A digital human marketing interaction method and system based on data processing

Abstract

The application provides a digital human marketing interaction method and system based on data processing, and belongs to the technical field of digital human marketing interaction. The method comprises the following steps: in response to detecting that a first operation of a first user on a man-machine interaction terminal does not meet expectations, displaying a digital human assisted marketing mode; in response to the first user activating the digital human assisted marketing mode, acquiring first operation data corresponding to the first operation of the first user on the man-machine interaction terminal; based on the first operation data, generating a first assisted marketing digital human for the first user, wherein the first assisted marketing digital human provides first assisted operation data for the first user; and when second operation data meets an association condition with the first operation data, reusing the first assisted marketing digital human for a second user. The technical scheme of the application can accelerate the efficiency of digital human generation and reproduction under the condition of a limited resource pool, and the timing meets user expectations.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of digital human marketing interaction, and particularly relates to a digital human marketing interaction method and system based on data processing, a computer readable storage medium for implementing the method, a computer program product, and an electronic device. BACKGROUND

[0002] A "digital human" refers to a virtual human that imitates humans in appearance, behavior, thinking, and emotion by relying on artificial intelligence technology. It has three main characteristics: (1) similar appearance, gender, personality, and other characteristics to humans; (2) can convey human behavior through language, expressions, and actions; (3) has the ability to recognize the external world and communicate and interact with people. The current market classification of digital humans mainly includes identity-type digital humans and service-type digital humans. Identity-type digital humans mainly include two types: virtual IP idols and virtual avatars of humans, and they focus more on creativity and aesthetics, and are widely used in the entertainment industry. Service-type digital humans have certain functionality and interactivity, and can replace real people to provide services in various application scenarios, with high automation, such as virtual anchors, virtual customer service, virtual assistants, etc.

[0003] Chinese invention patent application CN202210918660 proposes an intelligent digital human marketing interaction method, which can interact with users from multiple modalities, provide more rich visual, auditory, text, and chart information, and improve the final marketing experience and conversion rate. Chinese invention patent application CN202110175602 proposes a personalized generation method for virtual digital human appearance, sound effect, and service model, which analyzes the collected data to characterize customers, generates personalized virtual digital human appearance, sound effect, and service according to customer characteristics, and the appearance and interactive sound are personalized, meeting customer preferences, with thousands of faces for thousands of people, deep personalized service, and improved customer service experience.

[0004] When generating a multi-modal personalized experience digital human, the client or server needs to provide sufficient resource pool support on the local or cloud end, including hardware resources and software resources. As the number of customer groups expands, the number of concurrent users experiencing or using digital human services also increases significantly. How to continue to provide reliable digital human services under the support of limited local hardware resources and / or budget-constrained cloud available resources has become a new challenge. In addition, the determination of the intervention timing of the digital human in the marketing process is also a problem to be considered. SUMMARY

[0005] To solve the above technical problems, the present application proposes a digital human marketing interaction method and system based on data processing, a computer readable storage medium for implementing the method, a computer program product, and an electronic device.

[0006] In the first aspect of the present application, a digital human marketing interaction method based on data processing is proposed, which is realized based on a human-computer interaction terminal and includes the following steps:

[0007] In response to detecting that the first operation of the first user on the human-computer interaction terminal does not meet the expectation, a digital human assisted marketing mode is displayed.

[0008] In response to the first user activating the digital human assisted marketing mode, first operation data corresponding to the first operation of the first user on the human-computer interaction terminal is obtained.

[0009] Based on the first operation data, a first assisted marketing digital human is generated for the first user, which provides first assisted operation data for the first user.

[0010] After generating the assisted marketing digital human for the first user based on the first operation data, the method further includes:

[0011] At least one modification component is provided for the first user to modify the first assisted marketing digital human.

[0012] In response to detecting that the second operation of the second user on the human-computer interaction terminal does not meet the expectation, a digital human assisted marketing mode is displayed.

[0013] In response to the second user activating the digital human assisted marketing mode, second operation data corresponding to the second operation of the second user on the human-computer interaction terminal is obtained.

[0014] When the second operation data and the first operation data meet the association condition, the first assisted marketing digital human is reused for the second user.

[0015] The reuse of the first assisted marketing digital human for the second user specifically includes:

[0016] The first assisted marketing digital human is cloned for the second user to obtain a second assisted marketing digital human, which provides second assisted operation data for the second user.

[0017] The first assisted marketing digital human provides first assisted operation data for the first user, specifically including:

[0018] The first assisted marketing digital human provides a plurality of first assisted interaction channels, including at least one of a voice interaction channel, a visual interaction channel, or a text interaction channel.

[0019] Through the assisted interaction channel, the first assisted marketing digital human provides first assisted operation data for the first user.

[0020] The first auxiliary marketing digital person is reused for the second user, specifically including:

[0021] A first target auxiliary interaction channel that the first auxiliary marketing digital person currently activates is determined when the first user performs an interaction;

[0022] The first auxiliary marketing digital person is cloned for the second user to obtain a second auxiliary marketing digital person, and the second auxiliary marketing digital person provides available second auxiliary interaction channels, and the available second auxiliary interaction channels do not include the first target auxiliary interaction channel.

[0023] In a second aspect of the present application, in order to implement the method of the first aspect, a digital person marketing interaction system based on data processing is provided, and the system includes:

[0024] A user operation detection module is configured to obtain operation data of a user within a preset time period;

[0025] A user expectation value pre-judgment module is configured to judge whether the operation of the current user is in line with expectations based on the operation data obtained by the user operation detection module;

[0026] A digital person auxiliary marketing mode display module is configured to display a digital person auxiliary marketing mode when the operation of the current user is not in line with expectations;

[0027] A digital person auxiliary marketing mode activation module is configured to activate the digital person auxiliary marketing mode by the user;

[0028] An auxiliary marketing digital person generation module is configured to generate an auxiliary marketing digital person for the current user based on the operation data obtained by the user operation detection module, and the auxiliary marketing digital person provides auxiliary operation data for the current user;

[0029] An auxiliary marketing digital person modification module is configured to modify the auxiliary marketing digital person after the auxiliary marketing digital person generation module generates the auxiliary marketing digital person for the current user;

[0030] The modification includes adjusting an auxiliary interaction channel mode currently used by the auxiliary marketing digital person, and the auxiliary interaction channel mode includes at least one of a voice interaction channel, a visual interaction channel, or a text interaction channel.

[0031] The system further includes an auxiliary marketing digital person reuse module;

[0032] In a case where the system has generated a first auxiliary marketing digital person for the first user based on first operation data of the first user, the auxiliary marketing digital person reuse module performs a cloning operation on the first auxiliary marketing digital person to obtain a second auxiliary marketing digital person, and the second auxiliary marketing digital person provides auxiliary operation data for a second user.

[0033] The system is connected to a plurality of resource pools, and the resource pools provide resource support for generation of the auxiliary marketing digital person; each resource pool includes a plurality of channel resources corresponding to the auxiliary interaction channel mode.

[0034] The auxiliary marketing digital person reuse module performs a cloning operation on the first auxiliary marketing digital person to obtain a second auxiliary marketing digital person, and the cloning operation specifically includes:

[0035] Determining a target auxiliary interaction channel mode currently activated by the first auxiliary marketing digital person;

[0036] Determining channel resources currently available for the cloning operation based on the target auxiliary interaction channel mode;

[0037] Performing the cloning operation based on the channel resources currently available for the cloning operation.

[0038] The auxiliary marketing digital person reuse module performs a cloning operation on the first auxiliary marketing digital person to obtain a second auxiliary marketing digital person, and the cloning operation specifically includes:

[0039] Determining a first target auxiliary interaction channel currently activated by the first auxiliary marketing digital person;

[0040] Cloning the first auxiliary marketing digital person to obtain a second auxiliary marketing digital person for the second user, and the second auxiliary marketing digital person provides a second available auxiliary interaction channel, and the second available auxiliary interaction channel does not include the first target auxiliary interaction channel.

[0041] The aforementioned data processing-based digital person marketing interaction method can be implemented through various forms of electronic devices and computer program instructions automatically.

[0042] Therefore, in a third aspect of the present application, a computer-readable storage medium is also provided for storing computer instructions, which, when executed on an electronic device, cause the electronic device to perform the first aspect of the data processing-based digital person marketing interaction method.

[0043] In a fourth aspect of the present application, an electronic device is also proposed, which comprises a processor and a memory, the memory is used to store instructions, and the processor is used to call the instructions in the memory, so that the electronic device executes the data processing-based digital human marketing interaction method of the aforementioned first aspect.

[0044] In a fifth aspect of the present application, a computer program product is also proposed, which comprises a computer program, when the computer program is executed, the data processing-based digital human marketing interaction method of the aforementioned first aspect is implemented.

[0045] The technical solution of the present application displays the digital human assisted marketing mode only when it is detected that the operation of the user on the human-computer interaction terminal does not meet the expectation, avoids the process of the digital human assisted marketing mode from being resident, and only generates the assisted marketing digital human when the user activates the digital human assisted marketing mode, so that the intervention time of the digital human is more targeted. Finally, in the case that the system has generated a first assisted marketing digital human for a first user based on first operation data of the first user, the first assisted marketing digital human is subjected to a cloning operation by using the assisted marketing digital human reuse module to obtain a second assisted marketing digital human, the existing resource pool state can be fully reused, the processes of reactivation, calling, and allocation of the resource pool are avoided, the efficiency of the generation and reproduction of the digital human can be accelerated under the condition of the limited resource pool, and the occurrence time meets the expectation of the user.

[0046] Further advantages of the present application will be further embodied in detail in the specific embodiment part in combination with the drawings of the specification. BRIEF DESCRIPTION OF DRAWINGS

[0047] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments will be briefly introduced below. Obviously, the drawings in the following description only constitute some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without any creative effort on the basis of these drawings.

[0048] Figure 1 is a main flowchart of a data processing-based digital human marketing interaction method according to an embodiment of the present application;

[0049] Figure 2 is a main flowchart of a data processing-based digital human marketing interaction method according to a preferred embodiment of the present application;

[0050] Figure 3 is a schematic diagram of the assisted marketing digital humans generated in the technical solution of the present application under different scenarios;

[0051] Figure 4is a schematic diagram of a function button for a user to modify a currently generated auxiliary marketing digital person;

[0052] Figure 5 is a schematic diagram of a function structure of a digital person marketing interaction system based on data processing according to an embodiment of the present application. DETAILED DESCRIPTION

[0053] First of all, it needs to be pointed out that, Figure 1 or Figure 2 The embodiment of the digital person marketing interaction method based on data processing shown in the figure can be implemented on an electronic device, system configured with a memory and a processor through a computer program. The electronic device, system can be a physical machine, a virtual machine, a server, a cluster, or any combination thereof.

[0054] Preferably, the specific form of the electronic device can also be a human-computer interaction terminal. The human-computer interaction terminal can be a desktop terminal, a smart handheld terminal, a mobile terminal, etc. with a human-computer interaction interface, such as a desktop PC, a portable PDA, a smart phone, a vehicle-mounted terminal, etc.

[0055] The subsequent embodiments of the present application take a human-computer interaction terminal as an example for introduction. The human-computer interaction terminal is preferably a smart phone terminal or a vehicle-mounted terminal. The smart phone terminal or the vehicle-mounted terminal has a human-computer interaction interface. At least one marketing APP is configured on the human-computer interaction interface. The marketing APP includes at least one of a shopping APP, a financial APP, a social APP, or an entertainment APP, etc. When the marketing APP is running, the digital person marketing interaction method based on data processing provided by the present application can be called to provide digital person auxiliary marketing services for the current user in a timely manner.

[0056] In the prior art, the digital person auxiliary marketing service can provide services in various application scenarios for users with high automation, such as virtual voice anchors, virtual voice / video customer service, virtual shopping, financial management, social assistants, etc.

[0057] However, when generating a multi-modal personalized experience digital person, the client APP or the server needs to provide sufficient resource pool support on the local or cloud end. These resource pools include hardware resources and software resources. With the expansion of the customer group, the number of concurrent users experiencing or using the digital person service also increases significantly. How to continue to provide reliable digital person services under the support of limited local hardware resources and / or budget-constrained cloud available resources has become a new challenge. In addition, the determination of the intervention time of the digital person in the marketing process is also a problem that needs to be considered.

[0058] The present application proposes multiple embodiments to gradually solve the above technical problems. It should be noted that the present application subsequently gives multiple embodiments, each of which can constitute an independent technical solution and can contribute to the prior art and solve the corresponding technical problems. However, it should be pointed out that different embodiments can be combined with each other without violating the logic; at the same time, each embodiment can at least solve one technical problem mentioned in the background art, but it does not require each individual embodiment to solve multiple or all technical problems.

[0059] Firstly refer to Figure 1 , Figure 1 The main flow diagram of a digital human marketing interaction method based on data processing is shown, which is one embodiment of the present application. For the convenience of description, the subsequent description will give each step a step number (the relevant number is omitted in the related drawings).

[0060] Figure 1 The method embodiment includes the following steps S100-S300:

[0061] S100: in response to detecting that the first operation of the first user on the human-computer interaction terminal does not meet the expectation, displaying a digital human assisted marketing mode;

[0062] S200: in response to the first user activating the digital human assisted marketing mode, obtaining first operation data corresponding to the first operation of the first user on the human-computer interaction terminal;

[0063] S300: based on the first operation data, generating a first assisted marketing digital human for the first user, the first assisted marketing digital human providing first assisted operation data for the first user.

[0064] Next, the specific preferred implementation of each step of the method embodiment will be specifically introduced. Figure 1

[0065] Firstly, step S100: in response to detecting that the first operation of the first user on the human-computer interaction terminal does not meet the expectation, displaying a digital human assisted marketing mode.

[0066] In specific implementation, the step S100 includes the following sub-steps:

[0067] S110: obtaining first operation data of the first user on the human-computer interaction terminal within a preset time period;

[0068] As an example, the first operation data can be page operation data of the user within a preset time period (for example, within 5 minutes) after opening the marketing APP, including input operation, selection operation, closing operation, etc. ​

[0069] For example, for a shopping APP, the relevant page operation data can be searching for product names, sliding product page details, clicking to purchase / return to continue searching / close the current page to continue inputting search product names, etc.

[0070] For example, for a shopping APP, the relevant page operation data can be searching for product names, sliding product page details, clicking to purchase / return to continue searching / close the current page to continue inputting search product names, etc.

[0071] S120: determining whether the current first operation of the first user is expected based on the first operation data;

[0072] If not, the digital human assisted marketing mode is displayed.

[0073] If so, the digital human assisted marketing mode is not displayed.

[0074] Step S120 determines whether the current first operation of the first user is expected, mainly to determine the intervention timing of the digital human assisted marketing mode.

[0075] And determining whether the current first operation of the first user is expected means determining whether the user quickly hits the function he wants and completes the operation he wants after opening the marketing APP within a predetermined time period (for example, within 5 minutes); if the user often repeats the operation and finally does not complete the corresponding operation within the predetermined time period (for example, within 5 minutes), it means that the current first operation of the first user is likely not expected.

[0076] For example, for a shopping APP, the relevant page operation data can be searching for product names, sliding product page details, clicking to purchase / return to continue searching / close the current page to continue inputting search product names, etc.

[0077] If the user repeatedly performs the operation of "searching for product names-sliding product page details-returning to continue searching" within a predetermined time period, it means that the user is likely to have not hit the product he wants, i.e., the current first operation of the first user is not expected; and if the user finally performs the operation of "clicking to purchase-closing the page", it means that the user is likely to have hit the product he wants, i.e., the current first operation of the first user is expected.

[0078] For example, for a shopping APP, the relevant page operation data can be searching for product names, sliding product page details, clicking to purchase / return to continue searching / close the current page to continue inputting search product names, etc.

[0079] If a user repeatedly performs the action of "searching for financial product codes - scrolling through page details - returning to continue searching" within the predetermined time period, it means that the user is very likely not finding the financial product they want, i.e., the user's first action does not meet expectations. However, if the user eventually performs actions such as "clicking on risk test / clicking on purchase", it means that the user is very likely finding the financial product they want, i.e., the user's first action meets expectations.

[0080] Therefore, in a specific implementation, step S120 further includes:

[0081] Determine whether there is any page operation data in the first operation data with a repetition rate exceeding a threshold. If so, the first user's current first operation does not meet expectations.

[0082] The embodiment of the present invention enters the process of displaying the digital human-assisted marketing mode only when the first user's current first operation does not meet expectations.

[0083] Unlike existing digital human-assisted marketing models that display information online at all times, Figure 1 The embodiments of the present invention only enter the process of displaying the digital human assisted marketing mode when the first user's current first operation does not meet expectations, thus avoiding the phenomenon of the digital human generation process being resident and saving resource pool, especially system process resources, from the source.

[0084] After step S100 detects that the first user's first operation on the human-computer interaction terminal does not meet expectations, the digital human-assisted marketing mode is displayed. Figure 1 The method proceeds to step S200: in response to the first user activating the digital human-assisted marketing mode, the method obtains the first operation data corresponding to the first operation of the first user on the human-computer interaction terminal.

[0085] It is worth noting that although the digital human-assisted marketing mode is displayed when step S100 detects that the first user's first operation on the human-computer interaction terminal does not meet expectations, the digital human generation process is not started immediately, but the user is given the option.

[0086] This is because, in some scenarios, users may not be comfortable with displaying a virtual digital avatar on the current interface and prefer to continue using traditional search assistance methods. To address this issue, Figure 1 The example first shows that a digital human-assisted marketing mode is currently available, but whether or not the mode is activated needs to be actively activated by the user.

[0087] When a user actively activates this mode, they need to agree to the relevant terms of use, including personal data collection and terms of use. At the same time, the collection, use and processing of the relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0088] At this time, if the user agrees to activate the digital human-assisted marketing mode, in response to the first user activating the digital human-assisted marketing mode, the first operation data corresponding to the first user's first operation on the human-computer interaction terminal is obtained.

[0089] As an example, the first operation data includes not only the page operation data of the user within a preset time period (e.g., within 5 minutes) after opening the marketing APP, as mentioned above, but also other necessary data collection, such as user registration information and browsing history, in order to more accurately generate a digital human image for the current user.

[0090] Based on this, proceed to step S300: Based on the first operation data, generate a first auxiliary marketing digital human for the first user, and the first auxiliary marketing digital human provides the first auxiliary operation data to the first user.

[0091] It is understood that generating personalized digital humans for users based on known user data falls within the scope of existing technology (such as the relevant literature mentioned in the background art), and the embodiments of the present invention will not elaborate on this in detail, nor is this the focus of the present invention.

[0092] Figure 3 The schematic diagrams of the auxiliary marketing digital humans generated in different scenarios in the technical solution of the present invention have been given, including 8 different auxiliary marketing digital humans A, B, C, D, X, Y, Z and W.

[0093] The key difference between the above embodiments of the present invention and the prior art lies in the determination of the timing of digital human intervention to save system resources. In particular, the digital human-assisted marketing mode is only displayed when the user's operation on the human-computer interaction terminal is detected to be inconsistent with expectations, thus avoiding the process of the digital human-assisted marketing mode being constantly running and fundamentally saving system process resources.

[0094] Furthermore, in step S300, the first auxiliary marketing digital person provides the first auxiliary operation data to the first user, specifically including:

[0095] The first auxiliary marketing digital human provides multiple first auxiliary interaction channels, including at least one of voice interaction channels, visual interaction channels, or text interaction channels;

[0096] Through the auxiliary interaction channel, the first auxiliary marketing digital person provides the first auxiliary operation data to the first user.

[0097] Specifically, in an embodiment of the present invention, a resource pool consisting of a human-computer interaction terminal connected to a cloud server provides resource support for the generation of the assisted marketing digital human.

[0098] Specifically, each generation of an auxiliary marketing digital human corresponds to a parameter instantiation of the relevant resource pool.

[0099] Preferably, the marketing avatar can run in a virtual machine or container environment. Each marketing avatar corresponds to an independent virtual machine or container environment. The virtual machine or container has a basic resource template before instantiation. After obtaining the first operation data, the first operation data is filled into the resource template, thereby instantiating and generating the marketing avatar.

[0100] Each assisted marketing digital human consists of a basic kernel (which provides a basic reasoning model) and a facade interface. The facade interface is manifested in various assisted interaction channels.

[0101] Preferably, each generated auxiliary marketing digital human can display interfaces with voice interaction channels, visual interaction channels, and text interaction channels. However, in actual interaction with users, these three channels are not always activated simultaneously, depending on the availability of resources in the specific scenario.

[0102] As mentioned earlier, when a client or server generates a digital human with a multimodal personalized experience, it needs sufficient resource pool support from the local machine or the cloud. These resource pools include hardware resources and software resources.

[0103] Based on this, as a specific implementation, step S300 further includes:

[0104] S310: Obtain the current available resource status of the human-computer interaction terminal;

[0105] Currently available resources include audio channel resources (such as hardware resources like sound cards, microphones, and headphones, and software resources like voice decoders), current graphics and image interaction channel resources (such as hardware resources like GPUs and graphics cards, graphics and image decoder resources, and software resources like players), and input channel resources (hardware resources like keyboards and mice, and software resources like touch input).

[0106] S320: Based on the current available resources of the human-computer interaction terminal, the first auxiliary marketing digital human displays the currently available first auxiliary interaction channel;

[0107] In one scenario, the human-computer interaction terminal does not currently have video playback capabilities. Therefore, the first auxiliary interaction channel currently available is either a voice interaction channel or a text interaction channel; preferably, it is a voice interaction channel.

[0108] S330: Determine the auxiliary interaction channel currently used by the first auxiliary marketing digital human based on the user's selection;

[0109] When there are two or more optional first auxiliary interaction channels, the user can choose one of them; when there is only one optional first auxiliary interaction channel, the first auxiliary interaction channel is directly determined to be the auxiliary interaction channel currently used by the first auxiliary marketing digital human.

[0110] In the example above, the user selects the voice interaction channel;

[0111] S340: Through the currently used auxiliary interaction channel, the first auxiliary marketing digital person provides the first auxiliary operation data to the first user.

[0112] In other words, although the first auxiliary marketing digital human generated at this time displays three channel interfaces on the interface, namely voice interaction channel, visual interaction channel and text interaction channel, only the voice interaction channel and text interaction channel are currently available.

[0113] Visually, the visual interaction channel can be displayed as a dark gray state (unavailable), while the voice interaction channel and text interaction channel can be displayed as a highlighted state (available).

[0114] At this time, the first auxiliary marketing digital human outputs the first auxiliary operation data through voice interaction and displays it on the relevant input box interface or content details page of the human-computer interaction terminal.

[0115] The first auxiliary operation data is generated based on the user's expected operation on the current APP. Taking the aforementioned financial management APP as an example, the first auxiliary operation data can be based on the user's original input search keywords (financial product name, code, etc.), and expand to generate similar financial product names, codes, etc. (same risk level, same historical rate of return, same product portfolio allocation).

[0116] Of course, the first auxiliary marketing digital human provides different auxiliary operation data to the first user for different apps and different users' original operations, which cannot be listed in detail. Those skilled in the art can know that the auxiliary marketing digital human itself has self-learning ability and can make certain auxiliary decisions based on the actual situation.

[0117] Understandable. Figure 1The example scenario is actually the case where the client or server (cloud) generates the auxiliary marketing digital human for the first time.

[0118] When the client or server (cloud) already has a history of generating auxiliary marketing digital humans, this invention provides... Figure 2 The further preferred embodiments shown are as follows:

[0119] S100: In response to detecting that the first user's first operation on the human-computer interaction terminal is not as expected, display the digital human-assisted marketing mode;

[0120] S200: In response to the first user activating the digital human-assisted marketing mode, obtain the first operation data corresponding to the first operation of the first user on the human-computer interaction terminal;

[0121] S300: Based on the first operation data, generate a first auxiliary marketing digital human for the first user, and the first auxiliary marketing digital human provides the first auxiliary operation data to the first user.

[0122] S400: In response to detecting that a second user’s second operation on the human-computer interaction terminal is not as expected, display the digital human-assisted marketing mode;

[0123] S500: In response to the second user activating the digital human-assisted marketing mode, obtain the second operation data corresponding to the second operation of the second user on the human-computer interaction terminal;

[0124] S600: When the second operation data and the first operation data meet the association conditions, the first auxiliary marketing digital human is reused for the second user.

[0125] Specifically, the second operation data and the first operation data satisfy the association conditions, which may be: the similarity between the second operation data and the first operation data is greater than a preset threshold, the second operation data is a subset of the first operation data, the first operation data is a subset of the second operation data, etc.

[0126] exist Figure 2 In this embodiment, step S600, whereby the second user reuses the first auxiliary marketing digital human, specifically includes:

[0127] The second auxiliary marketing digital person is cloned to obtain a second auxiliary marketing digital person for the second user, and the second auxiliary marketing digital person provides the second user with second auxiliary operation data.

[0128] It is understandable that data cloning can directly produce a second auxiliary marketing digital human. This cloning method can be completed by directly migrating and copying the container group (POD) or virtual machine corresponding to the backend resource pool, without having to start from the initial state, making it faster and more efficient.

[0129] exist Figure 2 In the improved preferred embodiment, when the system has already generated a first auxiliary marketing digital human for the first user based on the first user's first operation data, the auxiliary marketing digital human reuse module is used to perform a cloning operation on the first auxiliary marketing digital human to obtain a second auxiliary marketing digital human. This can fully reuse the existing resource pool state, avoid the process of reactivating, calling, and allocating the resource pool, and accelerate the efficiency of digital human generation and reproduction under limited resource pool conditions.

[0130] In a more specific preferred embodiment, to further conserve system resources while avoiding resource conflicts, step S600 involves the second user reusing the first auxiliary marketing digital human, specifically including:

[0131] S610: Determine the first target auxiliary interaction channel that the first auxiliary marketing digital human interacting with the first user is currently activating;

[0132] S620: Cloning the first auxiliary marketing digital persona for the second user to obtain a second auxiliary marketing digital persona, the second auxiliary marketing digital persona provides an available second auxiliary interaction channel, the available second auxiliary interaction channel not including the first target auxiliary interaction channel.

[0133] To further explain the advantages of the above-mentioned improvement steps, the preceding text continues... Figure 1 Examples are described below.

[0134] In the foregoing Figure 1 In the embodiments, it has been noted that for the first auxiliary marketing digital human, the currently available first auxiliary interaction channel is either a voice interaction channel or a text interaction channel; the user ultimately selects the voice interaction channel.

[0135] In other words, the first target auxiliary interaction channel currently being activated by the first auxiliary marketing digital human interacting with the first user is determined to be the voice interaction channel.

[0136] If the first user and the second user concurrently use the same human-computer interaction terminal (e.g., a terminal with multiple inputs and multiple outputs), since the first auxiliary marketing digital human is currently activating the first target auxiliary interaction channel, which is the voice interaction channel, the second user clones the first auxiliary marketing digital human to obtain a second auxiliary marketing digital human. The second auxiliary marketing digital human provides an available second auxiliary interaction channel, which does not include the first target auxiliary interaction channel. That is, the second auxiliary marketing digital human provides an available second auxiliary interaction channel that does not include the voice interaction channel, in order to avoid resource conflicts and resource reallocation processes that consume system resources.

[0137] In another scenario, a first user uses a first human-computer interaction terminal, and a second user uses a second human-computer interaction terminal, but both have the same app open. This app provides digital human-assisted marketing services via a cloud server. In this case, the above approach can also avoid the conflict problem between similar cloud resources.

[0138] Furthermore, it should be noted that both the first and second auxiliary marketing avatars (as well as other batches of auxiliary marketing avatars generated and cloned each time) can display interfaces for voice interaction, visual interaction, and text interaction. However, in actual interaction with users, these three interfaces are not always activated simultaneously, depending on resource availability and potential resource conflicts in the specific scenario.

[0139] In the above scenario, the first auxiliary marketing digital human simultaneously displays three interaction channels on the interface: voice, visual, and text. However, only the voice and text interaction channels are currently available. Visually, the visual interaction channel can be displayed as grayed out (unavailable), while the voice and text interaction channels are highlighted (available). The second auxiliary marketing digital human also simultaneously displays the three interaction channels on the interface, but the voice interaction channel is currently unavailable. Visually, the voice interaction channel can be displayed as grayed out (unavailable), while the visual and text interaction channels are highlighted (available).

[0140] Combination Figures 1-2 To facilitate the computer in implementing the method flow described in this invention, a comprehensive embodiment is implemented as follows:

[0141] A digital human marketing interaction method based on data processing, the method being implemented using a human-computer interaction terminal, the method comprising the following steps:

[0142] Step 1: When the i-th operation of the i-th user on the human-computer interaction terminal is detected to be inconsistent with expectations, the digital human-assisted marketing mode is displayed;

[0143] Step 2: In response to the activation of the digital human-assisted marketing mode by the i-th user, obtain the i-th operation data corresponding to the i-th operation of the i-th user on the human-computer interaction terminal;

[0144] Step 3: Determine if there is a history of generating the j-th auxiliary marketing digital human for the j-th user. If there is, proceed to Step 4.

[0145] Otherwise, proceed to Step 7;

[0146] Step 4: Obtain the data of the j-th operation corresponding to the j-th operation of the j-th user when generating the j-th auxiliary marketing digital human;

[0147] Step 5: Determine whether the i-th operation data and the j-th operation data satisfy the association condition;

[0148] If so, proceed to Step 6;

[0149] Otherwise, proceed to Step 7;

[0150] Step 6: Reuse the j-th auxiliary marketing digital human for the i-th user;

[0151] Step 7: Based on the i-th operation data, generate the i-th auxiliary marketing digital human for the i-th user.

[0152] As can be seen, when the system has already generated the first auxiliary marketing digital human for the first user based on the first user's first operation data, the auxiliary marketing digital human reuse module performs a cloning operation on the first auxiliary marketing digital human to obtain the second auxiliary marketing digital human. This can fully reuse the existing resource pool state and avoid the processes of reactivating, calling, and allocating the resource pool. It can speed up the efficiency of digital human generation and reproduction under limited resource pool conditions. This is because simple operation data comparison (which can be quickly compared by comparing the segmented hash values ​​of the pre-calculated operation data) and data cloning (which can be directly copied and migrated through virtual machines or containers) operations require fewer process resources than regenerating the auxiliary marketing digital human from the beginning of resource allocation, and do not involve the complete reallocation of resources, making them faster and more efficient.

[0153] by Figure 3 The diagram shown illustrates different scenarios of assisted marketing digital humans generated in the example. Assisted marketing digital humans C, Y, and Z can all be cloned from an existing historically generated assisted marketing digital human B.

[0154] Of course, it can be seen that the marketing assisted digital humans B, C, Y, and Z are not completely "clones" but have certain differences and variations. However, the overall architecture is consistent (for example, they all simultaneously display three channel interfaces: voice interaction channel, visual interaction channel, and text interaction channel. However, each currently available channel interface needs to be specifically determined based on the resource configuration and / or user selection / modification operations in the actual scenario).

[0155] In practical applications, after generating an auxiliary marketing digital human for the first user based on the first operational data, the method further includes:

[0156] At least one modification component is provided, which is used by a first user to modify the first assisted marketing digital human.

[0157] like Figure 4 As shown, Figure 4 This is a diagram illustrating the function buttons for users to customize the currently generated digital marketing assistant.

[0158] The customization options include buttons for switching genders, interaction channels, hairstyles, and clothing. Of course, the actual user interface could include even more detailed adjustment modules. Figure 4 It is not fully shown in the text.

[0159] Based on this, when the second operation data and the first operation data meet the association conditions, reusing the first auxiliary marketing digital human for the second user further includes:

[0160] To obtain a second auxiliary marketing digital avatar, the first auxiliary marketing digital avatar is cloned for the second user. The second auxiliary marketing digital avatar is then modified using the modification component to meet the user's personalized needs. For example, auxiliary marketing digital avatar B' is cloned from auxiliary marketing digital avatar B, and then auxiliary marketing digital avatars C, Y, and Z are modified from B'.

[0161] exist Figures 1-2 Based on the given method implementation examples, see below. Figure 5 , Figure 5 This is a schematic diagram of the functional unit structure of a data processing-based digital human marketing interaction system according to an embodiment of the present invention.

[0162] Figure 5 Provides a data processing-based digital human marketing interaction system, the system comprising:

[0163] The user operation detection module is used to obtain user operation data within a preset time period;

[0164] The user expectation prediction module is used to determine whether the current user's operation meets expectations based on the operation data obtained by the user operation detection module.

[0165] The digital human-assisted marketing mode display module is used to display the digital human-assisted marketing mode when the current user's operation does not meet expectations.

[0166] The digital human-assisted marketing mode activation module is used for users to activate the digital human-assisted marketing mode; the digital human-assisted marketing generation module is used to generate a digital human-assisted marketing for the current user based on the operation data obtained by the user operation detection module, and the digital human-assisted marketing provides auxiliary operation data for the current user.

[0167] The auxiliary marketing digital human modification module is used to modify the auxiliary marketing digital human after the auxiliary marketing digital human generation module generates an auxiliary marketing digital human for the current user;

[0168] The modification includes adjusting the auxiliary interaction channel mode currently used by the auxiliary marketing digital human, the auxiliary interaction channel mode including at least one of voice interaction channel, visual interaction channel or text interaction channel.

[0169] The modifications also include gender switching, hairstyle switching, and clothing switching. For each modification, corresponding replaceable colors, hairstyles, and clothing are provided.

[0170] The system also includes a digital human reuse module for marketing assistance;

[0171] When the system has already generated a first auxiliary marketing digital persona for the first user based on the first user's first operation data, the auxiliary marketing digital persona reuse module performs a cloning operation on the first auxiliary marketing digital persona to obtain a second auxiliary marketing digital persona, and the second auxiliary marketing digital persona provides auxiliary operation data for the second user.

[0172] The system connects to multiple resource pools, which provide resource support for the generation of the assisted marketing digital human; each resource pool includes multiple channel resources, which correspond to the assisted interaction channel mode.

[0173] The auxiliary marketing digital human reuse module performs a cloning operation on the first auxiliary marketing digital human to obtain a second auxiliary marketing digital human, specifically including:

[0174] Determine the target auxiliary interaction channel mode currently activated by the first auxiliary marketing digital human;

[0175] The channel resources currently available for the cloning operation are determined based on the target auxiliary interaction channel mode;

[0176] The cloning operation is performed based on the channel resources currently available for the cloning operation.

[0177] The auxiliary marketing digital human reuse module performs a cloning operation on the first auxiliary marketing digital human to obtain a second auxiliary marketing digital human, specifically including:

[0178] Identify the first target-oriented auxiliary interaction channel that is currently being activated with the first auxiliary marketing digital human;

[0179] The second auxiliary marketing digital persona is obtained by cloning the first auxiliary marketing digital persona for the second user. The second auxiliary marketing digital persona provides a second auxiliary interaction channel, which does not include the first target auxiliary interaction channel.

[0180] The aforementioned data processing-based digital human marketing interaction method can be automatically implemented through various forms of electronic devices and computer program instructions; the computer program instructions can be stored in different forms of storage media and loaded into computer electronic devices for execution.

[0181] Therefore, further embodiments also provide a computer-readable storage medium for storing computer instructions that, when executed on an electronic device, cause the electronic device to perform... Figure 1 or Figure 2 The embodiment describes a digital human marketing interaction method based on data processing.

[0182] Further embodiments also propose a computer device including a processor and a memory, the memory for storing instructions, and the processor for calling the instructions in the memory to cause the computer device to perform... Figure 1 or Figure 2 The embodiment describes a digital human marketing interaction method based on data processing.

[0183] Further embodiments also propose a computer program product comprising a computer program that, when executed, performs the aforementioned... Figure 1 or Figure 2 A digital human marketing interaction method based on data processing is implemented according to the embodiment.

[0184] The technical solution of this application only displays the digital human-assisted marketing mode when it detects that the user's operation on the human-computer interaction terminal does not meet expectations, thus avoiding the process of the digital human-assisted marketing mode being constantly running. Furthermore, the digital human-assisted marketing mode is only generated when the user activates it, making the timing of the digital human's intervention more targeted. Finally, when the system has already generated a first digital human-assisted marketing mode for the first user based on the first user's first operation data, the digital human-assisted marketing mode reuse module is used to clone the first digital human-assisted marketing mode to obtain a second digital human-assisted marketing mode. This fully reuses the existing resource pool state, avoids the processes of reactivating, calling, and allocating the resource pool, and can accelerate the efficiency of digital human generation and reproduction under limited resource pool conditions, and the timing of its appearance meets the user's expectations.

[0185] In this application embodiment, the terms "module" or "unit" refer to a computer program or part of a computer program that has a predetermined function and works with other related parts to achieve a predetermined goal, and can be implemented wholly or partially using software, hardware (such as processing circuitry or memory), or a combination thereof. Similarly, a processor (or multiple processors or memory) can be used to implement one or more modules or units. Furthermore, each module or unit can be part of an overall module or unit that includes the functionality of that module or unit.

[0186] Furthermore, in specific embodiments of this application, if user-related data is involved, user permission or consent must be obtained when the embodiments of this application are applied to specific products or technologies, and the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0187] Other technologies, principles, algorithms, or models not elaborated in detail in this application can be found in the prior art. The prior art referenced in the embodiments is an integral part of the disclosure of this invention.

[0188] The foregoing has shown and described the method embodiments and systems of the present invention, but it will be understood by those skilled in the art that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A digital human marketing interaction method based on data processing, the method being implemented based on a human-computer interaction terminal, characterized in that, The method includes the following steps: Acquire the first user's first operation data on the human-computer interaction terminal within a preset time period; Determine whether there is page operation data in the first operation data with a repetition rate exceeding a threshold. If so, the first user's current first operation does not meet expectations. When the first user's first action on the human-computer interaction terminal is detected to be inconsistent with expectations, the digital human-assisted marketing mode is displayed; In response to the first user activating the digital human-assisted marketing mode, first operation data corresponding to the first user's first operation on the human-computer interaction terminal is obtained; Based on the first operation data, a first auxiliary marketing digital human is generated for the first user. The first auxiliary marketing digital human provides the first auxiliary operation data to the first user, specifically including: The first auxiliary marketing digital human provides multiple first auxiliary interaction channels, including at least one of voice interaction channels, visual interaction channels, or text interaction channels; Obtain the current available resource status of the human-computer interaction terminal; Based on the current available resources of the human-computer interaction terminal, the first auxiliary marketing digital human displays the currently available first auxiliary interaction channel; The auxiliary interaction channel currently used by the first auxiliary marketing digital human is determined based on the user's selection; Through the currently used auxiliary interaction channel, the first auxiliary marketing digital person provides the first auxiliary operation data to the first user.

2. The digital human marketing interaction method based on data processing as described in claim 1, characterized in that, After generating an auxiliary marketing digital human for the first user based on the first operational data, the method further includes: At least one modification component is provided, which is used by a first user to modify the first assisted marketing digital human.

3. The digital human marketing interaction method based on data processing as described in claim 1, characterized in that, After generating an auxiliary marketing digital human for the first user based on the first operational data, the method further includes: When a second user's second operation on the human-computer interaction terminal is detected to be inconsistent with expectations, the digital human-assisted marketing mode is displayed; In response to the second user activating the digital human-assisted marketing mode, the second operation data corresponding to the second operation of the second user on the human-computer interaction terminal is obtained; When the second operation data and the first operation data meet the association conditions, the first auxiliary marketing digital human is reused for the second user.

4. The digital human marketing interaction method based on data processing as described in claim 3, characterized in that, The reuse of the first auxiliary marketing digital human by the second user specifically includes: The second auxiliary marketing digital person is cloned to obtain a second auxiliary marketing digital person for the second user, and the second auxiliary marketing digital person provides the second user with second auxiliary operation data.

5. The digital human marketing interaction method based on data processing as described in claim 3, characterized in that, The reuse of the first auxiliary marketing digital human for a second user specifically includes: Determine the first target assisted interaction channel that the first assisted marketing digital human interacting with the first user is currently activating; The second user clones the first auxiliary marketing digital persona to obtain a second auxiliary marketing digital persona. The second auxiliary marketing digital persona provides a second auxiliary interaction channel, which does not include the first target auxiliary interaction channel.

6. A digital human marketing interaction system based on data processing, characterized in that, The system includes: The user operation detection module is used to obtain the first operation data of the first user on the human-computer interaction terminal within a preset time period. The user expectation prediction module is used to determine whether the current first user's operation meets expectations based on the first operation data obtained by the user operation detection module; if the current first user's operation has page operation data with a repetition rate exceeding a threshold, then the current first user's operation does not meet expectations. The digital human-assisted marketing mode display module is used to display the digital human-assisted marketing mode when the current first user's operation does not meet expectations; The digital human-assisted marketing mode activation module is used by the first user to activate the digital human-assisted marketing mode. The auxiliary marketing digital human generation module is used to generate a first auxiliary marketing digital human for the current first user based on the first operation data obtained by the user operation detection module, and the first auxiliary marketing digital human provides the current first user with first auxiliary operation data; The auxiliary marketing digital human modification module is used to modify the first auxiliary marketing digital human after the auxiliary marketing digital human generation module generates the first auxiliary marketing digital human for the current first user; The modification includes adjusting the auxiliary interaction channel mode currently used by the first auxiliary marketing digital human, the auxiliary interaction channel mode including at least one of voice interaction channel, visual interaction channel or text interaction channel; The first auxiliary marketing digital person provides the current first user with first auxiliary operation data, specifically including: Obtain the current available resource status of the human-computer interaction terminal; Based on the current available resources of the human-computer interaction terminal, the first auxiliary marketing digital human displays the currently available first auxiliary interaction channel; The auxiliary interaction channel currently used by the first auxiliary marketing digital human is determined based on the first user's selection. Through the currently used auxiliary interaction channel, the first auxiliary marketing digital person provides the first auxiliary operation data to the first user.

7. The digital human marketing interaction system based on data processing as described in claim 6, characterized in that, The system also includes a digital human reuse module for marketing assistance; When the system has already generated a first auxiliary marketing digital persona for the first user based on the first user's first operation data, the auxiliary marketing digital persona reuse module performs a cloning operation on the first auxiliary marketing digital persona to obtain a second auxiliary marketing digital persona, and the second auxiliary marketing digital persona provides auxiliary operation data for the second user.

8. The digital human marketing interaction system based on data processing as described in claim 7, characterized in that, The system connects to multiple resource pools, which provide resource support for the generation of the assisted marketing digital human; each resource pool includes multiple channel resources, which correspond to the assisted interaction channel mode. The auxiliary marketing digital human reuse module performs a cloning operation on the first auxiliary marketing digital human to obtain a second auxiliary marketing digital human, specifically including: Determine the target auxiliary interaction channel mode currently activated by the first auxiliary marketing digital human; The channel resources currently available for the cloning operation are determined based on the target auxiliary interaction channel mode; The cloning operation is performed based on the channel resources currently available for the cloning operation.

9. A digital human marketing interaction system based on data processing as described in claim 7, characterized in that, The auxiliary marketing digital human reuse module performs a cloning operation on the first auxiliary marketing digital human to obtain a second auxiliary marketing digital human, specifically including: Identify the first target-oriented auxiliary interaction channel that is currently being activated with the first auxiliary marketing digital human; The second user clones the first auxiliary marketing digital persona to obtain a second auxiliary marketing digital persona. The second auxiliary marketing digital persona provides a second auxiliary interaction channel, which does not include the first target auxiliary interaction channel.

Images