Cross-platform digital human behavior synchronization method and system based on AIGC and medium

Abstract

The invention belongs to the technical field of data processing, and particularly relates to an AIGC-based cross-platform digital human behavior synchronization method and system and a medium, and the method comprises the steps: obtaining interaction data with a user, obtaining a user demand, generating answer information based on the user demand by using an AIGC technology, determining a behavior action based on the answer information, and applying the behavior action to a digital human. The method comprises the following steps: collecting and storing related data of a digital human, naming a platform with changed related data as a first target platform, continuously monitoring the related data of the first target platform, starting timing after monitoring that the related data of the first target platform stops changing, and after a preset time interval, starting timing; the method comprises the following steps: performing synchronization processing on related data of a first target platform, acquiring difference data and dividing the difference data into a plurality of data blocks before the synchronization processing is performed, setting a temporary storage area, and sending the data blocks through the temporary storage area. According to the invention, the digital person with better experience feeling can be provided, and the digital person can keep behavior synchronization on a plurality of platforms.

Description

Technical Field

[0001] The present invention belongs to the field of data processing technology, and specifically relates to a cross-platform digital human behavior synchronization method, system and medium based on AIGC. Background Art

[0002] With the development and application of digital human technology, users have put forward higher requirements for the interactive experience of digital humans. Existing digital human behavior synchronization technology is often limited to a single platform, and it is difficult to achieve consistency and real-time performance across platforms. In addition, the demand for personalized content generation is also growing, which is difficult to meet with existing technologies. A similar prior art Chinese patent application with publication number CN111639137A provides a cross-platform data synchronization and exchange method, system, device and medium, wherein the method obtains matching first platform information according to preset synchronization parameters; obtains first commodity information from the first platform according to the first platform information; obtains matching standard commodity information from a preset standard database according to the obtained first commodity information; generates an inventory commodity database according to the standard commodity information; obtains preset push parameters, wherein the preset push parameters include second platform information and commodity information to be pushed; obtains push commodity information from the inventory commodity database according to the commodity information to be pushed; and sends the push commodity information to the second platform. Similar prior art also has a Chinese patent application with publication number CN117633107A, which discloses a cross-platform heterogeneous data synchronization method and device, the method comprising: after the transaction data is updated to the database of the host platform and synchronized to the database of the distributed platform, comparing the daily incremental data of the database of the distributed platform and the database of the host platform to obtain a first comparison result; reading and comparing the full data of the database of the distributed platform and the database of the host platform to obtain a second comparison result; comparing the message data of the message queue MQ of the database of the distributed platform and the database of the host platform synchronized daily to obtain a third comparison result; determining the difference data between the database of the distributed platform and the database of the host platform according to the first comparison result, the second comparison result, and the third comparison result; re-synchronizing the difference data between the database of the distributed platform and the database of the host platform through a batch backup mechanism. However, the above two methods do not consider the problem of how to improve the synchronization efficiency when the amount of synchronized data is large when realizing data synchronization. Therefore, the present invention provides a cross-platform digital human behavior synchronization method, system and medium based on AIGC. Summary of the invention

[0003] Cross-platform digital human behavior synchronization is mainly reflected in the ability to provide a coherent and consistent user experience and improve service efficiency on different platforms. To achieve the above goals, when implementing cross-platform synchronization, digital humans may generate a large amount of different data between different platforms, which greatly reduces the efficiency of data synchronization.

[0004] In order to solve the above problems, the present invention provides the following cross-platform digital human behavior synchronization method based on AIGC, which is implemented by executing the following steps:

[0005] Acquire interaction data with the user, pre-generate a user analysis model, use the user analysis model to analyze the interaction data to obtain user needs, use AIGC technology to generate answer information based on the user needs, determine behavioral actions based on the answer information, apply the behavioral actions to the digital human, and have the digital human present them to the user;

[0006] Collect and save the relevant data of the digital human, the relevant data including the interaction data, model data, user personalized data and corresponding user ID, the interaction data refers to the conversation information between the user and the digital human from the beginning of the conversation to the current time, the model data includes the user analysis model and model version number, and the user personalized data refers to the user's feature data;

[0007] Monitor the relevant data of each platform in real time, and when the relevant data of a certain platform is monitored to change, determine the change mode of the relevant data, and when the change mode is an active change, refer to the platform where the relevant data changes as a first target platform, continuously monitor the relevant data of the first target platform, and start timing after monitoring that the relevant data of the first target platform stops changing;

[0008] After a preset time interval, synchronization processing is performed on the relevant data of the first target platform. Before performing the synchronization processing, difference data between the relevant data in the first target platform and other platforms is obtained, the difference data is divided into multiple data blocks, a temporary storage area is set on the first target platform, the data blocks are stored in the temporary storage area, and a sending operation is performed, wherein the sending operation refers to sending the data blocks in the temporary storage area to other platforms.

[0009] As a preferred technical solution of the present invention, sending the data blocks in the temporary storage area to other platforms includes the following steps:

[0010] Analyze the first target user who causes the change in the relevant data based on the difference data, obtain the relevant data of the first target user, analyze the behavior pattern of the first target user based on the relevant data, obtain the second target platform of the first target user based on the behavior pattern, send the data blocks in the temporary storage area to the second target platform, delete the corresponding data block in the temporary storage area each time a data block is sent, and add the remaining data block that has not been sent to the temporary storage area, until all the data blocks are sent to the second target platform.

[0011] As a preferred technical solution of the present invention, after all the data blocks are sent to the second target platform, the following steps are further performed:

[0012] A temporary storage area is set on the second target platform, and the third target platform and the fourth target platform of the first target user are obtained based on the behavior pattern, and the first communication duration between the first target platform and the third target platform is obtained, and the second communication duration between the first target platform and the fourth target platform is also obtained, and the first communication duration is compared with the second communication duration. When the first communication duration is less than the second communication duration, the first target platform sends the data block to the third target platform, and the second target platform sends the data block to the fourth target platform; otherwise, the first target platform sends the data block to the fourth target platform, and the second target platform sends the data block to the third target platform, and then repeats this step, and all platforms that have received the data block send the data block to platforms that have not received the data block, until all platforms receive the data block.

[0013] As a preferred technical solution of the present invention, after all the data blocks are sent to the second target platform, the following steps are further performed:

[0014] The second target platform merges all the received data blocks to obtain the difference data. The second target platform determines whether the relevant data of itself has changed. If no active change occurs, a synchronization operation is performed. The synchronization operation refers to merging the difference data into the relevant data of the second target platform. If a change occurs, the synchronization operation is performed after the relevant data stops changing. During the execution of the synchronization operation, the change mode of the relevant data is set to passive change.

[0015] As a preferred technical solution of the present invention, while sending the data blocks in the temporary storage area to other platforms, the following steps are also performed:

[0016] Continuously monitor the relevant data of the first target platform, suspend the sending operation when the relevant data actively changes again, start timing when it is monitored that the relevant data stops actively changing, analyze the second target user who causes the relevant data to actively change, if the second target user is different from the first target user, after a preset time interval, update the difference data and continue to perform the sending operation, updating the difference data means obtaining new difference data between the current relevant data and the relevant data when performing the synchronization operation, adding the new difference data to the difference data, if the second target user is the same as the target user, increase the duration of the preset time interval, after the increased time interval, update the difference data and continue to perform the sending operation.

[0017] As a preferred technical solution of the present invention, generating a user analysis model includes the following steps:

[0018] The user personalized data is collected, the user personalized data and the interaction data are integrated to extract user features therefrom, key features are extracted from the user features using a feature selection algorithm, the key features are encoded and converted into a format that is convenient for processing by a machine learning algorithm, the integrated user personalized data, the interaction data and the key features are used as training data, and a user analysis model is generated based on the training data using a machine learning algorithm for training.

[0019] As a preferred technical solution of the present invention, after generating the user analysis model, the following steps are further performed:

[0020] The user analysis model is used to analyze user needs during the actual interaction process with the user, and the accuracy of the user analysis model is verified through the interaction data with the user. New training data is generated based on the new interaction data, and the user analysis model is optimized based on the new training data.

[0021] As a preferred technical solution of the present invention, the user personalized data includes user gender, user age, user location information, weather information when the user interacts with the digital human, and date information when the user interacts with the digital human.

[0022] The present invention also provides a cross-platform digital human behavior synchronization system based on AIGC, including the following modules:

[0023] The digital human module is used to obtain interaction data with the user, pre-generate a user analysis model, use the user analysis model to analyze the interaction data to obtain user needs, use AIGC technology to generate answer information based on the user needs, determine behavioral actions based on the answer information, apply the behavioral actions to the digital human, and the digital human presents them to the user;

[0024] A collection module, used to collect and save relevant data of the digital human, wherein the relevant data includes the interaction data, model data, user personalized data and corresponding user ID, wherein the interaction data refers to the conversation information between the user and the digital human from the beginning of the conversation to the current time, the model data includes the user analysis model and the model version number, and the user personalized data refers to the user's feature data;

[0025] A monitoring module is used to monitor the relevant data of each platform in real time, and when the relevant data of a certain platform is monitored to change, determine the change mode of the relevant data, and when the change mode is an active change, refer to the platform where the relevant data changes as a first target platform, continuously monitor the relevant data of the first target platform, and start timing after the relevant data of the first target platform stops changing;

[0026] A synchronization module is used to perform synchronization processing on the relevant data of the first target platform after a preset time interval, and before performing the synchronization processing, obtain the difference data between the relevant data in the first target platform and other platforms, divide the difference data into multiple data blocks, set a temporary storage area on the first target platform, store the data blocks in the temporary storage area, and perform a sending operation, which refers to sending the data blocks in the temporary storage area to other platforms.

[0027] The present invention also provides a medium storing program instructions, wherein when the program instructions are executed, the device where the medium is located is controlled to execute any one of the above-mentioned AIGC-based cross-platform digital human behavior synchronization methods.

[0028] Compared with the prior art, the beneficial effects of the present invention are at least as follows:

[0029] The present invention provides personalized services for users through digital humans, improves the user experience, synchronizes relevant data to other platforms in order to synchronize the behaviors of digital humans on various platforms, and synchronizes relevant data regularly by monitoring the changes of relevant data during synchronization, thereby reducing the synchronization frequency. When performing synchronization processing, only the difference data of relevant data between platforms is synchronized, thereby reducing the communication load and improving the synchronization efficiency. Finally, during the synchronization process, the difference data is divided into blocks, so that even if the amount of difference data is very large, the data can be quickly synchronized. After that, the difference data is sent through a temporary storage area to improve the communication efficiency and further improve the data synchronization efficiency. Through the above method steps, the digital human can provide users with unified, efficient, accurate and user-friendly services on multiple platforms. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 It is a flowchart of the steps of the cross-platform digital human behavior synchronization method based on AIGC of the present invention;

[0031] Figure 2 This is a structural diagram of the cross-platform digital human behavior synchronization system based on AIGC of the present invention. DETAILED DESCRIPTION

[0032] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with 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 intended to limit the present invention.

[0033] It is understood that the terms "first", "second", etc. used in this application may be used herein to describe various elements, but unless otherwise specified, these elements are not limited by these terms. These terms are only used to distinguish a first element from another element. For example, without departing from the scope of this application, a first xx script may be referred to as a second xx script, and similarly, a second xx script may be referred to as a first xx script.

[0034] The present invention provides Figure 1 The cross-platform digital human behavior synchronization method based on AIGC is mainly implemented by executing the following steps:

[0035] Step S1, obtain the interaction data with the user, pre-generate a user analysis model, use the user analysis model to analyze the interaction data to obtain user needs, use AIGC technology to generate answer information based on user needs, determine behavioral actions based on the answer information, apply the behavioral actions to the digital human, and have the digital human present them to the user.

[0036] Specifically, in order to provide a more intelligent digital human, by acquiring the user's interaction data, a user analysis model is generated based on the user's personalized information and interaction data combined with artificial intelligence algorithms and machine learning to analyze user needs. User needs refer to the user's input intentions. Personalized information includes date information, location information, weather information, etc. Based on these relevant information, the user's input intention in a specific situation can be more accurately inferred, and answer information can be generated based on the input intention using AIGC technology. Based on the answer information, the digital human's behavioral actions are designed and customized to produce a digital human that adapts to the user's preferences and the current dialogue scenario, providing users with a more accurate and natural interaction experience.

[0037] Step S2, collect and save the relevant data of the digital human, which includes interaction data, model data, user personalized data and corresponding user ID. The interaction data refers to the conversation information between the user and the digital human from the beginning of the conversation to the current time. The model data includes the user analysis model and the model version number. The user personalized data refers to the user's feature data.

[0038] Specifically, in order to achieve a high degree of consistency in the behavior of digital humans on different platforms and improve user experience, relevant data about digital humans are collected and saved, and the relevant data are deployed on different platforms. The relevant data include interaction data, model data, user personalized data and corresponding user IDs. The corresponding relevant data is saved for each user. The interaction data will continue to change with the interaction between each user and the digital human. The updated user analysis model based on the interaction data will also be continuously optimized. The interaction data, model data and user personalized data are collected, and based on the above data, the user's input intention can be analyzed more accurately.

[0039] Step S3, monitor the relevant data of each platform in real time. When the relevant data of a certain platform is monitored to change, determine the change mode of the relevant data. When the change mode is an active change, the platform where the relevant data changes is called the first target platform. Continuously monitor the relevant data of the first target platform, and start timing after the relevant data of the first target platform stops changing.

[0040] Specifically, in order to be able to use digital human services on multiple platforms, it is necessary to deploy the relevant data of digital humans on each platform. For example, when a user interacts with the first target platform, the corresponding interaction data will be updated, the model will be optimized, and personalized data may also be increased. Since the behavior of digital humans is determined based on the analysis of relevant data, in order to keep the behavior of digital humans on each platform synchronized, it is necessary to keep the relevant information of digital humans synchronized on each platform. Therefore, the relevant data of each platform is monitored in real time. When the relevant data of the first target platform is monitored to change, the changed relevant data can be synchronized to other platforms. There are two change modes for the change of relevant data. Active change refers to the change caused by the interaction between the user and the digital human, and passive change refers to the change caused by the platform where the active change occurs synchronizing data to the platform where the change does not occur. In the case of active change, if the generated difference data is directly synchronized to other platforms, if there are continuous changes in the future, it will lead to frequent synchronization processing, reducing the efficiency of synchronization. Therefore, after monitoring that the relevant data stops changing for a period of time, it is judged that the relevant data may not continue to change, and then synchronization processing is performed.

[0041] Step S4: after a preset time interval, perform synchronization processing on the relevant data of the first target platform. Before performing the synchronization processing, obtain the difference data between the relevant data in the first target platform and other platforms, divide the difference data into multiple data blocks, set a temporary storage area on the first target platform, store the data blocks in the temporary storage area, and perform a sending operation, which refers to sending the data blocks in the temporary storage area to other platforms.

[0042] Specifically, in order to quickly synchronize the relevant data of each platform so that the digital people on each platform can always keep their behaviors synchronized, after a preset time interval, synchronization processing is performed on the relevant data of the first target platform. The synchronization processing includes two steps: sending operation and synchronization operation. The sending operation is performed by the data sending party, and the synchronization operation is performed by the data receiving party. The difference data of the relevant data of the first target platform and other platforms are obtained. When performing synchronization processing, only the difference data is processed instead of all relevant data, which reduces the communication load. A temporary storage area is set on the first target platform. Although the storage space of the temporary storage area is much smaller than the storage space of the memory, the input and output speed is usually greater than the input and output speed of the target platform memory. Therefore, by dividing the difference data into multiple data blocks, the multiple data blocks are sent to other platforms in turn through the temporary storage area, thereby improving the efficiency of synchronization. The specific process of sending data blocks to other platforms will be explained in detail later.

[0043] The digital human is used to provide personalized services for users and improve the user experience. In order to synchronize the behaviors of the digital human on various platforms, the relevant data is synchronized to other platforms. During the synchronization process, the relevant data is synchronized regularly by monitoring the changes in the relevant data to reduce the synchronization frequency. During the synchronization process, only the difference data of the relevant data between the platforms is synchronized to reduce the communication load and thus improve the synchronization efficiency. Finally, during the synchronization process, the difference data is divided into blocks so that even if the amount of the difference data is very large, the data can be synchronized quickly. After that, the difference data is sent through the temporary storage area to improve the communication efficiency and further improve the efficiency of data synchronization. Through the above method steps, the digital human can provide users with unified, efficient, accurate and user-friendly services on multiple platforms.

[0044] Furthermore, sending the data blocks in the temporary storage area to other platforms includes the following steps:

[0045] Based on the difference data analysis, the first target user whose related data has changed is analyzed, the related data of the first target user is obtained, the behavior pattern of the first target user is analyzed based on the related data, and the second target platform of the first target user is obtained based on the behavior pattern, and the data blocks in the temporary storage area are sent to the second target platform. After each data block is sent, the corresponding data block in the temporary storage area is deleted, and the remaining data block that has not been sent is added to the temporary storage area until all data blocks are sent to the second target platform.

[0046] Specifically, in order to enable users to always experience the digital human service with synchronized behavior, when sending data blocks to other platforms, the first target user whose relevant data has changed is analyzed based on the difference data. That is to say, the first target user has a conversation with the digital human on the first target platform, which causes the relevant data of the first target platform to change. At this time, synchronization processing needs to be performed to other platforms. If the difference data is sent to other platforms at the same time, it may cause network congestion. In order to avoid network congestion and keep the behavior of the digital human synchronized, the relevant data of the first target user is analyzed and the behavior pattern is obtained. The behavior pattern refers to the order of platforms that the user is accustomed to using. For example, after the user interacts with the digital human on the first target platform, he is accustomed to interacting with the digital human on the second target platform. Therefore, the second target platform is used as the first platform for synchronization processing, and all data blocks are sent to the second target platform through the temporary storage area.

[0047] Furthermore, after all data blocks are sent to the second target platform, the following steps are also performed:

[0048] A temporary storage area is set on the second target platform, and the third target platform and the fourth target platform of the first target user are obtained based on the behavior pattern, and the first communication duration of the first target platform and the third target platform is obtained, and the second communication duration of the first target platform and the fourth target platform is also obtained, and the first communication duration and the second communication duration are compared. When the first communication duration is less than the second communication duration, the first target platform sends the data block to the third target platform, and the second target platform sends the data block to the fourth target platform; otherwise, the first target platform sends the data block to the fourth target platform, and the second target platform sends the data block to the third target platform, and then repeats this step, and all platforms that have received the data block send the data block to the platform that has not received the data block, until all platforms receive the data block.

[0049] Specifically, synchronizing data to other platforms only through the first target platform will reduce the synchronization efficiency. In order to improve the overall synchronization efficiency, after the second target platform receives the data block, the first target platform and the second target platform simultaneously send the data block to other platforms. Therefore, the third target platform and the fourth target platform are obtained based on the behavior pattern, and then the best communication method is obtained through the above method. The data block is sent to the third target platform and the fourth target platform through the best communication method. Then, this step is repeated, and all platforms that have received the difference data send the difference data to the platform that has not received the difference data until all platforms have received the data block. As the number of platforms that receive the data blocks increases, the number of senders also increases. As the number of senders increases, the efficiency of data synchronization also increases. By comparing the communication time, the sender and receiver with relatively short communication time are selected to transmit the data block, thereby further improving the efficiency of data synchronization.

[0050] Furthermore, after all data blocks are sent to the second target platform, the following steps are also performed:

[0051] The second target platform merges all the received data blocks to obtain differential data. The second target platform determines whether its own related data has changed. If no active change occurs, a synchronization operation is performed. The synchronization operation refers to merging the differential data into the related data of the second target platform. If a change occurs, the synchronization operation is performed after the related data stops changing. During the synchronization operation, the change mode of the related data is set to passive change.

[0052] Specifically, after receiving all the data blocks, the second target platform merges the data blocks to obtain differential data. At this time, it is necessary to synchronize the differential data with its own related data. However, if the related data of the second target platform is changing at this time, it means that the second target platform may be interacting with the user. At this time, if the differential data is merged into the changing related data, it may cause errors in the synchronization of the related data of the second target platform. Therefore, it is necessary to wait for the related data to stop changing before performing the synchronization operation. If the related data of the second target platform has not changed, the differential data can be immediately merged into the related data. The above method is used to synchronize the related data of the second target platform with the related data of the first target platform. In the process of performing the synchronization operation, the change mode of the related data is also set to passive change, so that it can be easily distinguished when monitoring the change mode of the related data.

[0053] Furthermore, while sending the data blocks in the temporary storage area to other platforms, the following steps are also performed:

[0054] Continuously monitor the relevant data of the first target platform, and suspend the sending operation when the relevant data changes actively again. Start timing when it is monitored that the relevant data stops changing actively, analyze the second target user who causes the relevant data to change actively, and if the second target user is different from the first target user, update the difference data and continue to execute the sending operation after a preset time interval. Updating the difference data means obtaining new difference data between the current relevant data and the relevant data when performing the synchronization operation, and adding the new difference data to the difference data. If the second target user is the same as the target user, increase the length of the preset time interval, and after the increased time interval, update the difference data and continue to execute the sending operation.

[0055] Specifically, in order to maintain a smooth experience of interaction between users and digital humans, while sending data blocks to other platforms, the relevant data of the first target platform is continuously monitored. If the relevant data changes actively again, it means that during the process of sending data blocks, another user interacted with the digital human of the first target platform. At this time, if the sending operation is continued, the first target platform must maintain a smooth interaction between the digital human and the user, and send data blocks, which may increase the CPU load of the first target platform, which may slow down the interaction between the user and the digital human, and may also reduce the efficiency of sending data blocks. Therefore, in order to reduce the load of the first target platform, the sending operation is suspended, and the first target platform focuses on the interaction between the digital human and the user. At the same time, it also analyzes the second target user who causes the active change of the relevant data. If the second target user is the same as the first target user, the preset time interval is increased during the next synchronization process to reduce the frequency of synchronization processing.

[0056] Furthermore, generating a user analysis model includes the following steps:

[0057] Collect user personalized data, integrate user personalized data and interaction data and extract user features from them, use feature selection algorithm to extract key features from user features, and encode and convert the key features into a format that is convenient for machine learning algorithm to process. Use the integrated user personalized data, interaction data and key features as training data, and use machine learning algorithm to train and generate user analysis model based on the training data.

[0058] Specifically, in order to obtain answer information that better meets user needs, personalized user data is collected. Usually, the amount of information collected in personalized user data is large, and it is necessary to integrate the personalized user collection and user interaction data and extract user features from them, and use a feature selection algorithm to extract key features from the user features. The feature selection algorithm can use existing technologies such as the chi-square test or correlation coefficient. The key features refer to user features that are highly correlated with user needs. The key features are also encoded and converted to a format that is convenient for machine learning algorithms to process. After that, the personalized data and key features are used as training data, and the machine learning algorithm is used to train the training data to generate a user analysis model.

[0059] Furthermore, after the user analysis model is generated, the following steps are also performed:

[0060] The user analysis model is used to analyze user needs during the actual interaction process with users, and the accuracy of the user analysis model is verified through interaction data with users. New training data is generated based on new interaction data, and the user analysis model is optimized based on the new training data.

[0061] Specifically, the user analysis model is continuously optimized through the above method to generate a user analysis model that can more accurately analyze user needs.

[0062] Furthermore, the user personalized data includes the user's gender, the user's age, the user's location information, the weather information when the user interacts with the digital human, and the date information when the user interacts with the digital human.

[0063] Specifically, consider adding the above personalized data to the training data, more fully consider the possible impact of various data, and generate a more accurate user analysis model.

[0064] According to another aspect of the embodiment of the present invention, Figure 2As shown, a cross-platform digital human behavior synchronization system based on AIGC is also provided, including a digital human module, a collection module, a monitoring module and a synchronization module, which are used to implement the cross-platform digital human behavior synchronization method based on AIGC as described above. The specific functions of each module are as follows:

[0065] The digital human module is used to obtain interaction data with users, pre-generate a user analysis model, use the user analysis model to analyze the interaction data to obtain user needs, use AIGC technology to generate answer information based on user needs, determine behavioral actions based on the answer information, apply the behavioral actions to the digital human, and have the digital human present them to the user;

[0066] The collection module is used to collect and save the relevant data of the digital human, including interaction data, model data, user personalized data and corresponding user ID. The interaction data refers to the conversation information between the user and the digital human from the beginning of the conversation to the current time. The model data includes the user analysis model and the model version number. The user personalized data refers to the user's feature data.

[0067] A monitoring module is used to monitor the relevant data of each platform in real time. When the relevant data of a certain platform is monitored to change, the change mode of the relevant data is determined. When the change mode is an active change, the platform whose relevant data has changed is called the first target platform. The relevant data of the first target platform is continuously monitored, and the timing is started after the relevant data of the first target platform stops changing;

[0068] The synchronization module is used to perform synchronization processing on the relevant data of the first target platform after a preset time interval. Before performing the synchronization processing, the difference data between the relevant data in the first target platform and other platforms is obtained, the difference data is divided into multiple data blocks, a temporary storage area is set on the first target platform, the data blocks are stored in the temporary storage area, and a sending operation is performed, which refers to sending the data blocks in the temporary storage area to other platforms.

[0069] According to another aspect of an embodiment of the present invention, a medium is further provided, wherein the medium stores program instructions, wherein when the program instructions are executed, the device where the medium is located is controlled to execute any one of the above-mentioned cross-platform digital human behavior synchronization methods based on AIGC.

[0070] In summary, the present invention provides a cross-platform digital human behavior synchronization method, system and medium based on AIGC, the method includes obtaining interaction data with users, obtaining user needs, using AIGC technology to generate answer information based on user needs, determining behavioral actions based on the answer information, applying the behavioral actions to digital humans, collecting and saving relevant data of digital humans, calling the platform where the relevant data changes as the first target platform, continuously monitoring the relevant data of the first target platform, starting timing after monitoring that the relevant data of the first target platform stops changing, performing synchronization processing on the relevant data of the first target platform after a preset time interval, obtaining difference data and dividing it into multiple data blocks before performing synchronization processing, and setting a temporary storage area to send data blocks through the temporary storage area. The present invention can provide a digital human with a better experience and can also enable the digital human to maintain behavioral synchronization on multiple platforms.

[0071] It should be understood that, although each step in the flow chart of each embodiment of the present invention is shown in sequence according to the indication of the arrow, these steps are not necessarily performed in sequence according to the order indicated by the arrow. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be performed in other orders. Moreover, at least a portion of the steps in each embodiment may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these sub-steps or stages is not necessarily performed in sequence, but can be performed in turn or alternately with at least a portion of other steps or sub-steps or stages of other steps.

[0072] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the above-mentioned program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0073] The technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above-mentioned embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0074] The above-mentioned embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

[0075] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A cross-platform digital human behavior synchronization method based on AIGC, characterized in that: The steps include: Acquire interaction data with the user, pre-generate a user analysis model, use the user analysis model to analyze the interaction data to obtain user needs, use AIGC technology to generate answer information based on the user needs, determine behavioral actions based on the answer information, apply the behavioral actions to the digital human, and have the digital human present them to the user; Collect and save the relevant data of the digital human, the relevant data including the interaction data, model data, user personalized data and corresponding user ID, the interaction data refers to the conversation information between the user and the digital human from the beginning of the conversation to the current time, the model data includes the user analysis model and model version number, and the user personalized data refers to the user's feature data; Monitor the relevant data of each platform in real time, and when the relevant data of a certain platform is monitored to change, determine the change mode of the relevant data, and when the change mode is an active change, refer to the platform where the relevant data changes as a first target platform, continuously monitor the relevant data of the first target platform, and start timing after monitoring that the relevant data of the first target platform stops changing; After a preset time interval, synchronization processing is performed on the relevant data of the first target platform. Before performing the synchronization processing, difference data between the relevant data in the first target platform and other platforms is obtained, the difference data is divided into multiple data blocks, a temporary storage area is set on the first target platform, the data blocks are stored in the temporary storage area, and a sending operation is performed, wherein the sending operation refers to sending the data blocks in the temporary storage area to other platforms.

2. The method according to claim 1, characterized in that: Sending the data block in the temporary storage area to other platforms comprises the following steps: Analyze the first target user who causes the change in the relevant data based on the difference data, obtain the relevant data of the first target user, analyze the behavior pattern of the first target user based on the relevant data, obtain the second target platform of the first target user based on the behavior pattern, send the data blocks in the temporary storage area to the second target platform, delete the corresponding data block in the temporary storage area each time a data block is sent, and add the remaining data block that has not been sent to the temporary storage area, until all the data blocks are sent to the second target platform.

3. The method according to claim 2, characterized in that After all the data blocks are sent to the second target platform, the following steps are further performed: A temporary storage area is set on the second target platform, and the third target platform and the fourth target platform of the first target user are obtained based on the behavior pattern, and the first communication duration between the first target platform and the third target platform is obtained, and the second communication duration between the first target platform and the fourth target platform is also obtained, and the first communication duration is compared with the second communication duration. When the first communication duration is less than the second communication duration, the first target platform sends the data block to the third target platform, and the second target platform sends the data block to the fourth target platform; otherwise, the first target platform sends the data block to the fourth target platform, and the second target platform sends the data block to the third target platform, and then repeats this step, and all platforms that have received the data block send the data block to platforms that have not received the data block, until all platforms receive the data block.

4. The method according to claim 2, characterized in that: After all the data blocks are sent to the second target platform, the following steps are further performed: The second target platform merges all the received data blocks to obtain the difference data. The second target platform determines whether the relevant data of itself has changed. If no active change occurs, a synchronization operation is performed. The synchronization operation refers to merging the difference data into the relevant data of the second target platform. If a change occurs, the synchronization operation is performed after the relevant data stops changing. During the execution of the synchronization operation, the change mode of the relevant data is set to passive change.

5. The method according to claim 1, characterized in that While sending the data blocks in the temporary storage area to other platforms, the following steps are also performed: Continuously monitor the relevant data of the first target platform, and suspend the sending operation when the relevant data actively changes again. Start timing when it is monitored that the relevant data stops actively changing, analyze the second target user who causes the relevant data to actively change, and if the second target user is different from the first target user, update the difference data and continue to perform the sending operation after a preset time interval. Updating the difference data refers to obtaining new difference data between the current relevant data and the relevant data during the synchronization operation, and adding the new difference data to the difference data. If the second target user is the same as the target user, increase the duration of the preset time interval, and after the increased time interval, update the difference data and continue to perform the sending operation.

6. The method according to claim 1, characterized in that Generating a user analysis model includes the following steps: The user personalized data is collected, the user personalized data and the interaction data are integrated to extract user features therefrom, key features are extracted from the user features using a feature selection algorithm, the key features are encoded and converted into a format that is convenient for processing by a machine learning algorithm, the integrated user personalized data, the interaction data and the key features are used as training data, and a user analysis model is generated based on the training data using a machine learning algorithm for training.

7. The method according to claim 6, characterized in that After generating the user analysis model, the following steps are also performed: The user analysis model is used to analyze user needs during the actual interaction process with the user, and the accuracy of the user analysis model is verified through the interaction data with the user. New training data is generated based on the new interaction data, and the user analysis model is optimized based on the new training data.

8. The method according to claim 6, characterized in that The user personalized data includes user gender, user age, user location information, weather information when the user interacts with the digital human, and date information when the user interacts with the digital human.

9. A cross-platform digital human behavior synchronization system based on AIGC, used to implement a cross-platform digital human behavior synchronization method based on AIGC as claimed in any one of claims 1 to 8, characterized in that: Includes the following modules: The digital human module is used to obtain interaction data with the user, pre-generate a user analysis model, use the user analysis model to analyze the interaction data to obtain user needs, use AIGC technology to generate answer information based on the user needs, determine behavioral actions based on the answer information, apply the behavioral actions to the digital human, and the digital human presents them to the user; A collection module, used to collect and save relevant data of the digital human, wherein the relevant data includes the interaction data, model data, user personalized data and corresponding user ID, wherein the interaction data refers to the conversation information between the user and the digital human from the beginning of the conversation to the current time, the model data includes the user analysis model and the model version number, and the user personalized data refers to the user's feature data; A monitoring module is used to monitor the relevant data of each platform in real time, and when the relevant data of a certain platform is monitored to change, determine the change mode of the relevant data, and when the change mode is an active change, refer to the platform where the relevant data changes as a first target platform, continuously monitor the relevant data of the first target platform, and start timing after the relevant data of the first target platform stops changing; A synchronization module is used to perform synchronization processing on the relevant data of the first target platform after a preset time interval, and before performing the synchronization processing, obtain the difference data between the relevant data in the first target platform and other platforms, divide the difference data into multiple data blocks, set a temporary storage area on the first target platform, store the data blocks in the temporary storage area, and perform a sending operation, which refers to sending the data blocks in the temporary storage area to other platforms.

10. A medium, characterized in that The medium stores program instructions, wherein when the program instructions are executed, the device where the medium is located is controlled to execute the cross-platform digital human behavior synchronization method based on AIGC as described in any one of claims 1-8.

Images