Automatic audio and video system operation and maintenance method based on large model

Abstract

The invention is suitable for the technical field of audio and video system operation and maintenance, and particularly relates to an automatic audio and video system operation and maintenance method based on a large model, and the method comprises the steps: constructing an audio and video operation and maintenance platform, embedding a pre-trained operation and maintenance large model, collecting the use record of audio and video equipment, and defining a historical user and a current user, the method comprises the following steps: establishing a use scene of audio and video equipment, positioning a deployment position of video monitoring equipment, acquiring video monitoring data, and intercepting state data of the audio and video equipment, the state data at least comprising a state lamp and a system log. According to the method, whether the video system breaks down or not can be judged by intercepting the snapshots, so that the operation and maintenance efficiency of the audio and video system is greatly improved, the fault position can be quickly positioned and recognized by sending the state data to the historical user, the troubleshooting time is shortened, the interruption time and resource waste caused by the fault are reduced, and the user experience is improved. And the operation and maintenance efficiency of the audio and video system is greatly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of operation and maintenance of audio - video systems, and particularly to an automated audio - video system operation and maintenance method based on a large model. Background Art

[0002] An audio - video system is a system that integrates functions such as audio and video signal acquisition, processing, transmission, and control. It is widely used in scenarios such as meetings, education, entertainment, security, and broadcasting. It usually includes microphones, cameras, speakers, display screens, signal processing devices, and control terminals, etc. The audio - video system can achieve synchronous transmission and management of sound and images. Through the audio - video system, users can perform operations such as remote meetings, real - time monitoring, multimedia teaching, performance sound reinforcement, and audio - visual playback.

[0003] However, due to the high integration and specialization of audio - video system devices, the operation difficulty is also relatively large. If the user operates improperly, problems such as audio - video asynchronization, signal interference, and equipment failures may occur. Moreover, the maintenance and fault troubleshooting before using the audio - video system also require professional knowledge. Therefore, "how to use video monitoring devices and pick - up microphones to identify faults in the audio - video system" is the technical problem to be solved by the present invention. Summary of the Invention

[0004] The purpose of the present invention is to provide an automated audio - video system operation and maintenance method based on a large model to solve the problem of "how to use video monitoring devices and pick - up microphones to identify faults in the audio - video system" proposed in the above - mentioned background art.

[0005] To achieve the above - mentioned purpose, the present invention provides the following technical solutions:

[0006] An automated audio - video system operation and maintenance method based on a large model, the method includes:

[0007] Construct an audio - video operation and maintenance platform, embed a pre - trained operation and maintenance large model, collect the usage records of audio - video devices, define historical users and current users, construct the usage scenarios of audio - video devices, locate the deployment positions of video monitoring devices, collect video monitoring data, and intercept the status data of audio - video devices, where the status data at least includes: status lights and system logs;

[0008] Via the audio - video operation and maintenance platform, receive the boot instruction uploaded by the current user, start the self - inspection mechanism pre - integrated in the audio - video operation and maintenance platform, and determine whether there is a fault in the audio - video device. If so, traverse the preset fault query table to determine the disposal plan. If not, use the pick - up microphone pre - deployed in the usage scenario to collect audio data, extract human voices and sound effects, and determine whether both belong to the preset frequency range. If so, generate an audio system normal instruction and send it to the audio - video operation and maintenance platform. From the video surveillance data, intercept a snapshot containing the video playback device and determine whether there are abnormal features. If not, output a video system normal instruction;

[0009] Determine whether the audio - video operation and maintenance platform has received the normal instructions of the audio system and the video system. If so, send a boot instruction to the audio - video device. If not, send the status data to all historical users, subscribe to the feedback information sent by historical users, and establish a conversation;

[0010] Further, the steps of building the audio - video operation and maintenance platform, embedding the pre - trained operation and maintenance large model, and collecting the usage records of the audio - video device and defining historical users and current users include:

[0011] Embed a time stamp into the usage record to build a behavior record table, where the behavior record table consists of a personnel item and a time item;

[0012] Collect the usage data of the current user, where the usage data at least includes: start time, running duration, and historical faults, and integrate the usage data into the behavior record table.

[0013] Further, the steps of building the usage scenario of the audio - video device, locating the deployment location of the video surveillance device, and collecting the video surveillance data include:

[0014] Configure the unique identifier of the audio - video device, and divide the video surveillance data into several blocks, where each audio - video device corresponds to one block;

[0015] Establish the correspondence relationship between the blocks, the audio - video devices, and the unique identifiers.

[0016] Further, the steps of intercepting the status data of the audio - video device, where the status data at least includes: status lights and system logs, include:

[0017] Input the status data into the operation and maintenance large model, output the potential hazard features, and send the potential hazard features to the preset terminal;

[0018] Based on the potential hazard features, build an adjustment mechanism for the status data.

[0019] Further, the method further includes:

[0020] Based on the usage data, record the adjustment frequency of each audio - video device;

[0021] Based on the adjustment frequency, cluster the audio - video devices into several priorities, and correct the self - inspection mechanism in the order from high to low priority.

[0022] Further, the steps of collecting audio data and extracting human voices and sound effects include:

[0023] Input the audio data into the operation and maintenance large - model to extract human voices and sound effects;

[0024] Configure evaluation metrics for the audio data, where the evaluation metrics at least include: distortion item, echo item, and noise item. Determine whether there are defective items in the audio data. If so, send the defective items to the historical user and generate an audio system exception instruction.

[0025] Further, the steps of intercepting a snapshot containing a video playback device from the video surveillance data and determining whether there are abnormal features, and if not, outputting a video system normal instruction include:

[0026] Construct an abnormal feature set, compare the snapshot with the abnormal feature set, determine whether there are abnormal features. If so, integrate the abnormal features into the status data.

[0027] Further, the steps of subscribing to feedback information sent by historical users and establishing a conversation include:

[0028] Extract processing steps from the feedback information;

[0029] Identify behavioral features from the video surveillance data and verify the processing steps.

[0030] Further, the method further includes:

[0031] Create an operation and maintenance community and integrate the operation and maintenance community into the audio - video operation and maintenance platform;

[0032] Publish the status data to the operation and maintenance community, select the best reply, and integrate the best reply into the feedback information.

[0033] Further, the method further includes:

[0034] Open the usage permissions of the video surveillance device and the pickup to the historical user, establish a control link, and establish a conversation;

[0035] Create a dialogue path and send access requests to the current user and historical users.

[0036] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0037] By determining the status data of the audio-visual device, anomalies can be detected in a timely manner, providing a data basis for troubleshooting the audio-visual system. By starting the self-check mechanism, potential problems in the audio-visual system can be quickly identified, saving time and labor costs and improving the operation and maintenance efficiency. By using a pickup, it is possible to determine whether there is a fault in the audio system. By capturing snapshots, it is possible to determine whether there is a fault in the video system, thus greatly improving the operation and maintenance efficiency of the audio-visual system. By sending the status data to historical users, the fault location can be quickly located and identified, shortening the troubleshooting time and reducing the interruption time and resource waste caused by faults, and greatly improving the operation and maintenance efficiency of the audio-visual system. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 It is a flowchart of the method for automatically operating and maintaining an audio-visual system based on a large model provided by an embodiment of the present invention;

[0039] Figure 2 It is a first sub-flowchart of the method for automatically operating and maintaining an audio-visual system based on a large model provided by an embodiment of the present invention;

[0040] Figure 3 It is a second sub-flowchart of the method for automatically operating and maintaining an audio-visual system based on a large model provided by an embodiment of the present invention;

[0041] Figure 4 It is a third sub-flowchart of the method for automatically operating and maintaining an audio-visual system based on a large model provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0042] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0043] In Embodiment 1, Figure 1 The implementation process of the method for automatically operating and maintaining an audio-visual system based on a large model provided by an embodiment of the present invention is shown, and the details are as follows:

[0044] S100: Build an audio - video operation and maintenance platform, embed a pre - trained operation and maintenance large model, collect the usage records of audio - video devices, define historical users and current users, construct the usage scenarios of audio - video devices, locate the deployment positions of video surveillance devices, collect video surveillance data, and intercept the status data of audio - video devices, where the status data at least includes: status lights and system logs.

[0045] Create an audio - video operation and maintenance platform, connect all audio - video devices to the audio - video operation and maintenance platform. The audio - video operation and maintenance platform is mainly used to overall manage all audio - video devices, and can also perform operations such as status monitoring, fault diagnosis, and maintenance record of audio - video devices; Embed an operation and maintenance large model into the audio - video operation and maintenance platform. The operation and maintenance large model has strong data processing, anomaly detection, and fault prediction capabilities. It can read the operation data of audio - video devices from the audio - video operation and maintenance platform, identify potential risks, and determine solutions. It can also continuously optimize itself by learning new operation and maintenance data and user operations, gradually improving prediction accuracy and data processing capabilities.

[0046] Use the audio - video operation and maintenance platform to record the usage records of audio - video devices, define the users who have used audio - video devices before as historical users, and the users who are currently using audio - video devices as current users. It should be noted that the identification of user identities needs to use video surveillance data and combine face recognition technology; Determine the usage scenarios of audio - video devices, where the usage scenarios can be meeting rooms, classrooms, studios, or other scenarios; In the usage scenario, determine the deployment positions of video surveillance devices. The video surveillance devices are cameras. Combine the layout of the usage scenario to determine the installation position, angle, and coverage area of the cameras, etc.; Use the video surveillance devices to collect video surveillance data in the usage scenario and intercept the status data of audio - video devices. The status data refers to the data that can show the status of audio - video devices. For example, status lights and system logs, etc.; By analyzing the status data, abnormal situations can be quickly identified and the causes of faults can be traced.

[0047] S200: Via the audio - video operation and maintenance platform, receive the power - on instruction uploaded by the current user, start the self - inspection mechanism pre - integrated in the audio - video operation and maintenance platform, and determine whether there is a fault in the audio - video device. If so, traverse the preset fault query table to determine the disposal plan. If not, use the pick - up microphone pre - deployed in the usage scenario to collect audio data, extract human voices and sound effects, and determine whether they belong to the preset frequency range. If so, generate an audio system normal instruction and send it to the audio - video operation and maintenance platform. Intercept a snapshot containing the video playback device from the video surveillance data and determine whether there are abnormal features. If not, output a video system normal instruction.

[0048] After receiving the boot instruction uploaded by the current user, the audio-video operation and maintenance platform will use the self-check mechanism to check the status of audio-video devices, determine whether the functions of audio-video devices are normal. The specific functions include video output, audio input, network connection, and system load, etc. At the same time, it will also self-check whether the camera is working properly and whether the microphone is connected, etc. The self-check mechanism is the self-check program, which is pre-established by the audio-video device management personnel. The self-check results will be recorded in the form of system logs. If a fault is detected in the audio-video device (where the fault is stored in the system log in the form of a fault code), query the pre-established fault query table to determine the disposal plan corresponding to the fault code. The fault query table is pre-established by the audio-video device management personnel.

[0049] If it is found that the audio-video device can work normally after self-check, the audio data is collected by the pick-up microphone pre-deployed in the usage scenario. The pick-up microphone can be integrated in the video surveillance device. The pick-up microphone will collect the audio signals in the surrounding environment in real time, including human voices and sound effects. Using the audio signal processing algorithm, the collected audio data is separated to extract the spectral features of human voices and sound effects, and determine whether they belong to the preset frequency range.

[0050] When the sound of the current user is collected by the pick-up microphone, that is, the spectral features of the human voice are extracted. At this time, continue to use the pick-up microphone to determine whether the sound effect can be collected. If so, and both belong to the preset frequency range, it means that both the pick-up microphone and the sound system can work normally, generate an audio system normal instruction, and upload it to the audio-video operation and maintenance platform. If the pick-up microphone cannot collect the sound effect, it means that there is a fault in the pick-up microphone or the sound system.

[0051] Using a similar method, a snapshot of the video playback device is intercepted from the video surveillance data. The video playback device is a projector or a display, etc. If the intercepted snapshot contains abnormal features such as a black screen, a flashing screen, and other abnormal displays that cannot be displayed normally, it means that the video system is abnormal at this time. If no abnormal features are intercepted in the snapshot, generate a video system normal instruction and upload it to the audio-video operation and maintenance platform.

[0052] S300: Determine whether the audio-video operation and maintenance platform has received the normal instructions of the audio system and the video system. If so, send a boot instruction to the audio-video device. If not, send the status data to all historical users, subscribe to the feedback information sent by the historical users, and establish a dialogue.

[0053] If the audio - video operation and maintenance platform receives the above - mentioned normal instruction of the audio system and the normal instruction of the video system successively, it indicates that all audio - video devices are normal. Then, send a power - on instruction to the audio - video devices to start the audio - video devices. If the normal instruction of the audio system or the normal instruction of the video system is not received, or neither of them is received, it indicates that there is a fault in the audio system or the video system or both. Then, send the fault data to all online historical users in the audio - video operation and maintenance platform, and at the same time receive the feedback information sent by the historical users. The feedback information is the solution to the fault. Use video monitoring devices, pick - up microphones or mobile terminals, etc. to establish a dialogue between historical users and current users.

[0054] In Embodiment 2, Figure 2 The implementation process of the automated audio - video system operation and maintenance method based on a large model provided by the embodiment of the present invention is shown. The following details the steps of constructing an audio - video operation and maintenance platform, embedding a pre - trained operation and maintenance large model, collecting the usage records of audio - video devices, and defining historical users and current users, as follows:

[0055] S101: Embed a timestamp into the usage record to construct a behavior record table, where the behavior record table consists of a personnel item and a time item.

[0056] By embedding a timestamp into the usage record, the specific times of operations such as power - on, operation, troubleshooting, maintenance, and shutdown of audio - video devices can be recorded, and a behavior record table is integrated and generated. The behavior record table consists of a personnel item and a time item. The personnel item is determined by the video monitoring device through face recognition technology, and the time item is determined by the timestamp.

[0057] S102: Collect the usage data of the current user, where the usage data at least includes: startup time, running duration, and historical faults, and integrate the usage data into the behavior record table.

[0058] Use the system log to record the usage data of the current user and write the usage data into the behavior record table.

[0059] In Embodiment 3, Figure 2 The implementation process of the automated audio - video system operation and maintenance method based on a large model provided by the embodiment of the present invention is shown. The following details the steps of constructing the usage scenario of audio - video devices, locating the deployment position of video monitoring devices, and collecting video monitoring data, as follows:

[0060] S103: Configure the unique identifier of the audio - video device, and divide the video monitoring data into several blocks, where each audio - video device corresponds to one block.

[0061] Set a unique identifier for each audio - video device, and split the video surveillance data into multiple chunks. The chunks are used to display the real - time status of the corresponding audio - video device. The advantage of doing this is that it can monitor all audio - video devices, avoid monitoring dead spots, and at the same time improve the data traceability ability.

[0062] S104: Establish the correspondence between the chunks, audio - video devices, and unique identifiers.

[0063] According to the correspondence, label the unique identifier to the corresponding chunk.

[0064] In Embodiment 4, Figure 2 The implementation process of the automated audio - video system operation and maintenance method based on the large model provided by the embodiment of the present invention is shown. The following details the step of extracting the status data of the audio - video device, where the status data at least includes: status lights and system logs, as follows:

[0065] S105: Input the status data into the operation and maintenance large model, output the potential hazard features, and send the potential hazard features to a preset terminal.

[0066] Input the status data of the audio - video device into the operation and maintenance large model. Through deep learning and data analysis, identify the potential hazard features of the audio - video device. The potential hazard features can be hardware failures, unstable connections, system anomalies, high temperatures, etc. Send the potential hazard features to the audio - video device management personnel for timely handling of the potential hazard features.

[0067] S106: Based on the potential hazard features, construct an adjustment mechanism for the status data.

[0068] When the potential hazard features are detected, use the adjustment mechanism to adjust the status data; where the adjustment mechanism can be: adjust the status light to red.

[0069] In Embodiment 5, different from Embodiment 1, in the embodiment of the present invention, the method further includes:

[0070] Based on the usage data, record the adjustment frequency of each audio - video device;

[0071] Based on the adjustment frequency, cluster the audio - video devices into several priorities, and correct the self - inspection mechanism in the order from high to low priority.

[0072] According to the usage data, record the operations of the current user on each audio - video device, where the operations include: increasing or decreasing the volume, software update, device connection, etc.; calculate the adjustment frequency of each audio - video device; for example, in a certain use, the current user adjusted the speaker volume 15 times within 3 hours, then the adjustment frequency of the speaker is 5 times per hour.

[0073] Determine the priority of each audio - video device according to the adjustment frequency. The higher the adjustment frequency, the higher the priority of the audio - video device. Adjust the self - inspection order, self - inspection frequency, etc. in the order from high to low priority. It should be noted that the self - inspection mechanism can also be started when the audio - video device is not in use.

[0074] In Embodiment 6, Figure 3 The implementation process of the automated audio - video system operation and maintenance method based on a large model provided by the embodiment of the present invention is shown. The steps of collecting audio data and extracting human voices and sound effects are described in detail as follows:

[0075] S201: Input the audio data into the operation and maintenance large model to extract human voices and sound effects.

[0076] S202: Configure evaluation indicators for the audio data. The evaluation indicators at least include: distortion item, echo item, and noise item. Determine whether there are defect items in the audio data. If so, send the defect items to the historical user and generate an audio system exception instruction.

[0077] Determine the evaluation indicators for the audio data; that is, evaluate the audio data from multiple aspects such as distortion, echo, and noise. If it is found that the audio data has distortion, the defect item is distortion. Send the evaluation result and the corresponding audio data to the historical user, and the historical user judges how to process the audio device according to the defect item. When a defect item is detected, generate an audio system exception instruction and upload it to the audio - video operation and maintenance platform.

[0078] In Embodiment 7, Figure 3 The implementation process of the automated audio - video system operation and maintenance method based on a large model provided by the embodiment of the present invention is shown. The steps of intercepting a snapshot containing a video playback device from the video surveillance data, judging whether there are abnormal features, and if not, outputting a video system normal instruction are described in detail as follows:

[0079] S203: Construct an abnormal feature set, compare the snapshot with the abnormal feature set, and judge whether there are abnormal features. If so, integrate the abnormal features into the status data.

[0080] Construct an abnormal feature set, extract the image data containing the video device from the snapshot, and judge whether there are abnormal features in the image data. If there are, integrate the abnormal features into the status data. The advantage of doing this is to enable rapid response to faults.

[0081] In Embodiment 8, Figure 4The implementation process of the automated audio - video system operation and maintenance method based on a large model provided by an embodiment of the present invention is shown. The steps of subscribing to the feedback information sent by historical users and establishing a dialogue are described in detail as follows:

[0082] S301: Extract the processing steps from the feedback information.

[0083] Read the processing steps from the feedback information sent by historical users, where the processing steps are the specific methods for handling faults.

[0084] S302: Identify the behavior characteristics from the video monitoring data and verify the processing steps.

[0085] Use a video monitoring device to record the process of the current user handling the fault, select the behavior characteristics from it, and determine whether the behavior characteristics are the same as the processing steps. If they are the same, it means that the current user handles the fault according to the prompts of the historical user. If they are not the same, use a speaker to send a reminder voice to the current user, informing that there is a deviation between the actual operation and the processing steps.

[0086] In Embodiment 9, different from Embodiment 1, in the embodiment of the present invention, the method further includes:

[0087] Create an operation and maintenance community and integrate the operation and maintenance community into the audio - video operation and maintenance platform;

[0088] Publish the status data to the operation and maintenance community, select the best reply, and integrate the best reply into the feedback information.

[0089] In the audio - video operation and maintenance platform, create an operation and maintenance community where current users can publish usage problems, fault phenomena, and other relevant feedback about audio - video devices. After the current user publishes the status data of the audio - video device to the operation and maintenance community, community members can provide replies with various possible solutions based on actual experience or technical background. Other community members can like and comment on the solutions. Define the reply with the most likes as the best reply, integrate the best reply into the feedback information, and send it to the current user.

[0090] In Embodiment 10, different from Embodiment 1, in the embodiment of the present invention, the method further includes:

[0091] Open the usage permissions of the video monitoring device and the pick - up microphone to the historical user, build a control link, and establish a dialogue;

[0092] Create a dialogue path and send an access request to the current user and the historical user.

[0093] Historical users establish a conversation with current users through video surveillance devices and pick-up microphones. In other words, both historical users and current users are connected to the conversation path, which can be a video surveillance device or a pick-up microphone.

[0094] For example, historical users establish a voice call with current users through a video surveillance device or the speakers and pick-up microphones in the usage scenario. At the same time, historical users can also monitor the real-time status of audio and video devices through the video surveillance device.

[0095] The technical features of the above-described embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0096] The above-described embodiments only represent several implementation manners of the present invention. The description is relatively specific and detailed, but it should not be construed as a limitation on the scope of the patent of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several modifications and improvements can still be made, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention should be subject to the appended claims.

[0097] The above is only the preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent replacements, and improvements made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. An automated audio and video system operation and maintenance method based on a large model, characterized in that: The method comprises: Build an audio and video operation and maintenance platform, and embed a pre-trained operation and maintenance large model, collect the usage records of audio and video equipment, define historical users and current users, build usage scenarios of audio and video equipment, and locate the deployment location of video surveillance equipment, collect video surveillance data, and intercept the status data of audio and video equipment, where the status data at least includes: status lights and system logs; The power-on command uploaded by the current user is received via the audio and video operation and maintenance platform, and the self-check mechanism pre-integrated in the audio and video operation and maintenance platform is started to determine whether the audio and video equipment has a fault. If so, the preset fault query table is traversed to determine the disposal plan. If not, the audio data is collected by using the microphone pre-deployed in the usage scenario, and the human voice and the sound are extracted to determine whether the two belong to the preset frequency range. If so, a normal audio system instruction is generated and sent to the audio and video operation and maintenance platform. A snapshot containing the video playback device is intercepted from the video monitoring data to determine whether there are abnormal features. If not, a normal video system instruction is output; Determine whether the audio and video operation and maintenance platform receives normal instructions from the audio system and the video system. If so, issue a power-on instruction to the audio and video equipment. If not, send the status data to all historical users, subscribe to feedback information sent by historical users, and establish a dialogue.

2. The large model-based automated audio and video system operation and maintenance method according to claim 1, characterized in that: The steps of constructing an audio and video operation and maintenance platform, embedding a pre-trained operation and maintenance large model, collecting usage records of audio and video equipment, and defining historical users and current users include: Embed a timestamp into the usage record to construct a behavior record table, wherein the behavior record table consists of a person item and a time item; Collect usage data of the current user, wherein the usage data at least includes: startup time, running time and historical failures, and integrate the usage data into the behavior record table.

3. The large model-based automated audio and video system operation and maintenance method according to claim 1, characterized in that: The steps of constructing the usage scenario of the audio and video equipment, locating the deployment location of the video surveillance equipment, and collecting the video surveillance data include: Configuring a unique identifier of the audio and video device, and dividing the video surveillance data into a plurality of blocks, wherein each audio and video device corresponds to a block; Establish the correspondence between the blocks, audio and video devices and unique identifiers.

4. The large model-based automated audio and video system operation and maintenance method according to claim 1, characterized in that: The step of intercepting the status data of the audio and video device, wherein the status data at least includes: a status light and a system log comprises: Input the state data into the operation and maintenance big model, output hidden danger features, and send the hidden danger features to a preset terminal; Based on the hidden danger characteristics, a mechanism for adjusting status data is constructed.

5. The large model-based automated audio and video system operation and maintenance method according to claim 2, characterized in that: The method further comprises: Based on the usage data, record the adjustment frequency of each audio and video device; Based on the adjustment frequency, the audio and video devices are clustered into a number of priorities, and the self-check mechanism is modified in descending order of priority.

6. The large model-based automated audio and video system operation and maintenance method according to claim 1, characterized in that: The steps of collecting audio data and extracting human voice and sound include: Input the audio data into the operation and maintenance big model to extract the human voice and the sound; An evaluation index for the audio data is configured, wherein the evaluation index includes at least a distortion item, an echo item, and a noise item, and it is determined whether there is a defect item in the audio data. If so, the defect item is sent to a historical user, and an audio system abnormality instruction is generated.

7. The large model-based automated audio and video system operation and maintenance method according to claim 1, characterized in that: The step of extracting a snapshot of the video playback device from the video surveillance data and determining whether there are abnormal features, and if not, outputting a normal instruction of the video system comprises: Construct an abnormal feature set, compare the snapshot with the abnormal feature set, determine whether abnormal features exist, and if so, integrate the abnormal features into the state data.

8. The large model-based automated audio and video system operation and maintenance method according to claim 1, characterized in that: The steps of subscribing to the feedback information sent by the historical user and establishing a dialogue include: Extracting processing steps from the feedback information; From the video surveillance data, behavioral features are identified and the processing steps are verified.

9. The large model-based automated audio and video system operation and maintenance method according to claim 1, characterized in that: The method further comprises: Creating an operation and maintenance community, and integrating the operation and maintenance community into the audio and video operation and maintenance platform; The status data is published to the operation and maintenance community, the best response is selected, and the best response is integrated into the feedback information.

10. The large model-based automated audio and video system operation and maintenance method according to claim 1, characterized in that: The method further comprises: Open the use rights of video surveillance equipment and microphones to the historical users, build a control link, and establish a dialogue; A dialog path is created, and access requests are sent to the current user and the historical user.

Images