Virtual digital human calling method and device across application scenarios

Abstract

This invention relates to a method and apparatus for invoking a virtual digital human across application scenarios. The target terminal is configured with a Windows program containing multiple .exe processes. These processes share a single window for integrated display. Each .exe process includes both virtual human software and external software processes, enabling the virtual human software to combine with the external software for cross-application scenario virtual digital human invocation. The target terminal also has a background configuration page for splitting the user interface of the virtual human software and external software. This splitting can be either vertical or horizontal. The target terminal also configures the positional ratio and screen resolution of the virtual human software and external software within the background configuration page. The self-service registration machine software has already integrated with the hospital's HIS system; the virtual human software only needs to interface with the self-service registration machine software and does not require repeated integration with the hospital's underlying interface.

Description

Technical Field

[0001] This invention relates to the field of software technology, and in particular to a method and apparatus for invoking virtual digital humans across application scenarios. Background Technology

[0002] AI 3D virtual human software and the all-in-one machine equipped with it are often positioned as a universal solution for all application scenarios. However, this vision is difficult to realize in actual commercialization. AI 3D virtual human software and the all-in-one machine equipped with it need to be deeply integrated with existing hardware and software systems in different application scenarios, rather than simply being added on top of each other. For example, hospital outpatient halls already have self-service registration machines, community health centers are equipped with integrated physical examination machines, government service halls (taxation, social security) deploy queuing and calling machines, and bank branches have differentiated equipment such as ATMs, VTMs, and STMs.

[0003] However, in the healthcare sector, self-service kiosks in outpatient halls have already been implemented, and hospitals often lack the budget to purchase all-in-one machines equipped with AI 3D virtual human software. In the financial sector, VTMs, ATMs, and STMs are already deeply integrated with bank internal systems, making it difficult to integrate AI 3D virtual human software without relying on existing equipment suppliers. Whether in finance, healthcare, or government settings, AI 3D virtual human software and the all-in-one machines equipped with it pose challenges to existing business all-in-one machines, and both scenario decision-makers and existing suppliers lack the motivation to cooperate in implementation. Summary of the Invention

[0004] This invention provides a method and apparatus for invoking virtual digital humans across application scenarios, aiming to solve at least one of the technical problems existing in the prior art.

[0005] The technical solution of this invention is a method for invoking virtual digital humans across application scenarios, comprising the following steps: S100. The target terminal sets up a Windows program, and fills in multiple exe processes in the Windows program. The multiple exe processes share a window for integrated display. The exe processes include virtual human software processes and external software processes to realize the invocation of virtual digital humans across application scenarios. S200: The target terminal sets a background configuration page, in which the user interface of the virtual human software and the external software is split into two screens, the split screen form including at least vertical split screen and horizontal split screen; the target terminal sets the position ratio relationship and screen resolution of the virtual human software and the external software in the background configuration page.

[0006] According to some embodiments of the present invention, the following steps are included: S300, The target terminal is configured with intent recognition, which includes standard intent and business operation; the virtual digital human combines the chat history context to detect the intent of the target object and obtain the standard intent or the business operation.

[0007] According to some embodiments of the present invention, the following steps are included: S400, The target terminal sets system prompt words to define the role positioning, capability boundaries, context management, and behavioral constraints of the virtual digital human; the target terminal uses dynamic placeholders in the system prompt words, and during actual operation, the dynamic placeholders are replaced with current page information and current status information to define the capability boundaries and context management.

[0008] According to some embodiments of the present invention, the following steps are included: S500, the target terminal sets a background list configuration, the background list configuration includes multiple system prompt words; the target terminal sets a menu list prompt word slot to define the current page context that the virtual digital human needs to know when processing the request of the target object, and injects the information of the menu list prompt word slot into the system prompt words so that the virtual digital human can make a reasonable response when processing the request of the target object.

[0009] According to some embodiments of the present invention, the following steps are included: S600, the intent decision includes an action field, an extrainfo field, and a rewrite field. The action field is a required field, while the extrainfo field and the rewrite field are optional fields. The target terminal uses the system prompt word, and after the central control module selects the standard intent or the business operation, it determines the standard intent or the business operation as the value of the action field, outputs the intent decision in JSON format, and controls the corresponding functional module to process the standard intent or the business operation.

[0010] According to some embodiments of the present invention, the virtual human software, AI server, and external software are applied to generate the virtual digital human, and in step S100, The target object performs a voice input operation. The virtual digital human generates transcribed text based on the target object's voice input operation. The virtual digital human sends the transcribed text to the AI ​​server. The AI ​​server processes the transcribed text to obtain a response message, which defines an execution instruction to instruct the external software to respond to the target object's operation. The AI ​​server sends the response message to the external software through the virtual digital human message sending interface. The external software parses the redirection logic based on the response message and executes the redirection to update the page based on the response message. The external software sends the page update content to the virtual digital human through the virtual digital human message receiving interface. The virtual digital human forwards the page update content to the AI ​​server, and the AI ​​server temporarily stores the page update content.

[0011] According to some embodiments of the present invention, the response message includes a data field, which includes a pageId field and an action field. The pageId field is used to inform the external software of the target page to which it will be redirected, and the action field is used to inform the external software of the standard intent or business operation to be processed.

[0012] According to some embodiments of the present invention, in step S100, In response to the target object's operation on the external software, the external software updates the page based on the target object's operation. The external software sends the page update content to the virtual digital human through the virtual digital human's message receiving interface. The virtual digital human forwards the page update content to the AI ​​server, and the AI ​​server temporarily stores the page update content.

[0013] According to some embodiments of the present invention, for the virtual digital human receiving message interface, the target terminal sets API request parameter specifications to define the fields and requirements that must be passed in when calling AI services; The API request parameters include a backgroundInfo field, a sessionId field, and an organizationalId field. The backgroundInfo field contains page context information, and the sessionId field is a unique identifier for the session. The target terminal loads the prompt word template corresponding to the external software based on the organizationalId field. The backgroundInfo field and the organizationalId field are required fields, while the sessionId field is optional.

[0014] The present invention also relates to a computer device, including a memory and a processor, wherein the processor executes the method described above when executing a computer program stored in the memory.

[0015] The present invention also relates to a computer-readable storage medium having program instructions stored thereon, which, when executed by a processor, implement the method described above.

[0016] The beneficial effects of this invention include: setting up a Windows program on the target terminal, filling in multiple .exe processes within the Windows program, with multiple .exe processes sharing a single window for integrated display; the .exe processes include virtual human software processes and external software processes, to enable virtual digital human invocation across application scenarios; setting up a background configuration page on the target terminal, in which the user interface of the virtual human software and external software is split-screened, with the split-screen format including at least vertical split-screen and horizontal split-screen; and setting the position ratio and screen resolution of the virtual human software and external software on the target terminal in the background configuration page.

[0017] The target terminal is configured with a Windows program. Within this program, the virtual human software process and the external software process are integrated. These two processes share a single window for unified display, allowing for seamless integration between the virtual human and external software. The self-service registration machine software has already completed its interface integration with the hospital's HIS system. The virtual human software only needs to interface with this self-service registration machine software, eliminating the need for repeated integration with the hospital's underlying interface. Adopting a software ODM model transforms challengers (potential substitutes) into partners: it doesn't add new hardware to the application scenario, nor does it replace existing software. Instead, it deeply integrates with existing hardware and software, empowering them with AI and virtual human capabilities. This significantly reduces resistance from existing integrated machine suppliers. Users can complete various operations of the original software through natural language interaction with the AI ​​virtual human, achieving an upgrade in experience from click-based interaction to voice interaction.

[0018] Furthermore, additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0019] Figure 1 This is an optional flowchart of a method for invoking virtual digital humans across application scenarios in an embodiment of the present invention.

[0020] Figure 2 This is a schematic diagram of a conventional business all-in-one machine and the business all-in-one machine of the present invention in the embodiments of the present invention.

[0021] Figure 3a This is an optional flowchart of a method for invoking virtual digital humans across application scenarios in an embodiment of the present invention.

[0022] Figure 3b This is an optional flowchart of a method for invoking virtual digital humans across application scenarios in an embodiment of the present invention.

[0023] Figure 4 Is placed Figure 3b , Figure 5 , Figure 9 and Figure 10 This is a diagram of the list above the UE universal interface call.

[0024] Figure 5 This is an optional flowchart of a method for invoking virtual digital humans across application scenarios in an embodiment of the present invention.

[0025] Figure 6 This is an optional flowchart of a method for invoking virtual digital humans across application scenarios in an embodiment of the present invention.

[0026] Figure 7 Is placed Figure 6 and Figure 8 This is a diagram of the list above the UE universal interface call.

[0027] Figure 8 This is an optional flowchart of a method for invoking virtual digital humans across application scenarios in an embodiment of the present invention.

[0028] Figure 9 This is an optional flowchart of a method for invoking virtual digital humans across application scenarios in an embodiment of the present invention.

[0029] Figure 10 This is an optional flowchart of a method for invoking virtual digital humans across application scenarios in an embodiment of the present invention.

[0030] Figure 11 This is an optional flowchart of a split-screen cross-application data processing method combining virtual digital humans in an embodiment of the present invention.

[0031] Figure 12 This is an optional flowchart of a split-screen cross-application data processing method combining virtual digital humans in an embodiment of the present invention.

[0032] Figure 13a This is an optional flowchart of a split-screen cross-application data processing method combining virtual digital humans in an embodiment of the present invention.

[0033] Figure 13b This is an optional flowchart of a split-screen cross-application data processing method combining virtual digital humans in an embodiment of the present invention.

[0034] Figure 13c This is an optional flowchart of a split-screen cross-application data processing method combining virtual digital humans in an embodiment of the present invention.

[0035] Figure 13d This is an optional flowchart of a split-screen cross-application data processing method combining virtual digital humans in an embodiment of the present invention.

[0036] Figure 13e This is an optional flowchart of a split-screen cross-application data processing method combining virtual digital humans in an embodiment of the present invention.

[0037] Figure 14a This is an optional flowchart of a split-screen cross-application data processing method combining virtual digital humans in an embodiment of the present invention.

[0038] Figure 14b This is an optional flowchart of a split-screen cross-application data processing method combining virtual digital humans in an embodiment of the present invention.

[0039] Figure 14c This is an optional flowchart of a split-screen cross-application data processing method combining virtual digital humans in an embodiment of the present invention.

[0040] Figure 15a This is an optional flowchart of a split-screen cross-application data processing method combining virtual digital humans in an embodiment of the present invention.

[0041] Figure 15b This is an optional flowchart of a split-screen cross-application data processing method combining virtual digital humans in an embodiment of the present invention.

[0042] Figure 16 This is an optional flowchart of a split-screen cross-application data processing method combining virtual digital humans in an embodiment of the present invention.

[0043] Figure 17a This is an optional flowchart of a split-screen cross-application data processing method combining virtual digital humans in an embodiment of the present invention.

[0044] Figure 17b This is an optional flowchart of a split-screen cross-application data processing method combining virtual digital humans in an embodiment of the present invention.

[0045] Figure 18a This is an optional flowchart of a split-screen cross-application data processing method combining virtual digital humans in an embodiment of the present invention.

[0046] Figure 18b This is an optional flowchart of a split-screen cross-application data processing method combining virtual digital humans in an embodiment of the present invention.

[0047] Figure 18c This is an optional flowchart of a split-screen cross-application data processing method combining virtual digital humans in an embodiment of the present invention.

[0048] Figure 18d This is an optional flowchart of a split-screen cross-application data processing method combining virtual digital humans in an embodiment of the present invention. Detailed Implementation

[0049] The following will provide a clear and complete description of the concept, specific structure, and technical effects of the present invention in conjunction with the embodiments and accompanying drawings, so as to fully understand the purpose, solution, and effects of the present invention. It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other.

[0050] It should be noted that, unless otherwise specified, when a feature is referred to as "fixed" or "connected" to another feature, it can be directly fixed or connected to the other feature, or indirectly fixed or connected to the other feature. Furthermore, the descriptions of "upper," "lower," "left," "right," "top," and "bottom" used in this invention are only relative to the relative positional relationships of the various components of the invention in the accompanying drawings.

[0051] Furthermore, unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art. The terminology used in this specification is for the purpose of describing particular embodiments only and not for limiting the invention. The term "and / or" as used herein includes any combination of one or more of the associated listed items.

[0052] It should be understood that although the terms first, second, third, etc., may be used to describe various elements in this invention, these elements should not be limited to these terms. These terms are only used to distinguish elements of the same type from each other. For example, a first element may also be referred to as a second element without departing from the scope of this invention, and similarly, a second element may also be referred to as a first element.

[0053] Reference Figures 1 to 18d In some embodiments, the cross-application scenario virtual digital human invocation method of the present invention includes at least the following steps: S100A, the target terminal is set up with a Windows program, in which multiple exe processes are filled in. Multiple exe processes share a single window for integrated display. The exe processes include virtual human software processes and external software processes to enable the invocation of virtual digital humans across application scenarios. S200A, the target terminal sets up a background configuration page, in which the user interface of the virtual human software and external software is split into two screens, including at least vertical split screen and horizontal split screen; the target terminal sets the position ratio and screen resolution of the virtual human software and external software in the background configuration page.

[0054] Specifically, the target audience is the user.

[0055] As can be seen, the target terminal is configured with a Windows program, within which the virtual human software process and the external software process are integrated. These two processes share a single window for unified display, allowing for seamless integration between the virtual human and external software. The self-service registration machine software has already completed its interface integration with the hospital's HIS system; the virtual human software only needs to interface with this software, eliminating the need for repeated integration with the hospital's underlying interface. Adopting a software ODM model transforms challengers—potential substitutes—into partners: it doesn't add new hardware to the application scenario, nor does it replace existing software; instead, it deeply integrates with existing hardware and software, empowering them with AI and virtual human capabilities. This significantly reduces the resistance from existing integrated machine suppliers. Users can complete various operations of the original software through natural language interaction with the AI ​​virtual human, achieving an upgrade in experience from click-based interaction to voice interaction.

[0056] See Figure 2 , Figure 2 The image on the left shows a traditional business all-in-one machine, with the main operating area located at the bottom and a beanie at the top displaying some guide diagrams. The overall software interface and screen utilization are not high. Figure 2 The image on the right shows the business all-in-one machine of this invention. This device is about 1.9 meters tall. It is inconvenient for users to operate the UI on the top. Therefore, the unused area on the top is just right to place a virtual digital human that only requires voice interaction, and the virtual digital human and chat history can be displayed.

[0057] Different manufacturers use different screen hardware sizes and aspect ratios; some use portrait mode, some use landscape mode, and some use two screens. Adapting to each screen type individually would be extremely costly. Therefore, the target terminal of this invention has a background configuration page. In this background configuration page, the user interface for the virtual human software and external software is split-screened. The split-screen format includes at least vertical and horizontal split-screen. The target terminal also sets the positional ratio and screen resolution of the virtual human software and external software in the background configuration page.

[0058] Specifically, when the business all-in-one machine's screen is portrait orientation, the virtual human software and external software use a top-to-bottom split-screen configuration; when the business all-in-one machine's screen is landscape orientation, the virtual human software and external software use a left-to-right split-screen configuration; when the business all-in-one machine has two screens, the virtual human software and external software each occupy one screen. This invention develops a framework that only requires modifying the configuration in the backend to meet all forms of split-screen and embedding requirements, without requiring further development.

[0059] Specifically, the virtual human software is divided into three versions, each adapted to different platforms. The high-end version uses UE5 for rendering and is packaged as an EXE file. The mid-to-low-end version uses Unity for rendering; on Windows, it's packaged as an EXE file, while on other platforms, it's packaged as a WebGL file. The mid-to-low-end 2D CV type virtual human requires no game engine; it can be rendered in a browser without needing to be packaged and can be opened via a URL.

[0060] On the Windows side, develop a container program to house the virtual human software's EXE process and the EXE processes of external software, displaying them all in a single window. Develop a Windows program using C++ that can accept multiple process names and display them within the Windows application. These processes can communicate via sockets to transfer data. Finally, abstract a standardized interface for the virtual human software.

[0061] On Android, Linux, or domestic IT systems, a low-spec version of the virtual digital human is rendered using the Unity engine. The virtual digital human software is then packaged into WebGL format, embedded into external software, and a standardized interface is provided to facilitate embedding by external software.

[0062] In a specific embodiment, the target terminal defines a standard interface for the virtual human software. External software controls the virtual digital human's behavior by calling the standard interface, and the virtual human software reports the target object's behavior to the external software by calling the standard interface, thereby realizing data interaction between the virtual human software and the external software. Specifically, the virtual human software exposes multiple interfaces, including controlling the virtual human to speak and perform actions, as well as controlling some UI and asset aspects of the software, such as the Speak interface, the action playback interface, the language switching interface, and the information update interface.

[0063] In some embodiments, the virtual digital human invocation method across application scenarios includes the following steps: S300A, target terminal is configured with intent recognition settings, which include standard intent and business operation; virtual digital human combines chat history context to detect the intent of the target object and obtain standard intent or business operation; The S400A and target terminal set system prompt words to define the role positioning, capability boundaries, context management, and behavioral constraints of the virtual digital human; the target terminal uses dynamic placeholders in the system prompt words, and the dynamic placeholders are replaced with the current page information and current status information during actual operation to define capability boundaries and context management; S500A, target terminal sets background list configuration, background list configuration includes multiple system prompt words; target terminal sets menu list prompt word slots to define the current page context that the virtual digital human needs to know when processing the request of the target object, and injects the information of the menu list prompt word slots into the system prompt words so that the virtual digital human can make a reasonable response when processing the request of the target object; The S600A intent decision includes the action field, extrainfo field, and rewrite field. The action field is mandatory, while the extrainfo and rewrite fields are optional. The target terminal uses system prompts. After the central control module selects a standard intent or business operation, it determines the standard intent or business operation as the value of the action field and outputs the intent decision in JSON format to control the corresponding functional module to process the standard intent or business operation.

[0064] Specifically, the target terminal obtains the standard intent or business operation based on the action field, and controls the corresponding functional module to process the standard intent or business operation based on the extrainfo field and the rewrite field.

[0065] Specifically, the code example for intent recognition configuration is as follows: "optionalActions":["knowledge","registration","medical-guidance","login","way-finding","Dir-button","unclear-action","Dir-refuse","chat"payment, printing, price inquiry, general inquiry, department inquiry, file creation] When the external software is used in a medical application scenario, standard intents include medical knowledge Q&A, medical-guidance, hospital navigation, Dir-button, Dir-refuse, unclear-action, and chat. Business operations include registration, login, payment, printing, price inquiry, comprehensive inquiry, department inquiry, and file creation.

[0066] Specifically, the code example for the System Prompt is as follows: <background> You are currently in the central control module. After you select a behavior, other modules will process the user's query. Therefore, you need to select the correct behavior module. However, from a user experience perspective, we need to ensure users understand our architecture. You are currently in... <Page.name> <Page.description> <status> Please select the most appropriate action from the following list. For action selection, you are required to output in JSON format. When outputting the action, you also need to consider whether to include `extraInfo`. The output format is: {"action":"", "extrainfo":""(optional),"rewrite":""(optional)}. <action> Specific examples are as follows: You are currently in the central control module. After you select a behavior, other modules will process the user's query. Therefore, you need to select the correct behavior module. However, from a user experience perspective, we need to ensure users understand our architecture. You are currently in... Menu page.

[0067] On this page, the self-service machine will guide users to register.

[0068] The user is not currently logged in, and some operations are restricted. Please select the most appropriate action from the following list. For action selection, you are required to output in JSON format. When outputting the action, you also need to consider whether to include `extraInfo`. The output format is: {"action":"", "extrainfo":""(optional),"rewrite":""(optional)}. Specifically, the extrainfo field is used to pass additional parameters (such as department ID and user ID) to downstream modules, while the rewrite field is used to rewrite queries: optimize the original user query for easier downstream processing.

[0069] In some embodiments, the software is applied to virtual human software, an AI server, and external software. The virtual human software generates a virtual digital human. In step S100A... The target object performs voice input on the target object. The virtual digital human generates transcribed text based on the target object's voice input and the transcribed text is sent to the AI ​​server. The AI ​​server processes the transcribed text to obtain a response message, which defines the execution instructions and instructs external software to respond to the target object's operation. The AI ​​server sends response messages to external software through the virtual digital human message sending interface. The external software parses the redirection logic based on the response message and executes the redirection to update the page based on the response message. The external software sends the page update content to the virtual digital human through the virtual digital human message receiving interface. The virtual digital human forwards the page update content to the AI ​​server, and the AI ​​server temporarily stores the page update content.

[0070] In some embodiments, the response message includes a data field, which includes a pageId field and an action field. The pageId field is used to inform the external software of the target page to which it will be redirected, and the action field is used to inform the external software of the standard intent or business operation to be processed.

[0071] In some embodiments, in step S100A, In response to the target object's operation on the external software, the external software updates the page based on the target object's operation. The external software sends the page update content to the virtual digital human through the virtual digital human message receiving interface. The virtual digital human forwards the page update content to the AI ​​server, and the AI ​​server temporarily stores the page update content.

[0072] In some embodiments, for the virtual digital human message receiving interface, the target terminal sets API request parameter specifications to define the fields and requirements that must be passed when calling AI services; The API request parameters include the backgroundInfo, sessionId, and organizationalId fields. The backgroundInfo field contains page context information, the sessionId field is a unique identifier for the session, and the target terminal loads the corresponding prompt word template from external software based on the organizationalId field. The backgroundInfo and organizationalId fields are required, while the sessionId field is optional.

[0073] Specifically, Figure 3a , Figure 3b , Figure 5 , Figure 6 , Figure 8 , Figure 9 and Figure 10 They can be pieced together sequentially. See also Figure 3b and Figure 4 The user clicked on the ID card login option on the login page, as shown in the example below: In response to the target object's operation on the external software, the external software updates the page based on the target object's operation. The external software sends the page update content to the virtual digital human through the virtual digital human message receiving interface. The virtual digital human forwards the page update content to the AI ​​server, and the AI ​​server temporarily stores the page update content. If external software needs the virtual digital human to speak, the external software prepares the text and calls the speaking interface, the virtual digital human speaks, and the virtual digital human inserts the spoken content into the chat history; if external software needs the virtual digital human to perform actions, the external software calls the action interface, and the virtual digital human performs the action.

[0074] The `backgroundInfo` field provides page context information, including the current page ID and state ID. The `sessionId` field is a unique session identifier used to maintain conversation continuity. The `organizationalId` field is the organization / institution ID used to distinguish different hospital / tenant configurations.

[0075] Figure 3b The code on it is an Action Command, which uses JSON format to define the specific behaviors that the virtual digital human needs to perform.

[0076] Specifically, see Figure 5 Users can log in by voice inputting their ID card number, as shown in the example below: The target object performs voice input on the target object. The virtual digital human generates transcribed text based on the target object's voice input and the transcribed text is sent to the AI ​​server. The AI ​​server processes the transcribed text to obtain a response message, which defines the execution instructions and instructs external software to respond to the target object's operation. The AI ​​server sends response messages to external software through the virtual digital human message sending interface. The external software parses the redirection logic based on the response message and executes the redirection to update the page based on the response message. The external software sends the page update content to the virtual digital human through the virtual digital human message receiving interface. The virtual digital human forwards the page update content to the AI ​​server, and the AI ​​server temporarily stores the page update content.

[0077] Figure 5 The code on the screen represents the response message, which includes the `code`, `id`, `message`, and `data` fields. The `data` field includes the `pageId` and `action` fields. The `pageId` field informs the external software of the target page to navigate to, and the `action` field informs the external software of the standard intent or business operation to be performed. The `code` field contains the status code: 0 indicates success (non-zero indicates various errors); the `id` field identifies the message type: the message / instruction issued by the AI; the `message` field contains a human-readable description (for logging or debugging); and the `data` field contains the core instruction data.

[0078] Specifically, see Figure 6 and Figure 8 Users can directly click to the obesity clinic registration confirmation page, and then voice-input "I want to change the time / doctor / department", as shown in the example below: In response to the target object's operation on the external software (login + clicked to the registration confirmation page), the external software updates the page based on the target object's operation. The external software sends the page update content and Business additional content to the virtual digital human through the virtual digital human message receiving interface. The virtual digital human forwards the page update content to the AI ​​server, and the AI ​​server temporarily stores the page update content. If external software needs the virtual digital human to speak, the external software prepares the text and calls the speaking interface, the virtual digital human speaks, and the virtual digital human inserts the spoken content into the chat history; if external software needs the virtual digital human to perform an action, the external software calls the action interface, and the virtual digital human performs the action. The target object inputs a voice operation (I want to change to the dental department, Dr. Li, appointment the day after tomorrow morning). The virtual digital human generates a transcribed text based on the target object's voice input operation. The virtual digital human sends the transcribed text to the AI ​​server. The AI ​​server processes the transcribed text to obtain a response message, which defines the execution instructions and instructs the external software to respond to the target object operation. The AI ​​server sends response messages to external software through the virtual digital human message sending interface. The external software parses the redirection logic based on the response message and executes the redirection to update the page based on the response message. The external software sends the page update content to the virtual digital human through the virtual digital human message receiving interface. The virtual digital human forwards the page update content to the AI ​​server, and the AI ​​server temporarily stores the page update content.

[0079] The Business field describes the business status, recording the user's completed operations and current progress.

[0080] Specifically, see Figure 9 The user can voice-input "I want to get a SIM card for my child. The phone number is 123XXXXX910. The child's name is Li XX, and his birthday is January 1, 1900. He is a boy." An example is shown below: The target object inputs a voice operation (I want to get a card for my child, the phone number is 12345678910, the child's name is Li XX, the birthday is January 1, 1900, and he is a boy). The virtual digital human generates a transcribed text based on the target object's voice input operation. The virtual digital human sends the transcribed text to the AI ​​server. The AI ​​server processes the transcribed text and obtains a response message to define the execution instructions, which instruct external software to respond to the target object's operation. The AI ​​server sends response messages to external software through the virtual digital human message sending interface. The external software parses the redirection logic based on the response message and executes the redirection to update the page based on the response message. The external software sends the page update content to the virtual digital human through the virtual digital human message receiving interface. The virtual digital human forwards the page update content to the AI ​​server, and the AI ​​server temporarily stores the page update content.

[0081] In a specific embodiment, the external software periodically updates the departmental appointment data. The external software sends the departmental appointment data to the virtual digital human through the virtual digital human's message sending interface. The virtual digital human forwards the departmental appointment data to the AI ​​server, and the AI ​​server temporarily stores the departmental appointment data.

[0082] Specifically, see Figure 10 A user can use voice input to say, "I want to book an appointment with Dr. Chen XX in the dermatology department for 9 AM tomorrow," as shown in the example below: The target object inputs a voice command (I want to book an appointment with Dr. Chen XX in the dermatology department for 9 AM tomorrow). The virtual digital human generates a transcribed text based on the target object's voice input. The virtual digital human sends the transcribed text to the AI ​​server. The AI ​​server processes the transcribed text to obtain a response message, which defines the execution instructions and directs the external software to respond to the target object's operation. The AI ​​server sends response messages to external software through the virtual digital human message sending interface. The external software parses the redirection logic based on the response message and executes the redirection to update the page based on the response message. The external software sends the page update content to the virtual digital human through the virtual digital human message receiving interface. The virtual digital human forwards the page update content to the AI ​​server, and the AI ​​server temporarily stores the page update content.

[0083] Reference Figures 1 to 18d In some embodiments, the cross-application split-screen data processing method of the present invention, which combines virtual digital humans, is applied to intent control prompts, business prompts, virtual human software, and external software. The virtual human software generates a virtual digital human, which includes at least the following steps: S100B: When the virtual digital human recognizes the target object, the virtual digital human greets the target and the chat history is saved. S200B: The target object performs a voice input operation. The virtual digital human converts the voice input operation into text and inserts it into the chat history. The external software sends the current page information to the intent control prompt word. The intent control prompt word obtains the system prompt word based on the current page information and the chat history. The intent control prompt word recognizes the system prompt word and outputs the standard intent. The business prompt word receives the standard intent and calls the chat history. The business prompt word generates the intent result and sends the intent result to the virtual digital human. The virtual digital human outputs the intent result and the business prompt word inserts the intent result into the chat history.

[0084] As can be seen, when the virtual human recognizes a target object, it greets them, the chat history is stored, the target object inputs a voice command, the virtual human converts the voice input into text and inserts it into the chat history, the external software sends the current page information to the central control prompt, the central control prompt obtains the system prompt based on the current page information and the chat history, the central control prompt recognizes the system prompt and outputs the standard intent, the business prompt receives the standard intent and calls the chat history, the business prompt generates the intent result, the business prompt sends the intent result to the virtual human, the virtual human outputs the intent result, and the business prompt inserts the intent result into the chat history. This realizes data interaction between the virtual human software and the external software, thus achieving technical integration between the virtual human software and the external software. The self-service registration machine software has already completed interface integration with the hospital's HIS system; the virtual human software only needs to interface with the self-service registration machine software, without needing to repeatedly interface with the hospital's underlying interface. Adopting a software ODM model can transform challengers, i.e., potential substitutes, into partners: without adding new hardware to the application scenario or replacing existing software, but rather deeply integrating with existing hardware and software, empowering them with AI and virtual human capabilities. As a result, the resistance from existing all-in-one machine suppliers will be significantly reduced. Users can complete various operations of the original software by interacting with the AI ​​virtual human using natural language, achieving an experience upgrade from click-based interaction to voice interaction.

[0085] Specifically, when a user performs a voice operation, the AI ​​server invokes a virtual digital human to speak; when a user performs a click operation, external software invokes the virtual digital human to speak. Collier refers to the external software.

[0086] Specifically, Figure 11 , Figure 12 , Figures 13a-13e and Figures 14a-14c They can be pieced together one by one.

[0087] exist Figure 11 In the chat logs, chat log 1 reads: "assistant: 'Hi! I'm your medical assistant, registration guide...'", chat log 2 reads: "assistant: 'Hi! I'm your medical assistant, registration guide...' user: 'I fell and my leg hurts, which department should I go to?'", chat log 3 reads: "assistant: 'Hi! I'm your medical assistant, registration guide...' user: 'I fell and my leg hurts, which department should I go to?'", chat log 4 reads: "assistant: 'Hi! I'm your medical assistant, registration guide...' user: 'I fell and my leg hurts, which department should I go to?'", a: You can go to orthopedics, or you can go to the emergency room...".

[0088] exist Figure 12 In the chat logs, chat log 5 reads: "assistant: 'Hi! I'm your medical assistant, registration navigation...' user: 'I fell and my leg hurts, which department should I go to?' a: You can go to orthopedics or the emergency room... user: I want to go to orthopedics." Chat log 6 reads: "assistant: 'Hi! I'm your medical assistant, registration navigation...' user: 'I fell and my leg hurts, which department should I go to?' a: You can go to orthopedics or the emergency room... user: I want to go to orthopedics." Chat log 7 reads: "assistant: 'Hi! I'm your medical assistant, registration navigation...' user: 'I fell and my leg hurts, which department should I go to?' a: You can go to orthopedics or the emergency room... user: I want to go to orthopedics. a: Okay, please tell me your login method first, and I'll help you...'"

[0089] exist Figures 13a-13e In the chat logs, chat log 8 reads: "assistant: 'Hi! I'm your medical assistant, registration navigation...' user: 'I fell and my leg hurts, which department should I go to?' a: You can go to orthopedics or the emergency room... user: I want to go to orthopedics a: Okay, please tell me your login method first, I'll help you... user: ID card login." Chat log 9 reads: "assistant: 'Hi! I'm your medical assistant, registration navigation...' user: 'I fell and my leg hurts, which department should I go to?' a: You can go to orthopedics or the emergency room... user: I want to go to orthopedics a: Okay, please tell me your login method first, I'll help you... user: ID card login." Chat log 10 reads: "assistant: 'Hi! I'm your medical assistant, registration navigation...' user: 'I fell and my leg hurts, which department should I go to?'" "A: You can register for orthopedics or the emergency room... User: I want to register for orthopedics. A: Okay, please tell me your login method first, and I'll help you... User: ID card login A: Okay, please put your ID card into the recognition area as I instructed." Chat log eleven is "assistant: "Hi! I'm your medical assistant, registration navigation..." User: "I fell and my leg hurts, which department should I register for?" A: You can register for orthopedics or the emergency room... User: I want to register for orthopedics. A: Okay, please tell me your login method first, and I'll help you... User: ID card login A: Okay, please put your ID card into the recognition area as I instructed. A: Okay, please tell me the date and doctor you want to book." Chat log twelve is "assistant: "Hi! I'm your medical assistant, registration navigation..." User: "I fell and my leg hurts, which department should I register for?" "A: You can register for orthopedics or the emergency room... User: I want to register for orthopedics. A: Okay, please tell me your login method first, and I'll help you... User: Login with ID card. A: Okay, please place your ID card in the recognition area as instructed. A: Okay, please tell me the date and doctor you want to book. User: I want to book an appointment with Dr. Wu XX on Wednesday." Chat history 13 is "assistant: "Hi! I'm your medical assistant, registration navigation..." User: "I fell and my leg hurts, which department should I go to?" A: You can register for orthopedics or the emergency room... User: I want to register for orthopedics. A: Okay, please tell me your login method first, and I'll help you..."User: Login with ID card A: Okay, please place your ID card in the recognition area as instructed. A: Okay, please tell me the date and doctor you wish to book. User: I want to book an appointment with Dr. Wu XX on Wednesday. A: Okay, I've booked an appointment for you with Dr. Wu XX on Wednesday afternoon from 2:00 PM to 2:30 PM. If you confirm the appointment information is correct, please click "Confirm". Chat history fourteen is "assistant: "Hi! I'm your medical assistant, registration navigation..." User: "I fell and my leg hurts, which department should I go to?" A: You can... I can book an appointment with the orthopedics department, or the emergency room... User: I want to book an appointment with the orthopedics department. A: Okay, please tell me your login method first, and I'll help you... User: I'm logging in with my ID card. A: Okay, please place your ID card in the recognition area as instructed. A: Okay, please tell me the date and doctor you'd like to book. User: I want to book an appointment with Dr. Wu XX on Wednesday. A: Okay, I've booked an appointment for you with Dr. Wu XX on Wednesday from 2:00 PM to 2:30 PM. If you confirm the appointment information is correct, please click "Confirm". A: Okay, please select or tell me your payment method.

[0090] exist Figures 14a-14c In the chat log 15, the message reads: "assistant: 'Hi! I'm your medical assistant, registration navigation...' user: 'I fell and my leg hurts, which department should I go to?' a: You can go to orthopedics or the emergency room... user: I want to go to orthopedics a: Okay, please tell me your login method first, I'll help you... user: ID card login a: Okay, please put your ID card in the recognition area as I instructed a: Okay, please tell me the date and doctor you want to book an appointment with user: I want to book an appointment with Dr. Wu XX on Wednesday a: Okay, I've booked an appointment for you with Dr. Wu XX on Wednesday afternoon from 2:00 PM to 2:30 PM. If you confirm the appointment information is correct, please click 'Confirm' a: Okay, please select or tell me your payment method user: Alipay payment." The next chat log 16 reads: "assistant: 'Hi! I'm your medical assistant, registration navigation...' user: 'I fell and my leg hurts, which department should I go to?'" "A: You can register for orthopedics or emergency... User: I want to register for orthopedics. A: Okay, please tell me your login method first, and I'll help you... User: ID card login A: Okay, please place your ID card in the recognition area as instructed. A: Okay, please tell me the date and doctor you want to book. User: I want to book an appointment with Dr. Wu XX on Wednesday. A: Okay, I've booked an appointment for you with Dr. Wu XX on Wednesday from 2:00 PM to 2:30 PM. If you confirm the appointment information is correct, please click 'Confirm'. A: Okay, please select or tell me your payment method. User: Alipay payment A: Okay, please scan the QR code below to complete the payment." Chat history seventeen is "assistant: "Hi! I'm your medical assistant, registration navigation..." User: "I fell and my leg hurts, which department should I go to?" "A: You can register for orthopedics or the emergency room... User: I want to register for orthopedics. A: Okay, please tell me your login method first, and I'll help you... User: ID card login A: Okay, please place your ID card in the recognition area as instructed. A: Okay, please tell me the date and doctor you want to book. User: I want to book an appointment with Dr. Wu XX on Wednesday. A: Okay, I've booked an appointment for you with Dr. Wu XX on Wednesday from 2:00 PM to 2:30 PM. If you confirm the appointment information is correct, please click "Confirm". A: Okay, please select or tell me your payment method. User: Alipay payment A: Okay, please scan the QR code below to complete the payment. A: You have successfully registered. There is one person ahead of you in the queue. Don't forget to take your receipt." Chat history 18 is "assistant: "Hi! I'm your medical assistant, registration navigation..." User: "I fell and my leg hurts. What department should I go to?" A: You can register for orthopedics or the emergency room... User: I want to register for orthopedics. A: Okay, please tell me your login method first, and I'll help you...User: Login with ID card A: Okay, please place your ID card in the recognition area as instructed. A: Okay, please tell me the date and doctor you wish to book. User: I want to book an appointment with Dr. Wu XX on Wednesday. A: Okay, I've booked an appointment for you with Dr. Wu XX on Wednesday afternoon from 2:00 PM to 2:30 PM. If you confirm the appointment information is correct, please click "Confirm". A: Okay, please select or tell me your payment method. User: Alipay payment A: Okay, please scan the QR code below to complete the payment. A: You have successfully booked an appointment. There is one person ahead of you in the queue. Don't forget to pick up your receipt. A: The Respiratory Medicine Department is on the second floor, first door on the right. (You can zoom in to view the map.)

[0091] See Figure 11 A user can voice input "I fell and my leg hurts, which department should I go to?", as shown in the example below: The virtual digital human is in standby mode. When the virtual digital human recognizes a target, it greets the target and the chat history is saved. The user inputs a voice command (e.g., "I fell and my leg hurts, what department should I go to?"). The virtual android converts the user's voice input into text and inserts it into the chat history. External software sends the current page information to the central intent control prompt. The central intent control prompt obtains the system prompt based on the current page information and the chat history. The central intent control prompt recognizes the system prompt and outputs a standard intent (intelligent triage intent). The business prompt receives the standard intent (intelligent triage intent) and accesses the chat history. The business prompt generates the intent result (e.g., "You can go to orthopedics or the emergency room. If it's just leg pain, go to orthopedics"). The business prompt sends the intent result to the virtual android, which outputs the intent result. The business prompt inserts the intent result into the chat history and waits for the user's second round of input.

[0092] In some embodiments, the split-screen cross-application data processing method incorporating virtual digital humans includes the following steps: In the S300B system, the target user inputs a voice command to perform an operation. The virtual digital human converts the voice-input into text and inserts it into the chat history. External software sends the current page information to the intent control prompt. The intent control prompt obtains a system prompt based on the current page information and the chat history. The intent control prompt recognizes the system prompt and outputs a business operation. When the intent control prompt recognizes the system prompt and outputs a business operation, the target terminal determines whether the target user is logged in. If the target user is not logged in, the business prompt receives the business operation and retrieves the chat history. The business prompt generates a business result and sends it to the virtual digital human. The virtual digital human outputs the business result, and the business prompt inserts the result into the chat history. The business result instructs the target user to make a login intent and instructs the business prompt to suspend the business operation.

[0093] Specifically, see Figure 12 The user can input "I want to register for orthopedics" via voice, as shown in the following example: During the virtual digital human listening process, the target user inputs a voice command (e.g., "I want to register for orthopedics"). The virtual digital human converts this voice input into text and inserts it into the chat history. External software sends the current page information to the intent control prompt. The intent control prompt obtains a system prompt based on the current page information and the chat history. The intent control prompt recognizes the system prompt and outputs the business operation. When the intent control prompt recognizes the system prompt and outputs the business operation, the target terminal determines whether the target user is logged in. If the target user is not logged in, the business prompt receives the business operation and retrieves the chat history. The business prompt generates a business result (e.g., "Okay, please tell me your login method so I can help you register"). The business prompt sends the business result to the virtual digital human, which outputs the business result. The business prompt inserts the business result into the chat history and waits for the user's second round of input. In this process, the business result instructs the target user to make a login intention and instructs the business prompt to suspend the business operation.

[0094] In some embodiments, step S300B further includes: The target user inputs a voice command to perform an action. The virtual android converts this voice input into text and inserts it into the chat history. The central intent control prompt word obtains system prompt words based on the current page information and the chat history. The central intent control prompt word recognizes the system prompt words and outputs a standard intent. The business prompt word receives the standard intent and retrieves the chat history. When the standard intent is login, the business prompt generates a response message and intent result. The business prompt message sends a response message to external software. The external software executes redirection logic based on the response message to update the page based on the target object. The external software then sends the updated page content to the virtual avatar. The virtual avatar forwards the updated page content to the intent control prompt message, which temporarily stores the updated page content. The business prompt sends the intent result to the virtual digital human, the virtual digital human outputs the intent result, the business prompt inserts the intent result into the chat history, the intent result instructs the target object to perform a login operation, the external software verifies the login operation performed by the target object, and if the verification is successful, the target object logs in successfully.

[0095] Specifically, see Figures 13a-13e and Figures 14a-14c Triage + registration + payment + navigation, example as follows: During the virtual digital human's listening process, the target user inputs a voice command (e.g., logging in with their ID card). The virtual digital human converts this voice input into text and inserts it into the chat history. The central intent control prompt word obtains system prompt words from the chat history. The central intent control prompt word recognizes the system prompt words and outputs a standard intent. The business prompt word receives the standard intent and retrieves the chat history. When the standard intent is login intent, the business prompt generates a response message (redirecting to the login page) and intent result (OK, please place your ID card in the recognition area as instructed). The business prompt message sends a response message to external software. The external software executes redirection logic based on the response message to update the page based on the target object. The external software then sends the updated page content to the virtual avatar. The virtual avatar forwards the updated page content to the intent control prompt message, which temporarily stores the updated page content. The business prompt sends the intent result to the virtual digital human, the virtual digital human outputs the intent result, the business prompt inserts the intent result into the chat history, the intent result instructs the target object to perform a login operation, and the external software verifies the login operation performed by the target object (login by swiping an ID card). If the verification is successful, the target object logs in successfully. After the target user successfully logs in, the target terminal resumes the suspended business operations in the business results. If the operation instruction information of the business operation is incomplete, the business prompt word sends a response message (redirecting to the registration page after login) to the external software. The external software executes the redirection logic based on the response message to update the page based on the business operation. The external software sends the updated page content (including department, doctor, and hospital information) to the virtual digital human. The virtual digital human forwards the updated page content to the intent control prompt word. The intent control prompt word temporarily stores the updated page content. The external software calls the speech interface to output supplementary information instructions (Okay, please tell me the date and doctor you want to book). The virtual digital human outputs supplementary information instructions and sends them to the intent control prompt word and inserts them into the chat history. The virtual digital human waits for the next round of input. Information supplementation instructions guide the target object's voice input for operation. The target user inputs a voice command ("I want to book an appointment with Dr. Wu XX on Wednesday"). The virtual avatar converts the voice input into text and inserts it into the chat history. The intent control prompt word obtains a system prompt word from the chat history. The intent control prompt word recognizes the system prompt word and outputs the business operation (booking an appointment). The business prompt word receives the business operation and generates a response message. The external software executes the business operation (including selecting the doctor and time) based on the response message and updates the page. The external software waits for the user to click on the operation. The business prompt word outputs the execution result through the virtual avatar ("Okay, I have booked an appointment with Dr. Wu XX on Wednesday from 2:00 PM to 2:30 PM. If you confirm the appointment information is correct, please click 'OK'") and inserts the execution result into the chat history. Once the target user confirms the information is correct, the external software redirects to the payment page. The external software sends the updated page content to the intent control prompt, which temporarily stores the updated content and waits for the next round of input. The external software then calls the speech interface to output the target user's next action instruction ("Okay, please select or tell me your payment method"). The virtual avatar outputs the target user's next action instruction and sends it to the intent control prompt and inserts it into the chat history. The target user inputs a voice action (Alipay payment). The virtual avatar converts the voice input into text and inserts it into the chat history. The central control prompt word obtains a system prompt word from the chat history. The central control prompt word recognizes the system prompt word and outputs the business operation (payment / payment). When the central control prompt word recognizes the system prompt word and outputs the business operation, the target terminal determines whether the target user is logged in. If the target user is logged in, the business prompt word receives the business operation, generates a response message (redirecting to the payment page) and a business result (OK, please scan the QR code below to complete the payment). The business prompt word sends the response message to external software. The external software executes the redirection logic based on the response message to update the page based on the target user's operation. The external software sends the updated page content to the virtual avatar. The virtual avatar forwards the updated page content to the central control prompt word. The central control prompt word temporarily stores the updated page content. The business prompt word sends the business result to the virtual avatar. The virtual avatar outputs the business result and inserts the business result into the chat history. The business result instructs the target object to perform a business operation. The target object performs the business operation (user scans QR code to pay). The external software verifies the business operation performed by the target object. If the verification is successful, the external software completes the business operation (payment / fee payment). The external software updates the page and calls the speech interface to output the execution result (You have successfully registered. There is one person ahead of you in the queue. Don't forget to pick up your receipt). The virtual avatar outputs the execution result and sends the execution result to the intent control prompt and inserts it into the chat history. The external software calls the speech interface to output navigation content (The respiratory medicine department is on the second floor, the first door on the right. You can zoom in to view the map). The virtual avatar outputs the navigation content and inserts it into the chat history.

[0096] In some embodiments, step S300B further includes: After the target object successfully logs in, the target terminal resumes the suspended business operation in the business results. If the operation instruction information of the business operation is incomplete, the business prompt word will send a response message to the external software. The external software executes the jump logic based on the response message to update the page based on the business operation. The external software sends the page update content to the virtual digital human. The virtual digital human forwards the page update content to the intention control prompt word. The intention control prompt word temporarily stores the page update content. The external software calls the speech interface to output information supplement instructions. The virtual digital human outputs information supplement instructions and sends the information supplement instructions to the intention control prompt word and inserts them into the chat history.

[0097] In some embodiments, step S300B further includes: Information supplementation instructions guide the target object's voice input for operation. The target user inputs a voice command to perform an action. The virtual android converts the voice command into text and inserts it into the chat history. The intent control prompt word obtains system prompt words from the chat history. The intent control prompt word recognizes the system prompt words and outputs business operations. The business prompt word receives the business operations and generates a response message. External software executes the business operations based on the response message and updates the page. The business prompt word outputs the execution result through the virtual android and inserts the execution result into the chat history.

[0098] In some embodiments, the split-screen cross-application data processing method incorporating virtual digital humans includes the following steps: In the S400B system, the target user inputs a voice command to perform an operation. The virtual digital human converts the voice-input operation into text and inserts it into the chat history. External software sends the current page information to the intent control prompt. The intent control prompt obtains a system prompt based on the current page information and the chat history. The intent control prompt recognizes the system prompt and outputs a business operation. When the intent control prompt recognizes the system prompt and outputs a business operation, the target terminal determines whether the target user is logged in. If the target user is logged in, the business prompt receives the business operation and retrieves the chat history. The business prompt generates a response message and a business result, and sends the response message to the external software. The external software executes redirection logic based on the response message to update the page based on the target user's operation. The external software sends the updated page content to the virtual digital human, which forwards the updated page content to the intent control prompt. The intent control prompt temporarily stores the updated page content. The business prompt sends the business result to the virtual digital human, which outputs the business result. The business prompt inserts the business result into the chat history.

[0099] In some embodiments, step S400B further includes: The business result instructs the target object to perform a business operation. The target object performs the business operation, and the external software verifies the business operation performed by the target object. If the verification is successful, the external software completes the business operation, updates the page, and calls the speech interface to output the execution result. The virtual digital human outputs the execution result and sends the execution result to the intent control prompt and inserts it into the chat history.

[0100] In some embodiments, step S400B further includes: The business result instructs the target user to perform a voice input operation. The virtual avatar then converts the voice input into text and inserts it into the chat history. The intent control prompt word obtains system prompt words based on the chat history. The intent control prompt word recognizes the system prompt words and outputs business operations. The business prompt word receives the business operations and calls the chat history. The business prompt word generates a response message. The external software executes the business operations based on the response message and updates the page. The business prompt word outputs the execution result through a virtual digital human and inserts the execution result into the chat history.

[0101] In some embodiments, the split-screen cross-application data processing method incorporating virtual digital humans includes the following steps: In the S500B, the target object input box is used to input the target object operation. The external software executes the jump logic to update the page based on the target object operation. The external software jumps to the information confirmation page and calls the speech interface to output a confirmation instruction. The virtual digital human outputs the confirmation instruction and the external software inserts the confirmation instruction into the chat history.

[0102] Specifically, Figure 15a , Figure 15b , Figure 16 , Figure 17a , Figure 17b and Figures 18a-18d They can be pieced together one by one.

[0103] exist Figure 15a and Figure 15b In the chat log, chat log 19 reads "assistant: "Hi! I'm your medical assistant, registration navigation..."; chat log 20 reads "assistant: "Hi! I'm your medical assistant, registration navigation..."; user: "ID card login"; chat log 21 reads "assistant: "Hi! I'm your medical assistant, registration navigation..."; user: "ID card login"; and chat log 22 reads "assistant: "Hi! I'm your medical assistant, registration navigation..."; user: "ID card login". A: Okay, please place your ID card in the recognition area as instructed.

[0104] exist Figure 16 , Figure 17a and Figure 17b In the chat log 23, it reads: "assistant: "Hi! I'm your medical assistant, registration navigation..." user: "ID card login" a: Okay, please put your ID card into the recognition area as I instructed a: You can tell me directly what service you need, and you can ask me any questions." Chat log 24 reads: "assistant: "Hi! I'm your medical assistant, registration navigation..." user: "ID card login" a: Okay, please put your ID card into the recognition area as I instructed a: You can tell me directly what service you need, and you can ask me any questions." user: "Children's medical record registration." Chat log 25 reads: "assistant: "Hi! I'm your medical assistant, registration navigation..." user: "ID card login" a: Okay, please put your ID card into the recognition area as I instructed a: You can tell me directly what service you need, and you can ask me any questions." user: "Children's medical record registration." Chat log 26 reads: "assistant: "Hi! "I am your medical assistant, registration navigation..." user: "ID card login" a: Okay, please follow my instructions and place your ID card in the recognition area. a: You can tell me directly what service you need, and you can ask me any questions. user: Child registration a: Please enter the child's name, date of birth, and gender information below."

[0105] exist Figures 18a-18d Chat log 27 reads: "assistant: "Hi! I'm your medical assistant, registration navigation..." user: "ID card login" a: Okay, please place your ID card in the recognition area as instructed. a: You can tell me directly what service you need, and you can ask me any questions. user: Child registration a: Please enter the child's name, date of birth, and gender information below. user: 159XXXXX678 Zhang XX..." Chat log 28 reads: "assistant: "Hi! I'm your medical assistant, registration navigation..." user: "ID card login" a: Okay, please place your ID card in the recognition area as instructed. a: You can tell me directly what service you need, and you can ask me any questions. user: Child registration a: Please enter the child's name, date of birth, and gender information below. user: 159XXXXX678 Zhang XX..." Chat log 29 reads: "assistant: "Hi! "I am your medical assistant, registration navigation..." user: "ID card login" a: Okay, please follow my instructions to place your ID card in the recognition area. a: You can directly tell me what service you need, and you can ask me any questions. user: Child registration a: Please enter the child's name, date of birth, and gender information below. user: 159XXXXX678 Zhang XX... a: Please enter the child's name and date of birth. After entering, please click "OK". Chat history 30 is "assistant: "Hi! I am your medical assistant, registration navigation..." user: "ID card login" a: Okay, please follow my instructions to place your ID card in the recognition area. a: You can directly tell me what service you need, and you can ask me any questions. user: Child registration a: Please enter the child's name, date of birth, and gender information below. user: 159XXXXX678 Zhang XX... a: Please enter the child's name and date of birth. After entering, please click "OK". a: Please double-check that the information is correct. If it is correct, click the "OK" button. Chat history 31 is "assistant: "Hi! "I am your medical assistant, registration navigation..." user: "ID card login" a: Okay, please place your ID card in the recognition area as instructed. a: You can tell me directly what service you need, and you can ask me any questions. user: Child registration a: Please enter the child's name, date of birth, and gender information below. user: 159XXXXX678ZhangXX... a: Please enter the child's name and date of birth. After entering, please click "OK". a: Please double-check that the information is correct. If it is correct, click the "OK" button. a: Registration successful. You can pick up your medical card in the area indicated below.

[0106] Specifically, see Figure 15a , Figure 15b , Figure 16 , Figure 17a , Figure 17b and Figures 18a-18d For example, the following is a record of a child's medical record: The virtual digital human is in standby mode. When the virtual digital human recognizes a target, it greets the target and the chat history is saved. The target user inputs a voice action (ID card login). The virtual avatar converts the voice input into text and inserts it into the chat history. The external software sends the current page information to the central intent control prompt. The central intent control prompt obtains the system prompt based on the current page information and the chat history. The central intent control prompt recognizes the system prompt and outputs a standard intent (login intent). The business prompt receives the standard intent and calls the chat history. The business prompt generates a response message (redirecting to the login page) and an intent result (OK, please place your ID card in the recognition area as instructed). The business prompt sends the response message to the external software. The external software executes the redirection logic based on the response message to update the page based on the target user's action. The external software sends the updated page content to the virtual avatar. The virtual avatar forwards the updated page content to the central intent control prompt. The central intent control prompt temporarily stores the updated page content. The business prompt sends the intent result to the virtual avatar. The virtual avatar outputs the intent result. The business prompt inserts the intent result into the chat history. The external software waits for the user to insert their card. The intent result instructs the target object to perform a login operation (login by swiping its ID card). External software verifies the login operation performed by the target object. If the verification is successful, the target object logs in successfully. The external software identifies user information. If there are no to-do items, the external software opens the menu page and sends the updated page content to the virtual avatar. The virtual avatar forwards the updated page content to the central intent control prompt, which temporarily stores the updated page content and waits for user input. External software calls the speech interface to output the inquiry content (You can directly tell me what business you want to do? You can ask me if you have any questions). The virtual digital human outputs the inquiry content and sends the inquiry content to the intent control prompt and inserts it into the chat history. During the virtual digital human listening process, the target subject inputs a voice command (child registration). The virtual digital human converts the voice input into text and inserts it into the chat history. The intent control prompt word obtains system prompt words based on the chat history. The intent control prompt word recognizes the system prompt words and outputs the business operation (registration). The business prompt word receives the business operation and calls the chat history. The business prompt word generates a response message (redirecting to the registration page) and a business result (please enter the child's name, date of birth, and gender information below). The business prompt message sends a response message to external software. The external software executes redirection logic based on the response message to update the page according to the business operation. The external software sends the updated page content to the virtual avatar. The virtual avatar forwards the updated page content to the intent control prompt message. The intent control prompt message temporarily stores the updated page content and waits for user input. The business prompt sends the business result to the virtual avatar, the virtual avatar outputs the business result, and the virtual avatar inserts the business result into the chat history; During the virtual avatar's listening process, the business result instructs the target user to input a voice command (e.g., 159XXXXX678, Zhang XX, November 10, 2001, baby girl). The virtual avatar then converts the voice input into text and inserts it into the chat history. The intent control prompt word obtains system prompt words based on the chat history. The intent control prompt word recognizes the system prompt words and outputs business operations (file creation). The business prompt word receives the business operations and calls the chat history. The business prompt word generates a response message (information is entered on the file creation page). The external software executes the business operations based on the response message and updates the page. The business prompt word outputs the execution result through a virtual digital human (after input, please click "OK") and inserts the execution result into the chat history. The target object input box is used to input the target object operation (click to modify the information on the file creation page). The external software executes the jump logic to update the page based on the target object operation. The external software jumps to the information confirmation page (file creation information confirmation page). The external software sends the updated page content to the intent control prompt. The intent control prompt temporarily stores the file creation information confirmation page. The external software calls the speech interface to output a confirmation instruction (please confirm the information is correct again. If it is correct, click the "OK" button). The virtual digital human outputs the confirmation instruction, and the external software inserts the confirmation instruction into the chat history. When the target clicks the "OK" button, the external software redirects to the file creation success page. The external software then sends the updated page content to the intent control prompt, which temporarily stores the file creation success page. The external software then calls the speech interface to output a closing statement ("File creation successful, you can retrieve your medical card in the area below"). The virtual digital human outputs the closing statement and inserts it into the chat history.

[0107] It's important to note that the menu list prompt slots are context variable templates dynamically populated into the system prompts, used to inform the AI ​​in real time about the available functions and interface elements on the current page. The Intent MasterControl Prompt is defined as the central decision-making hub of the AI ​​system, responsible for understanding the user's natural language input, recognizing their intent, and determining which business module to route the request to. Business Prompts (or Domain-Specific Prompts) are defined as specialized execution modules within a vertical domain, receiving the intent and parameters distributed by the master control, executing specific business logic, and generating professional responses.

[0108] This invention also provides a computer device including a memory and a processor, wherein the processor performs the above-described method when executing a computer program stored in the memory.

[0109] This invention also provides a computer-readable storage medium storing program instructions thereon, which, when executed by a processor, implement the method described above.

[0110] It should be understood that the method steps in the embodiments of the present invention can be implemented or carried out by computer hardware, a combination of hardware and software, or by computer instructions stored in a non-transitory computer-readable storage medium. The method can use standard programming techniques. Each program can be implemented in a high-level procedural or object-oriented programming language to communicate with the computer system. However, if necessary, the program can be implemented in assembly or machine language. In any case, the language can be a compiled or interpreted language. Furthermore, for this purpose, the program can run on a programmed application-specific integrated circuit (ASIC).

[0111] Furthermore, the procedures described herein may be performed in any suitable order unless otherwise indicated herein or otherwise clearly contradicted by the context. The procedures described herein (or variations and / or combinations thereof) may be executed under the control of one or more computer systems configured with executable instructions, and may be implemented by hardware or a combination thereof as code (e.g., executable instructions, one or more computer programs, or one or more applications) that commonly executes on one or more processors. The computer program comprises a plurality of instructions executable by one or more processors.

[0112] Furthermore, the method can be implemented in any suitable type of computing platform, including but not limited to personal computers, minicomputers, mainframes, workstations, networked or distributed computing environments, standalone or integrated computer platforms, or in communication with charged particle tools or other imaging devices, etc. Aspects of the invention can be implemented as machine-readable code stored on a non-transitory storage medium or device, whether removable or integrated into a computing platform, such as a hard disk, optical read and / or write storage medium, RAM, ROM, etc., such that it is readable by a programmable computer, and when the storage medium or device is read by the computer, it can be used to configure and operate the computer to perform the processes described herein. Furthermore, the machine-readable code, or portions thereof, can be transmitted via wired or wireless networks. The invention described herein includes these and other different types of non-transitory computer-readable storage media when such media comprises instructions or programs that implement the steps described above in conjunction with a microprocessor or other data processor. When programmed according to the methods and techniques described in the invention, the invention may also include the computer itself.

[0113] A computer program can be applied to input data to perform the functions described herein, thereby transforming the input data to generate output data stored in non-volatile memory. The output information can also be applied to one or more output devices, such as a display. In a preferred embodiment of the invention, the transformed data represents physical and tangible objects, including specific visual depictions of physical and tangible objects generated on the display.

[0114] The above description is merely a preferred embodiment of the present invention. The present invention is not limited to the above-described embodiments. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention, as long as they achieve the technical effects of the present invention by the same means, should be included within the scope of protection of the present invention. Within the scope of protection of the present invention, the technical solutions and / or implementation methods can have various modifications and variations.< / action> < / status> < / background>

Claims

1. A method for virtual digital human invocation across application scenarios, characterized in that, Includes the following steps: S100. The target terminal sets up a Windows program, and fills in multiple exe processes in the Windows program. The multiple exe processes share a window for integrated display. The exe processes include virtual human software processes and external software processes to realize the invocation of virtual digital humans across application scenarios. S200: The target terminal sets a background configuration page, in which the user interface of the virtual human software and the external software is split into two screens, the split screen form including at least vertical split screen and horizontal split screen; the target terminal sets the position ratio relationship and screen resolution of the virtual human software and the external software in the background configuration page.

2. The method for invoking virtual digital humans across application scenarios according to claim 1, characterized in that, Includes the following steps: S300, The target terminal is configured with intent recognition, which includes standard intent and business operation; the virtual digital human combines the chat history context to detect the intent of the target object and obtain the standard intent or the business operation.

3. The method for invoking virtual digital humans across application scenarios according to claim 2, characterized in that, Includes the following steps: S400, The target terminal sets system prompt words to define the role positioning, capability boundaries, context management and behavioral constraints of the virtual digital human; The target terminal uses dynamic placeholders in the system prompts. During actual operation, the dynamic placeholders are replaced with current page information and current status information to define the capability boundaries and context management.

4. The method for invoking virtual digital humans across application scenarios according to claim 3, characterized in that, Includes the following steps: S500, the target terminal sets a background list configuration, the background list configuration includes multiple system prompt words; the target terminal sets a menu list prompt word slot to define the current page context that the virtual digital human needs to know when processing the request of the target object, and injects the information of the menu list prompt word slot into the system prompt words so that the virtual digital human can make a reasonable response when processing the request of the target object.

5. The method for invoking virtual digital humans across application scenarios according to claim 4, characterized in that, Includes the following steps: S600, the intent decision includes an action field, an extrainfo field, and a rewrite field. The action field is a required field, while the extrainfo field and the rewrite field are optional fields. The target terminal uses the system prompt word, and after the central control module selects the standard intent or the business operation, it determines the standard intent or the business operation as the value of the action field, outputs the intent decision in JSON format, and controls the corresponding functional module to process the standard intent or the business operation.

6. The method for invoking virtual digital humans across application scenarios according to claim 2, characterized in that, Applied to the virtual human software, AI server, and external software, the virtual human software generates the virtual digital human. In step S100, The target object performs a voice input operation. The virtual digital human generates transcribed text based on the target object's voice input operation. The virtual digital human sends the transcribed text to the AI ​​server. The AI ​​server processes the transcribed text to obtain a response message, which defines an execution instruction to instruct the external software to respond to the target object's operation. The AI ​​server sends the response message to the external software through the virtual digital human message sending interface. The external software parses the redirection logic based on the response message and executes the redirection to update the page based on the response message. The external software sends the page update content to the virtual digital human through the virtual digital human message receiving interface. The virtual digital human forwards the page update content to the AI ​​server, and the AI ​​server temporarily stores the page update content.

7. The method for invoking virtual digital humans across application scenarios according to claim 6, characterized in that, The response message includes a data field, which includes a pageId field and an action field. The pageId field is used to inform the external software of the target page to which it will be redirected, and the action field is used to inform the external software of the standard intent or business operation to be processed.

8. The method for invoking virtual digital humans across application scenarios according to claim 6, characterized in that, In step S100, In response to the target object's operation on the external software, the external software updates the page based on the target object's operation. The external software sends the page update content to the virtual digital human through the virtual digital human's message receiving interface. The virtual digital human forwards the page update content to the AI ​​server, and the AI ​​server temporarily stores the page update content.

9. The method for invoking virtual digital humans across application scenarios according to claim 8, characterized in that, For the virtual digital human message receiving interface, the target terminal sets API request parameter specifications to define the fields and requirements that must be passed when calling AI services; The API request parameters include a backgroundInfo field, a sessionId field, and an organizationalId field. The backgroundInfo field contains page context information, and the sessionId field is a unique identifier for the session. The target terminal loads the prompt word template corresponding to the external software based on the organizationalId field. The backgroundInfo field and the organizationalId field are required fields, while the sessionId field is optional.

10. A computer device comprising a memory and a processor, characterized in that, When the processor executes a computer program stored in the memory, it performs the method as described in any one of claims 1 to 9.

Images