quality analysis module 13 is used to perform audio quality analysis on the time-frequency data based on the audio analysis reference information to obtain the audio quality analysis results of the audio signal data; The abnormal instruction triggering module 14 is used to determine the audio quality deviation value of the audio signal data based on the audio quality analysis results and the audio analysis benchmark information; if the audio quality deviation value is not within the preset quality threshold range, the device abnormality positioning instruction of the audio system is triggered. The inspection report generation module 15 is used to generate an inspection report for the audio system based on the device anomaly location command and the real-time audio feature data of each audio channel in the audio system.

[0056] The audio system inspection device 10 for video conferencing provided in this embodiment of the invention can realize all the processes of the audio system inspection method for video conferencing in the above embodiment. The functions and technical effects of each module in the device are the same as those of the audio system inspection method for video conferencing in the above embodiment, and will not be repeated here.

[0057] See Figure 3 , Figure 3 This is a schematic diagram of the structure of an audio system inspection device 20 for video conferencing provided in an embodiment of the present invention. The audio system inspection device 20 for video conferencing in this embodiment includes: a processor 21, a memory 22, and a computer program stored in the memory 22 and executable on the processor 21. When the processor 21 executes the computer program, it implements the steps in the above-described embodiment of the audio system inspection method for video conferencing. Alternatively, when the processor 21 executes the computer program, it implements the functions of each module in the above-described embodiment of the audio system inspection device for video conferencing.

[0058] For example, the computer program may be divided into one or more modules, which are stored in the memory 22 and executed by the processor 21 to complete the present invention. The one or more modules may be a series of computer program instruction segments capable of performing specific functions, which describe the execution process of the computer program in the audio system inspection device 20 of the video conferencing system.

[0059] The audio system inspection device 20 for video conferencing can be a computing device such as a desktop computer, laptop, handheld computer, or cloud server. The audio system inspection device 20 may include, but is not limited to, a processor 21 and a memory 22. Those skilled in the art will understand that the schematic diagram is merely an example of the audio system inspection device 20 for video conferencing and does not constitute a limitation on the device. It may include more or fewer components than illustrated, or combine certain components, or use different components. For example, the audio system inspection device 20 for video conferencing may also include input / output devices, network access devices, buses, etc.

[0060] The processor 21 may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor. The processor 21 is the control center of the video conferencing audio system inspection equipment 20, connecting all parts of the video conferencing audio system inspection equipment 20 via various interfaces and lines.

[0061] The memory 22 can be used to store the computer programs and / or modules. The processor 21 implements various functions of the audio system inspection device 20 for video conferencing by running or executing the computer programs and / or modules stored in the memory 22 and calling the data stored in the memory 22. The memory 22 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the mobile phone (such as audio data, phonebook, etc.). In addition, the memory 22 may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0062] The integrated module of the audio system inspection device 20 for video conferencing, if implemented as a software functional unit and sold or used as an independent product, can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by the processor 21, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content contained in the computer-readable medium may be appropriately added to or subtracted from the content as required by the legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium may not include electrical carrier signals and telecommunication signals.

[0063] It should be noted that the device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the accompanying drawings of the device embodiments provided by this invention, the connection relationships between modules indicate that they have communication connections, which can be specifically implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without any creative effort.

[0064] This invention also provides a computer-readable storage medium comprising a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to perform an audio system inspection method for video conferencing as described in the above embodiments.

[0065] Furthermore, embodiments of the present invention also provide a computer program product, which is stored in a storage medium and executed by at least one processor to implement the steps of the audio system inspection method for video conferencing described above.

[0066] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for inspecting the audio system of a video conferencing system, characterized in that, include: Acquire the inspection configuration information and audio signal data of the audio system during video conferencing; The audio analysis baseline information of the audio system is determined based on the inspection configuration information; Based on the audio analysis reference information, time-domain analysis and frequency-domain transformation are performed on the audio signal data to obtain the time-frequency data of the audio signal data. Based on the audio analysis benchmark information, audio quality analysis is performed on the time-frequency data to obtain the audio quality analysis results of the audio signal data; Based on the audio quality analysis results and the audio analysis benchmark information, the audio quality deviation value of the audio signal data is determined; if the audio quality deviation value is not within the preset quality threshold range, the device abnormality location command of the audio system is triggered. Based on the device anomaly location command and the real-time audio characteristic data of each audio channel in the audio system, an inspection report of the audio system is generated. The step of performing audio quality analysis on the time-frequency data based on the audio analysis benchmark information to obtain the audio quality analysis result of the audio signal data includes: Based on the signal-to-noise ratio (SNR) benchmark value in the audio analysis benchmark information, a preset SNR separation algorithm is used to perform SNR separation processing on the time-frequency data to obtain the SNR separation result of the audio signal data; Based on the frequency domain transformation parameters in the audio analysis reference information, a preset howling recognition algorithm is used to perform howling recognition processing on the time-frequency data to obtain the howling recognition result of the audio signal data; Based on the echo suppression benchmark value in the audio analysis benchmark information, the time-frequency data is processed by a preset echo perception algorithm to obtain the echo detection result of the audio signal data; The signal-to-noise separation result, the howling identification result, and the echo detection result are combined to obtain the audio quality analysis result of the audio signal data; The step of generating an inspection report for the audio system based on the device anomaly location command and the real-time audio characteristic data of each audio channel in the audio system includes: According to the device anomaly location command, real-time audio feature data of each audio channel in the audio system is obtained; based on the real-time audio feature data and the audio analysis benchmark information, audio feature deviation information of each audio channel is determined. Based on the audio feature deviation information, each audio channel is compared and analyzed to obtain the channel comparison analysis result information of the audio system; Based on the channel comparison analysis results and the audio feature deviation information, the current abnormal state information of the audio system is obtained; Based on the current abnormal state information and the real-time audio feature data, obtain the fault device identification parameter information of the audio system; An inspection report for the audio system is generated based on the faulty equipment identification parameter information. The step of obtaining the current abnormal state information of the audio system based on the channel comparison analysis results and the audio feature deviation information includes: Based on the channel comparison analysis results, the timestamp sequence information of each abnormal channel in the audio system is obtained; Based on the timestamp sequence information, the abnormal propagation path of each abnormal channel is traced to obtain the path tracing result information of each abnormal channel; Based on the path tracing results and the audio feature deviation information, the primary fault channel identification information of the audio system is obtained; Based on the original fault channel identification information, the fault source is determined from the abnormal channel of the audio system to obtain the current abnormal state information of the audio system; The step of determining the audio analysis benchmark information of the audio system based on the inspection configuration information includes: Based on the inspection mode of the inspection configuration information, determine the time-domain analysis window parameters and frequency-domain transformation parameters of the audio system; Based on the audio quality level of the inspection configuration information, determine the signal-to-noise ratio reference value and echo suppression reference value of the audio system; By correlating the time-domain analysis window parameters, the frequency-domain transformation parameters, the signal-to-noise ratio reference value, and the echo suppression reference value, audio analysis reference information of the audio system is obtained. The inspection modes include a pre-meeting rapid inspection mode, a meeting break standard inspection mode, and a non-working time in-depth inspection mode. A key-value pair structure is used to store the parameter range corresponding to each mode, and a preset mode parameter mapping table is established. Based on the input inspection mode identifier, the corresponding window length and overlap rate parameters are quickly retrieved from the mapping table using a hash lookup algorithm, and dynamically adjusted within the parameter range according to the current system load.

2. The method for inspecting the audio system of a video conferencing system as described in claim 1, characterized in that, The step of determining the time-domain analysis window parameters and frequency-domain transformation parameters of the audio system based on the inspection mode of the inspection configuration information includes: Based on the inspection mode of the inspection configuration information and the preset mode parameter mapping table, the window length and window overlap rate of the audio system are obtained; The temporal analysis window parameters of the audio system are determined based on the window length and window overlap rate. Based on the window length and preset frequency domain analysis rules, the FFT transform length and frequency resolution of the audio system are obtained; The frequency domain transformation parameters of the audio system are determined based on the FFT transform length and frequency resolution.

3. An audio system inspection device for video conferencing, characterized in that, include: The inspection data acquisition module is used to acquire the inspection configuration information and audio signal data of the audio system in video conferencing; The time-frequency data acquisition module is used to determine the audio analysis reference information of the audio system based on the inspection configuration information; and to perform time-domain analysis and frequency-domain transformation on the audio signal data based on the audio analysis reference information to obtain the time-frequency data of the audio signal data. The audio quality analysis module is used to perform audio quality analysis on the time-frequency data based on the audio analysis benchmark information to obtain the audio quality analysis results of the audio signal data. An abnormal instruction triggering module is used to determine the audio quality deviation value of the audio signal data based on the audio quality analysis results and the audio analysis benchmark information; if the audio quality deviation value is not within the preset quality threshold range, the device abnormality location instruction of the audio system is triggered. The inspection report generation module is used to generate an inspection report for the audio system based on the device anomaly location command and the real-time audio feature data of each audio channel in the audio system. The audio quality analysis module is used for: Based on the signal-to-noise ratio (SNR) benchmark value in the audio analysis benchmark information, a preset SNR separation algorithm is used to perform SNR separation processing on the time-frequency data to obtain the SNR separation result of the audio signal data; Based on the frequency domain transformation parameters in the audio analysis reference information, a preset howling recognition algorithm is used to perform howling recognition processing on the time-frequency data to obtain the howling recognition result of the audio signal data; Based on the echo suppression benchmark value in the audio analysis benchmark information, the time-frequency data is processed by a preset echo perception algorithm to obtain the echo detection result of the audio signal data; The signal-to-noise separation result, the howling identification result, and the echo detection result are combined to obtain the audio quality analysis result of the audio signal data; The inspection report generation module is used for: According to the device anomaly location command, real-time audio feature data of each audio channel in the audio system is obtained; based on the real-time audio feature data and the audio analysis benchmark information, audio feature deviation information of each audio channel is determined. Based on the audio feature deviation information, each audio channel is compared and analyzed to obtain the channel comparison analysis result information of the audio system; Based on the channel comparison analysis results and the audio feature deviation information, the current abnormal state information of the audio system is obtained; Based on the current abnormal state information and the real-time audio feature data, obtain the fault device identification parameter information of the audio system; An inspection report for the audio system is generated based on the faulty equipment identification parameter information. The step of obtaining the current abnormal state information of the audio system based on the channel comparison analysis results and the audio feature deviation information includes: Based on the channel comparison analysis results, the timestamp sequence information of each abnormal channel in the audio system is obtained; Based on the timestamp sequence information, the abnormal propagation path of each abnormal channel is traced to obtain the path tracing result information of each abnormal channel; Based on the path tracing results and the audio feature deviation information, the primary fault channel identification information of the audio system is obtained; Based on the original fault channel identification information, the fault source is determined from the abnormal channel of the audio system to obtain the current abnormal state information of the audio system; The step of determining the audio analysis benchmark information of the audio system based on the inspection configuration information includes: Based on the inspection mode of the inspection configuration information, determine the time-domain analysis window parameters and frequency-domain transformation parameters of the audio system; Based on the audio quality level of the inspection configuration information, determine the signal-to-noise ratio reference value and echo suppression reference value of the audio system; By correlating the time-domain analysis window parameters, the frequency-domain transformation parameters, the signal-to-noise ratio reference value, and the echo suppression reference value, audio analysis reference information of the audio system is obtained. The inspection modes include a pre-meeting rapid inspection mode, a meeting break standard inspection mode, and a non-working time in-depth inspection mode. A key-value pair structure is used to store the parameter range corresponding to each mode, and a preset mode parameter mapping table is established. Based on the input inspection mode identifier, the corresponding window length and overlap rate parameters are quickly retrieved from the mapping table using a hash lookup algorithm, and dynamically adjusted within the parameter range according to the current system load.

4. An audio system inspection device for video conferencing, characterized in that, The method includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor executes the computer program to implement the audio system inspection method for video conferencing as described in any one of claims 1-2.

5. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to perform the audio system inspection method for video conferencing as described in any one of claims 1-2.

6. A computer program product, characterized in that, The computer program product is stored in a storage medium, and the program product is executed by at least one processor to implement the steps of the audio system inspection method for video conferencing as described in any one of claims 1-2.

Images