Control method, storage medium, electronic apparatus, and computer program product

Abstract

Provided in the embodiments of the present disclosure are a control method, a storage medium, an electronic apparatus, and a computer program product. The method comprises: acquiring an AI interaction instruction; and on the basis of a network environment of a user equipment, determining, from among a plurality of AI modules, a target AI module for processing the AI interaction instruction, wherein the plurality of AI modules comprise an AI module located in a cloud environment and an AI module located in a local environment.

Description

Control methods, storage media, electronic devices and computer program products

[0001] Cross-references to related applications

[0002] This disclosure is based on and claims priority to Chinese patent application CN2024114497738, filed on October 16, 2024, entitled “Control Method, Storage Medium, Electronic Device and Computer Program Product”, the entire contents of which are incorporated herein by reference. Technical Field

[0003] This disclosure relates to the field of communications, and more specifically, to a control method, storage medium, electronic device, and computer program product. Background Technology

[0004] With the rapid development of cloud computing and artificial intelligence technologies, cloud computing has gradually become popular as a new computing model, providing users with flexible and efficient computing resources. However, traditional cloud computing has significant limitations in the integration and application of artificial intelligence (AI) assistants. Specifically, the activation and interaction of AI assistants are mostly limited to single scenarios outside the cloud (i.e., the user's local device or mobile device environment) or inside the cloud (i.e., the cloud environment), lacking the ability to seamlessly switch between the two, resulting in inconsistent user experiences. Summary of the Invention

[0005] This disclosure provides a control method, storage medium, electronic device, and computer program product to at least address the issue of inconsistent user experience of AI assistants in cloud computers in cloud-based and cloud-based environments in the related art.

[0006] According to one embodiment of this disclosure, a control method is provided, comprising: acquiring AI interaction instructions; determining a target AI module for processing the AI ​​interaction instructions from a plurality of AI modules based on the network environment of a user device, wherein the plurality of AI modules includes: an AI module located in a cloud environment and an AI module located in a local environment.

[0007] According to another embodiment of this disclosure, a user device is provided, including: an AI agent module configured to acquire AI interaction instructions, and to determine a target AI module for processing the AI ​​interaction instructions from a plurality of AI modules based on the network environment of the user device, wherein the plurality of AI modules includes: an AI module located in a cloud environment and an AI module located in a local environment.

[0008] According to another embodiment of this disclosure, a control system is also provided, including: a user device located in a local environment and at least one cloud host located in a cloud environment. The user device includes an AI agent module and an AI module located in the local environment, and the cloud host includes an AI module located in the cloud environment. The AI ​​agent module is configured to acquire AI interaction instructions and determine a target AI module for processing the AI ​​interaction instructions from a plurality of AI modules based on the network environment of the user device. The plurality of AI modules includes an AI module located in the cloud environment and an AI module located in the local environment.

[0009] According to yet another embodiment of this disclosure, a computer-readable storage medium is also provided, which stores a computer program configured to perform the steps in any of the above method embodiments when executed.

[0010] According to yet another embodiment of this disclosure, an electronic device is also provided, including a memory and a processor, the memory storing a computer program and the processor being configured to run the computer program to perform the steps in any of the above method embodiments.

[0011] According to yet another embodiment of this disclosure, a computer program product is also provided, including a computer program that, when executed by a processor, implements the steps in any of the above method embodiments. Attached Figure Description

[0012] Figure 1 is a hardware structure block diagram of a computer terminal for a control method according to an embodiment of the present disclosure;

[0013] Figure 2 is a flowchart of a control method for an AI agent module according to an embodiment of the present disclosure;

[0014] Figure 3 is a schematic diagram of a user equipment according to an embodiment of the present disclosure;

[0015] Figure 4 is a schematic diagram of a control system according to an embodiment of the present disclosure;

[0016] Figure 5 is a schematic flowchart of the control method in one embodiment of this disclosure;

[0017] Figure 6 is a schematic diagram of the process of waking up an AI module located in the local environment in one embodiment of the present disclosure;

[0018] Figure 7 is a schematic diagram of the process of waking up an AI module located in a cloud environment according to an embodiment of the present disclosure;

[0019] Figure 8 is a schematic diagram of the interaction of AI modules in a local environment according to an embodiment of this disclosure;

[0020] Figure 9 is a schematic diagram of the interaction of AI modules in a cloud environment according to an embodiment of this disclosure;

[0021] Figure 10 is a schematic diagram of the interaction between an AI module located in a local environment and an AI module located in a cloud environment in one embodiment of this disclosure.

[0022] Figure 11 is a schematic diagram of the process of a user device in a local environment performing remote work in an embodiment of this disclosure;

[0023] Figure 12 is a schematic diagram of the process of a cloud host in a cloud environment performing remote work in an embodiment of this disclosure;

[0024] Figure 13 is a schematic diagram of the process of a user device in a local environment performing online learning in an embodiment of this disclosure. Detailed Implementation

[0025] The embodiments of this disclosure will be described in detail below with reference to the accompanying drawings and examples.

[0026] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this disclosure are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should also be understood that in the embodiments of this disclosure, "a plurality of" can refer to two or more, and "at least one" can refer to one, two, or more.

[0027] The methods and embodiments provided in this disclosure can be executed in a mobile terminal, a computer terminal, or a similar computing device. Taking a computer terminal as an example, FIG1 is a hardware structure block diagram of a computer terminal in which the methods and embodiments of this disclosure are run. As shown in FIG1, the computer terminal may include one or more (only one is shown in FIG1) processors 102 (processors 102 may include, but are not limited to, microprocessors MCUs or programmable logic devices FPGAs, etc.) and a memory 104 for storing data. The computer terminal may also include a transmission device 106 for communication functions and an input / output device 108. Those skilled in the art will understand that the structure shown in FIG1 is only illustrative and does not limit the structure of the computer terminal. For example, the computer terminal may also include more or fewer components than shown in FIG1, or have a different configuration than shown in FIG1.

[0028] The memory 104 can be used to store computer programs, such as application software programs and modules, like the computer program corresponding to the control method in this embodiment. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, thus implementing the above-described method. The memory 104 may include high-speed random access memory and non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor 102, and these remote memories can be connected to a computer terminal via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0029] The transmission device 106 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the mobile terminal's communication provider. In one example, the transmission device 106 includes a Network Interface Controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission device 106 may be a Radio Frequency (RF) module, used for wireless communication with the Internet.

[0030] In one embodiment of this disclosure, a control method is provided, applied to an AI agent module. Figure 2 is a flowchart of the control method for the AI ​​agent module according to an embodiment of this disclosure. As shown in Figure 2, the process includes the following steps:

[0031] Step S202: Obtain AI interaction commands;

[0032] Step S204: Determine the target AI module for processing the AI ​​interaction command from multiple AI modules based on the network environment of the user device.

[0033] In this embodiment, the plurality of AI modules include: an AI module located in a cloud environment and an AI module located in a local environment.

[0034] In some embodiments, user equipment includes, but is not limited to, user equipment located in a local environment and cloud hosts located in a cloud environment. The user equipment located in a local environment can be any terminal device capable of running a cloud computer client, able to connect to the cloud server through specific client software to achieve cloud computer functions, such as a personal computer, laptop, tablet, or smartphone. The cloud host located in the cloud environment provides the core infrastructure for cloud computer services, requiring high-performance computing capabilities and stable network connectivity to ensure a smooth user experience, and is responsible for handling user computing tasks, storing data, and running AI assistants, etc. This disclosure does not impose any limitations in this regard.

[0035] In some embodiments, the network environment in which the user equipment resides includes, but is not limited to, a local environment and a cloud environment. The user equipment located in a local environment includes, but is not limited to, user equipment connected to home Wi-Fi, a corporate network, or a mobile data network. The cloud environment is the data center network environment where the user equipment resides, supporting high-speed, low-latency network connections.

[0036] In some embodiments, a user device located in a local environment includes an AI agent module and an AI module located in a local environment, and a cloud host located in a cloud environment includes an AI module located in a cloud environment. This disclosure does not limit this.

[0037] In some embodiments, the AI ​​agent module can determine the target AI module by judging the user device's network environment based on information such as the user device's current network status and device connectivity. This disclosure does not limit the conditions for judging the user device's network environment.

[0038] In some embodiments, the AI ​​agent module can also acquire real-time connection information between user devices located in the local environment and cloud hosts. For example, it can identify whether a user is using a cloud host through existing protocols, thereby determining whether the user is in a cloud environment. Alternatively, it can determine whether the user is in a local environment by whether the user is only using local applications.

[0039] In some embodiments, AI interaction commands include, but are not limited to, AI wake-up commands and commands based on specific user needs.

[0040] In some embodiments, the AI ​​module includes an AI assistant (front-end) and an AI service (back-end), configured to perform corresponding processing based on AI interaction instructions. The AI ​​assistant may employ AI technologies such as Natural Language Processing (NLP), speech recognition and synthesis, and computer vision to achieve interaction with the user in various forms, including voice, text, images, and video. It is configured to understand the user's specific needs, perform corresponding operations, and generate feedback results. The AI ​​service is configured to convert the AI ​​interaction instructions received by the AI ​​assistant into system instructions that the system can understand, and process them in a local or cloud environment, or connect to large cloud models on demand to provide corresponding feedback. This disclosure does not impose any limitations on this.

[0041] In some embodiments, due to the different environments in which the AI ​​assistant is located, it can be divided into a local AI assistant in an AI module located in a local environment, and a cloud AI assistant in an AI module located in a cloud environment.

[0042] Through the embodiments of this disclosure, the AI ​​agent module can intelligently determine the target AI module in the corresponding environment based on the network environment of the user device, so that AI modules in different environments can process AI interaction commands accordingly, which enhances the interaction between inside and outside the cloud. This solves the problem of inconsistent user experience of AI assistants in cloud computers in cloud environments and outside the cloud in related technologies, making the user experience of AI assistants consistent in cloud environments and improving the user experience.

[0043] In some embodiments, step S202 may include any of the following steps:

[0044] Step S202A: Obtain the AI ​​interaction command input through the AI ​​button module, wherein the AI ​​button module includes: preset physical buttons or virtual buttons;

[0045] Step S202B: Obtain the AI ​​interaction command input through the interface interaction module, wherein the interface interaction module includes: a user interface located in the cloud environment and / or a user interface located in the local environment.

[0046] In some embodiments, the input methods in the interface interaction module include, but are not limited to, voice input, text input, image input, and video input.

[0047] In some embodiments, the interface interaction module utilizes an intuitive and easy-to-use user interface to display the interactive interface in real time, supporting user interaction with the AI ​​assistant. Interaction methods include, but are not limited to, voice, text, images, video, and feedback display.

[0048] In some embodiments, the AI ​​button module may be able to recognize user presses and trigger preset AI wake-up commands.

[0049] In one exemplary embodiment, after pressing the AI ​​button module, the AI ​​agent module receives the AI ​​wake-up command from the AI ​​button module, determines that the current user device is in a local environment, and wakes up the local AI assistant.

[0050] In some embodiments, the preset physical button can be a physical button. The pressing action of the physical button is recognized through a hardware interface or driver, and a wake-up signal is triggered. For example, the AI ​​button module can be physically integrated into the "F12" key on the keyboard, and an "AI" label can be added.

[0051] In some embodiments, virtual buttons can simulate the function of physical buttons on a touchscreen or software interface. For example, an AI button module can be virtually implemented on a touchscreen via software, including but not limited to a soft keyboard and touchpad gestures.

[0052] In some embodiments, step S204 may include the following steps:

[0053] Step S2041: In response to the AI ​​interaction command, determine the network environment of the user equipment;

[0054] Step S2042: Determine the first AI module based on the network environment of the user equipment.

[0055] Step S2043, determining the first AI module as the target AI module includes...

[0056] In this embodiment, the first AI module can independently process AI interaction commands. This disclosure does not limit the number of first AI modules.

[0057] In some embodiments, the network environment in which the first AI module is located affects the wake-up logic and interaction method of subsequent AI modules. For example, in response to an AI wake-up command, if the network environment of the user device is detected to be a local environment, the AI ​​agent module receives the AI ​​wake-up command, determines that the AI ​​module located in the local environment is the first module, and wakes up the first AI module.

[0058] In some embodiments, step S2042 may include any of the following steps:

[0059] Step S2042-2: If the network environment is the cloud environment, the AI ​​module located in the cloud environment is identified as the first AI module.

[0060] Step S2042-4: If the network environment is the local environment, the AI ​​module located in the local environment is identified as the first AI module.

[0061] Through the embodiments disclosed herein, the AI ​​module can be automatically switched when the network environment changes, ensuring that the AI ​​module can effectively process AI interaction commands regardless of whether the environment is cloud-based or local. Furthermore, when using user devices that support cloud server connections, the AI ​​module can be adaptively switched for different cloud servers, improving the user experience.

[0062] In some embodiments, step S2042-4 includes the following steps:

[0063] Step S2042-42: In the case that the network environment is the cloud environment, determine the target cloud host connected to the user equipment from multiple cloud hosts;

[0064] Steps S2042-44: The AI ​​module deployed on the target cloud host is identified as the first AI module.

[0065] In some embodiments, step S204 may include steps S2041 to S2043 described above, as well as the following steps:

[0066] Step S2044: In response to the AI ​​interaction command, determine the computing power resources of the multiple AI modules;

[0067] Step S2045: Determine the second AI module based on the computing power resources of the plurality of AI modules and the AI ​​interaction instructions;

[0068] Step S2046: Determine the first AI module and the second AI module as the target AI module.

[0069] In this embodiment, the first AI module and the second AI module can cooperate to process AI interaction commands. This disclosure does not limit the number of the first AI module and the second AI module.

[0070] In some embodiments, step S2045 may be executed only when the computing power of the first AI module is insufficient. For example, step S2045 may include: in response to the fact that the computing power resources of the first AI module do not meet the computing power requirements of the AI ​​interaction instruction, determining from the plurality of AI modules the AI ​​module whose computing power resources meet the computing power requirements of the AI ​​interaction instruction as the second AI module.

[0071] In one exemplary embodiment, if the AI ​​interaction command is set to real-time image recognition, the computing power resources of the first AI module awakened by the network environment may not be sufficient to meet the computing power requirements of image recognition. Therefore, a second AI module with a graphics processing unit (GPU) can be awakened by an AI agent module (i.e., the second AI module meets the computing power requirements of image recognition). The first AI module and the second AI module can be located in a local environment and a cloud environment, or they can be located on different cloud hosts within the cloud environment.

[0072] In some other embodiments, step S204 may include the following steps:

[0073] Step S2047: In response to the AI ​​interaction command, determine the network environment of the user equipment and the computing power resources of the multiple AI modules;

[0074] Step S2048: Determine the third AI module from the multiple AI modules based on the network environment of the user equipment, the computing power resources of the multiple AI modules, and the AI ​​interaction instructions;

[0075] Step S2049: Determine that the third AI module is the target AI module.

[0076] In this embodiment, the third AI module can independently process AI interaction commands. This disclosure does not limit the number of third AI modules.

[0077] In this embodiment, the AI ​​module may need to process large amounts of data or perform complex calculations when executing AI interaction commands. Therefore, an AI module that meets the computing power requirements of the AI ​​interaction command can be directly selected based on the command content and network environment. The selection range of the target AI module is determined by the network environment. For example, if the local user device has established a remote connection with cloud host 1 and cloud host 2, an AI module with sufficient computing power can be selected from the AI ​​modules deployed on the local user device, cloud host 1, and cloud host 2.

[0078] In an exemplary embodiment, step S2048 may be implemented in the following ways: first, determine a number of candidate AI modules from a number of AI modules based on the network environment of the user equipment, and then determine the candidate AI module whose computing power resources meet the computing power requirements of the AI ​​interaction instructions as the third AI module.

[0079] In an exemplary embodiment, the specific implementation of step S2048 can also refer to steps S2042 and S2045 above. First, a first AI module is determined based on the network environment. If the computing power of the first AI module is insufficient, a third AI module that meets the instruction computing power requirements is determined based on the computing power. The difference is that the second AI module in step S2045 needs to cooperate with the first AI module to complete the processing of AI interaction instructions, while the third AI module in step S2048 can complete the processing of AI interaction instructions independently.

[0080] In this embodiment of the disclosure, the target AI module can be located on a local user device or on a cloud host. Depending on the network environment / location of the target AI module, the interaction process between the AI ​​agent module and the target AI module may also differ.

[0081] In some embodiments, the method may further include: step S206A, sending the AI ​​interaction instruction to the target AI module located in the cloud environment, and receiving the response result of the AI ​​interaction instruction from the target AI module located in the cloud environment. For example, the response result of the AI ​​interaction instruction may be obtained by the AI ​​module in the cloud environment completing relevant processing based on the AI ​​interaction instruction. This relevant processing includes, but is not limited to, processing within the system, or processing through a large cloud-based model.

[0082] In an exemplary embodiment, the process of cloud interaction between the AI ​​agent module and the cloud AI module (including cloud AI assistant 1 and AI service 1) in the cloud environment in step S206A can be referred to steps TC1-2, TC1-3, TC1-4 and TC1-5 in Figure 5:

[0083] In step TC1-2, the AI ​​agent module forwards the digital signal to the cloud AI assistant 1 for processing;

[0084] In steps TC1-3, the cloud AI assistant 1 identifies the digital signal, converts the digital signal into a system command that the system can recognize, and forwards it to the cloud AI service 1.

[0085] In steps TC1-4, Cloud AI Service 1 completes the relevant processing according to the system instructions and returns the response result to Cloud AI Assistant 1. Cloud AI Service 1 can process system instructions internally within the cloud host system, or it can connect to the cloud-based large model as needed to process the system instructions and obtain the response result. For example, if the computing power within the cloud host system is insufficient to process the system instructions, Cloud AI Service 1 can forward the system instructions to the cloud-based large model and obtain the response result from the cloud-based large model.

[0086] In step TC1-5, Cloud AI Assistant 1 returns a response result to the AI ​​Agent Module.

[0087] In some embodiments, the method may further include: step S206B, sending the AI ​​interaction instruction to the target AI module located in the cloud environment, wherein the response result of the AI ​​interaction instruction is output through the user interface of the cloud environment.

[0088] In this embodiment, the response result of the AI ​​interaction command can be directly output through the user interface of the cloud environment without needing to be returned to the AI ​​agent module. For example, the response result of the AI ​​interaction command can be obtained by the AI ​​module of the cloud environment completing relevant processing based on the AI ​​interaction command. This processing includes, but is not limited to, processing within the system, or processing through a large cloud-based model.

[0089] In an exemplary embodiment, the process of cloud interaction between the AI ​​agent module and the cloud AI module (including cloud AI assistant 1 and AI service 1) in the cloud environment in step S206B can be referred to steps TC1-2, TC1-3, TC1-4 and TC1-9 in Figure 5:

[0090] In step TC1-2, the AI ​​agent module forwards the digital signal to the cloud AI assistant 1 for processing;

[0091] In steps TC1-3, the cloud AI assistant 1 identifies the digital signal, converts the digital signal into a system command that the system can recognize, and forwards it to the cloud AI service 1.

[0092] In steps TC1-4, Cloud AI Service 1 completes the relevant processing according to the system instructions and returns the response result to Cloud AI Assistant 1. Cloud AI Service 1 can process system instructions internally within the cloud host system, or it can connect to the cloud-based large model as needed to process the system instructions and obtain the response result. For example, if the computing power within the cloud host system is insufficient to process the system instructions, Cloud AI Service 1 can forward the system instructions to the cloud-based large model and obtain the response result from the cloud-based large model.

[0093] In steps TC1-9, the cloud AI assistant 1 directly sends the response result to the user interface of the cloud environment and outputs the response result through the user interface of the cloud environment.

[0094] In some embodiments, the method may further include: step S206C, sending the AI ​​interaction instruction to the target AI module located in the local environment, and receiving the response result of the AI ​​interaction instruction from the target AI module located in the local environment. For example, the response result of the AI ​​interaction instruction may be obtained by the local AI module completing relevant processing based on the AI ​​interaction instruction. This processing includes, but is not limited to, processing within the system, or processing through a large cloud-based model.

[0095] In an exemplary embodiment, the process of local interaction between the AI ​​agent module in step S206C and the local AI module (including the local AI assistant and AI services) in the local environment can be referred to steps TL-2, TL-3 and TL-4 in Figure 5:

[0096] In step TL-2, the AI ​​agent module forwards the digital signal to the local AI assistant for processing;

[0097] Step TL-3: The local AI assistant identifies the digital signal, converts the digital signal into a system command that the system can recognize, and forwards it to the local AI service;

[0098] In step TL-4, the local AI service completes the relevant processing according to the system instructions and returns the response result to the local AI assistant. The local AI service can process system instructions within the user device system or, as needed, connect to a cloud-based large model to process the system instructions and obtain the response result. For example, if the local system's internal computing power is insufficient to process the system instructions, the local AI service can forward the system instructions to the cloud-based large model and obtain the response result from the cloud-based large model.

[0099] Furthermore, after step TL-4, the local AI assistant can first return the response result to the AI ​​agent module, and then the AI ​​agent module can forward it to other cloud hosts or user devices according to the content of the response result, so as to output the response result through the corresponding user interface.

[0100] In some embodiments, the method may further include: step S206D, sending the AI ​​interaction instruction to the target AI module located in the local environment, wherein the response result of the AI ​​interaction instruction is output through the user interface of the local environment.

[0101] In this embodiment, the response result of the AI ​​interaction command can be directly output through the user interface of the local environment without returning it to the AI ​​agent module. For example, the response result of the AI ​​interaction command can be obtained by the local AI module completing relevant processing based on the AI ​​interaction command. Such processing includes, but is not limited to, processing within the system, or processing through a large cloud model.

[0102] In an exemplary embodiment, the process of local interaction between the AI ​​agent module in step S206D and the local AI module (including the local AI assistant and AI service) in the local environment can be referred to steps TL-2, TL-3, TL-4 and TL-5 in Figure 5:

[0103] In step TL-2, the AI ​​agent module forwards the digital signal to the local AI assistant for processing;

[0104] Step TL-3: The local AI assistant identifies the digital signal, converts the digital signal into a system command that the system can recognize, and forwards it to the local AI service;

[0105] Step TL-4: The local AI service completes the relevant processing according to the system instructions and returns the response result to the local AI assistant. The local AI service can process system instructions within the user device system or, as needed, connect to a cloud-based large model to process the system instructions and obtain the response result. For example, if the local system's internal computing power is insufficient to process the system instructions, the local AI service can forward the system instructions to the cloud-based large model and obtain the response result from the cloud-based large model.

[0106] In step TL-5, the local AI assistant returns the response result to the user interface located in the local environment.

[0107] In this embodiment, the local AI assistant can directly output response results through the user interface of the local environment, but this disclosure is not limited thereto.

[0108] In some embodiments, after the AI ​​agent module obtains the response result of the AI ​​interaction command in step S206A or step S206C, the method may further include at least one of the following:

[0109] Step S208A: Send the response result of the AI ​​interaction command to the AI ​​module located in the local environment, so that the AI ​​module located in the local environment outputs the response result through the user interface located in the local environment;

[0110] Step S208B: Send the response result of the AI ​​interaction command to the AI ​​module located in the cloud environment, so that the AI ​​module located in the cloud environment can output the response result through the user interface located in the cloud environment.

[0111] In this embodiment, the AI ​​agent module can interact with the AI ​​module in the local environment or the cloud environment, and output the response result through the user interface located in the local environment or the cloud environment. If the AI ​​module and the user interface of the output result are in different environments, the AI ​​agent module needs to forward the response result. If the AI ​​module and the user interface of the output result are in the same environment, the AI ​​module can directly output the response result to the user interface in the same environment.

[0112] In an exemplary embodiment, the process of outputting the response result through the user interface of the local environment in step S208A can be referred to steps TC1-6 and TC1-7 in FIG5:

[0113] In steps TC1-6, the AI ​​agent module forwards the response result to the local AI assistant;

[0114] In steps TC1-7, the local AI assistant returns the response result to the user interface located in the local environment.

[0115] In this embodiment, depending on the specific content of the AI ​​interaction command and the response result, the response result can be flexibly output or displayed in the user interface of the local environment or the cloud environment, and this disclosure does not impose any limitations on this. For example, if the response result of the AI ​​interaction command is to launch an application in the cloud environment, the AI ​​agent module will forward the response result to the corresponding cloud host. If the response result of the AI ​​interaction command requires operation on an application in the local environment, the AI ​​agent module will forward the response result to the local user device. Furthermore, the AI ​​agent module can directly interact with the user interface in the local environment or the cloud environment, or it can interact with the corresponding user interface through the AI ​​agent module in the cloud environment or the local environment; this disclosure is not limited to this.

[0116] In this embodiment, this flexible instruction processing and result reception mechanism ensures that users receive a consistent interactive experience and feedback regardless of whether the AI ​​module executes in a cloud environment or a local environment. Furthermore, when there are multiple AI interaction instructions, the target AI module processing each instruction can be different; the AI ​​agent module forwards each instruction to the corresponding target AI module in the network environment. Depending on the content of the AI ​​interaction instructions, the response result can be output to the user interface in the cloud environment or the user interface in the local environment.

[0117] In one exemplary embodiment, the AI ​​interaction command is an AI wake-up command. After receiving the AI ​​wake-up command, the AI ​​agent module determines that the current network environment is the local environment, and then forwards the AI ​​wake-up command to the local AI module. The local AI module is woken up after receiving the AI ​​wake-up command.

[0118] In another exemplary embodiment, the AI ​​interaction command is an AI wake-up command. After receiving the AI ​​wake-up command, the AI ​​agent module determines that the current network environment is a cloud environment and sends it to the cloud host located in the cloud environment through the cloud desktop communication protocol. The cloud host receives the AI ​​wake-up command and forwards it to the cloud AI module. The cloud AI module is woken up after receiving the AI ​​wake-up command.

[0119] In some embodiments, when an AI module has been activated, the AI ​​agent module can directly forward AI interaction commands to the activated AI module for processing, thereby ensuring processing efficiency. Furthermore, if the computing power of the activated AI module is insufficient to meet the computing power requirements of the AI ​​interaction command, the AI ​​agent module can continue to activate other AI modules with sufficient computing power to process the AI ​​interaction command. The user side can flexibly schedule AI computing resources in different environments without any additional action, and the scheduling process is completely imperceptible to the user, thus improving the user experience.

[0120] In some embodiments, the response results output by the user interface in the local environment and the user interface in the cloud environment may be presented in the form of voice, images, video, or text, and this disclosure does not limit this.

[0121] In one exemplary embodiment, the AI ​​agent module determines that the current environment is a local environment and sends the user's specific requirements to the AI ​​module located in the local environment. The AI ​​assistant (front end) in the AI ​​module forwards the instructions to the AI ​​service (back end) in the AI ​​module. The AI ​​service processes the user's specific requirements within the system or connects to the cloud-based large model as needed, and sends the response result to the AI ​​assistant. The AI ​​assistant sends the response result to the AI ​​agent module, and the AI ​​agent module outputs the response result through the user interface located in the local environment.

[0122] In one exemplary embodiment, the AI ​​agent module determines that the current environment is a cloud environment and sends the user's specific request instructions to the cloud host located in the cloud environment via the cloud desktop communication protocol. The cloud host receives the user's specific request instructions and forwards them to the AI ​​assistant located in the cloud host. The AI ​​assistant sends them to the AI ​​service. The AI ​​service processes the user's specific request instructions within the cloud host system or connects to the cloud-based large model as needed, and sends the response result to the AI ​​assistant in the cloud host. The AI ​​assistant in the cloud host sends the response result to the AI ​​agent module via the cloud desktop communication protocol. The AI ​​agent module outputs the response result through the user interface located in the local environment.

[0123] In one embodiment of this disclosure, a user equipment is provided. FIG3 is a schematic diagram of a user equipment according to an embodiment of this disclosure. As shown in FIG3, the user equipment includes:

[0124] AI agent module 32 is configured to acquire AI interaction instructions and determine the target AI module for processing the AI ​​interaction instructions from multiple AI modules based on the network environment of the user device. The multiple AI modules include AI modules located in the cloud environment and AI modules located in the local environment.

[0125] AI agent module 32 is configured to implement the above-described method embodiments and exemplary implementations, and will not be repeated hereafter. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function.

[0126] It should be noted that the above modules can be implemented using software or hardware.

[0127] Through the embodiments of this disclosure, the AI ​​agent module can intelligently determine the target AI module in the corresponding environment based on the network environment of the user device, so that AI modules in different environments can process AI interaction commands accordingly, which enhances the interaction between inside and outside the cloud. This solves the problem of inconsistent user experience of AI assistants in cloud computers in cloud environments and outside the cloud in related technologies, making the user experience of AI assistants consistent in cloud environments and improving the user experience.

[0128] In one embodiment of this disclosure, a control system is provided. Figure 4 is a schematic diagram of a control system according to an embodiment of this disclosure. As shown in Figure 4, the control system includes:

[0129] The user equipment 42 is located in a local environment and at least one cloud host 44 is located in a cloud environment. The user equipment includes an AI agent module 422 and an AI module 424 located in the local environment. The cloud host includes an AI module 442 located in the cloud environment.

[0130] The AI ​​agent module 422 is configured to acquire AI interaction instructions and determine the target AI module for processing the AI ​​interaction instructions from multiple AI modules based on the network environment of the user device. The multiple AI modules include the AI ​​module located in the cloud environment and the AI ​​module located in the local environment.

[0131] The local environment refers to the physical environment in which the user directly operates or the network environment in which the user's device is located. For the purposes of this disclosure, the local environment refers to the environment in which the user directly uses their cloud computer device or local device to operate without remotely connecting to the cloud host.

[0132] User equipment refers to terminal devices that users directly interact with and use, including but not limited to computer terminals, mobile terminals, or other devices with computing and network connectivity capabilities.

[0133] A cloud environment refers to a set of infrastructure and services provided by a cloud computing service provider, including but not limited to computing resources (such as cloud servers), storage resources, network resources, and various platform services (such as databases, security services, AI services, etc.). A cloud environment allows users to use these resources on demand without needing to worry about the maintenance and management of the underlying hardware, providing a flexible, scalable, and cost-effective computing solution. For the purposes of this disclosure, a cloud environment refers to a scenario where user devices are remotely connected to cloud servers.

[0134] A cloud server is a basic computing resource provided to users in cloud computing services. It resides in a remote data center and is made available to users via a network. A cloud server can be a virtual machine, a container, or other form of computing instance. Users can create and manage these computing resources in the cloud as needed. Users can interact with cloud servers on their user devices using technologies such as remote desktop protocols. In this disclosure, a local user device can establish a remote connection with one or more cloud servers, thereby enabling it to use the computing resources and AI services on one or more cloud servers.

[0135] In one exemplary embodiment, the AI ​​module can be divided into two parts: an AI assistant (front-end) and an AI service (back-end). The AI ​​assistant is directly displayed on the user interface, providing interactive functions such as receiving user input or outputting responses to AI commands. The AI ​​service is configured to perform related calculations and processing on the AI ​​interaction commands using computing resources, and is invisible to the user.

[0136] Through the embodiments of this disclosure, the AI ​​agent module can intelligently determine the target AI module in the corresponding environment based on the network environment of the user device, so that AI modules in different environments can process AI interaction commands accordingly, which enhances the interaction between inside and outside the cloud. This solves the problem of inconsistent user experience of AI assistants in cloud computers in cloud environments and outside the cloud in related technologies, making the user experience of AI assistants consistent in cloud environments and improving the user experience.

[0137] Figure 5 is a schematic diagram of the overall flow of the control method in one embodiment of this disclosure. As shown in Figure 5, the control method can be implemented in a system that includes a user device located in a local environment and a cloud host (including cloud host 1 and cloud host 2) located in a cloud environment. The user device includes a user interface, an AI agent module, and an AI module located in a local environment. The AI ​​module located in the local environment also includes a local AI assistant and a local AI service. The cloud host includes a user interface (cloud host 1 includes a user interface 1 located in the cloud environment, and cloud host 2 includes a user interface 2 located in the cloud environment) and an AI module located in the cloud environment (cloud host 1 includes an AI module 1 located in the cloud environment, and cloud host 2 includes an AI module 2 located in the cloud environment). The AI ​​module located in the cloud environment also includes a cloud AI assistant (AI module 1 located in the cloud environment includes a cloud AI assistant 1, and AI module 2 located in the cloud environment includes a cloud AI assistant 2) and a cloud AI service (cloud host 1 includes a cloud AI service 1, and cloud host 2 includes a cloud AI service 2).

[0138] The overall process also includes: wake-up process S and interaction process T.

[0139] Depending on the network environment of the user device, the wake-up process can be subdivided into a local wake-up process (SL) and a cloud wake-up process (SC). The local wake-up process wakes up the AI ​​module located in the local environment, while the cloud wake-up process wakes up the AI ​​module located in the cloud environment. In an exemplary embodiment of this disclosure, since the AI ​​modules being woken up in the cloud environment are different, the cloud wake-up process (SC) can be further subdivided into a cloud wake-up process (SC1) for waking up AI module 1 located in the cloud environment, and a cloud wake-up process (SC2) for waking up AI module 2 located in the cloud environment.

[0140] Depending on the network environment of the user device, the interaction process can be subdivided into a local interaction process (TL) and a cloud interaction process (TC). The local interaction process involves interaction with an AI module located in the local environment, while the cloud interaction process involves interaction with an AI module located in the cloud environment. In an exemplary embodiment of this disclosure, since the AI ​​modules performing the interaction in the cloud environment are different, the cloud interaction process (TC) can be further subdivided into a cloud interaction process (TC1) executed by AI module 1 in the cloud environment, and a cloud interaction process (TC2) executed by AI module 2 in the cloud environment.

[0141] The local wake-up process may include the following steps:

[0142] Step S-0: Trigger the AI ​​wake-up command and send the AI ​​wake-up command to the AI ​​agent module.

[0143] In some embodiments, the AI ​​wake-up command is triggered in ways including, but not limited to, the user pressing the AI ​​button module, or the user inputting wake-up information through a user interface (including a user interface in a local environment or a user interface in a cloud environment).

[0144] Step SL-1: The AI ​​agent module sends an AI wake-up command to the local AI assistant and wakes up the local AI assistant.

[0145] In some embodiments, when the computing power of the awakened AI module does not meet the computing power requirements of the AI ​​interaction command, it is determined whether the assistance of the unawakened local AI module is needed. If so, the AI ​​wake-up command is triggered.

[0146] In some embodiments, when the AI ​​agent module responds to the AI ​​wake-up command and determines that the network environment in which the user device is located is a local environment, it sends the AI ​​wake-up command to the local AI assistant and wakes up the local AI assistant.

[0147] The cloud wake-up process may include step S-0 and at least one of the following steps:

[0148] Step SC1-1: The AI ​​agent module sends an AI wake-up command to the cloud AI assistant 1 and wakes up the cloud AI assistant 1.

[0149] Step SC2-1: The AI ​​agent module sends an AI wake-up command to the cloud AI assistant 2 and wakes up the cloud AI assistant 2.

[0150] In some embodiments, when the AI ​​agent module responds to the AI ​​wake-up command and determines that the network environment in which the user device is located is a cloud environment, and the connected cloud host is not cloud host 1, it sends an AI wake-up command to the cloud AI assistant 1 and wakes up the cloud AI assistant 1.

[0151] In some embodiments, when the AI ​​agent module responds to the AI ​​wake-up command and determines that the network environment in which the user device is located is a cloud environment, and the connected cloud host is not cloud host 2, it sends an AI wake-up command to the cloud AI assistant 2 and wakes up the cloud AI assistant 2.

[0152] In some embodiments, when the computing power resources of the awakened AI module do not meet the computing power requirements of the AI ​​interaction command, it is determined whether the assistance of the unawakened cloud AI module is needed. If so, the AI ​​wake-up command is triggered.

[0153] The local interaction process may include the following steps:

[0154] Step T-0: Obtain the interactive information input by the user through the user interface;

[0155] Step T-1: The user interface converts the interactive information into digital signals and then forwards them to the AI ​​agent module for processing;

[0156] In step TL-2, the AI ​​agent module forwards the digital signal to the local AI assistant for processing;

[0157] Step TL-3: The local AI assistant identifies the digital signal, converts the digital signal into a system command that the system can recognize, and forwards it to the local AI service;

[0158] Step TL-4: The local AI service completes the relevant processing according to the system instructions and returns the response result to the local AI assistant;

[0159] Step TL-5: The local AI assistant returns the response result to the user interface located in the local environment;

[0160] In step TL-6, the local AI assistant forwards the digital signal to the AI ​​agent module.

[0161] In some embodiments, when the AI ​​agent module determines that the network environment in which the user device is located is a local environment, it forwards the digital signal to the local AI assistant for processing.

[0162] In some embodiments, step TL-6 is optional, meaning it can be omitted. In an exemplary embodiment, when the computing resources of the awakened AI module do not meet the computing power requirements of the AI ​​interaction instructions, it is determined whether the assistance of the awakened local AI module is needed. If so, step TL-6 is executed.

[0163] After waking up the cloud AI assistant 1, a cloud interaction process can be executed. The cloud interaction process with the cloud host 1 may include the following steps:

[0164] In step TC1-2, the AI ​​agent module forwards the digital signal to the cloud AI assistant 1 for processing;

[0165] In steps TC1-3, the cloud AI assistant 1 identifies the digital signal, converts the digital signal into a system command that the system can recognize, and forwards it to the cloud AI service 1.

[0166] In steps TC1-4, Cloud AI Service 1 completes the relevant processing according to the system instructions and returns the response result to Cloud AI Assistant 1;

[0167] In steps TC1-5, Cloud AI Assistant 1 returns a response result to the AI ​​Agent Module;

[0168] In steps TC1-6, the AI ​​agent module forwards the response result to the local AI assistant;

[0169] In steps TC1-7, the local AI assistant returns the response result to the user interface located in the local environment.

[0170] In some embodiments, after step TC1-7, step TC1-8 can also be executed, in which the cloud AI assistant 1 forwards the digital signal to the AI ​​agent module. For example, when the computing power resources of the awakened AI module do not meet the computing power requirements of the AI ​​interaction instructions, it is determined whether the assistance of other AI modules is needed. If so, step TC1-8 is executed.

[0171] In some embodiments, after step TC1-4, step TC1-9 can also be executed directly, and the cloud AI assistant 1 returns a response result to the user interface 1 located in the cloud environment.

[0172] In some embodiments, when the AI ​​agent module determines that the network environment in which the user device is located is a cloud environment and the connected host is cloud host 1, it forwards the digital signal to cloud AI assistant 1 for processing.

[0173] After waking up Cloud Assistant 2, a cloud interaction process can be executed. The cloud interaction process with Cloud Host 2 may include the following steps:

[0174] In step TC2-2, the AI ​​agent module forwards the digital signal to the cloud AI assistant 2 for processing.

[0175] In step TC2-3, the cloud AI assistant 2 identifies the digital signal, converts the digital signal into a system command that the system can recognize, and forwards it to the cloud AI service 2;

[0176] In step TC2-4, Cloud AI Service 2 completes the relevant processing according to the system instructions and returns the response result to Cloud AI Assistant 2;

[0177] In step TC2-5, Cloud AI Assistant 2 returns a response result to the AI ​​Agent Module;

[0178] In step TC2-6, the AI ​​agent module forwards the response result to the local AI assistant;

[0179] In step TC2-7, the local AI assistant returns the response result to the user interface located in the local environment.

[0180] In some embodiments, after step TC2-7, step TC2-8 can also be executed, in which the cloud AI assistant 2 forwards the digital signal to the AI ​​agent module. For example, when the computing power resources of the awakened AI module do not meet the computing power requirements of the AI ​​interaction instructions, it is determined whether the assistance of other AI modules is needed. If so, step TC2-8 is executed.

[0181] In some embodiments, after step TC2-4, step TC2-9 can also be executed directly, and the cloud AI assistant 2 returns a response result to the user interface 2 located in the cloud environment.

[0182] In some embodiments, the AI ​​agent module located in the local environment can also directly output the response results obtained from cloud AI assistant 1 or cloud AI assistant 2 to the local user interface, and this disclosure does not limit this.

[0183] In some embodiments, the cloud AI assistant can return relevant feedback results to the AI ​​agent module via the cloud desktop communication protocol. This disclosure does not impose any limitations on this.

[0184] In some embodiments, after performing step T1, the AI ​​agent module may also initiate a service request based on the cloud host connection status, and then perform other steps. This disclosure does not impose any limitations on this.

[0185] In some embodiments, different processes can be executed by combining the above steps. The embodiments will be described in detail below with reference to the above steps.

[0186] Figure 6 is a schematic diagram of the process of waking up an AI module located in the local environment according to an embodiment of the present disclosure. As shown in Figure 6, the process includes the following steps:

[0187] Step S-0: Trigger the AI ​​wake-up command and send the AI ​​wake-up command to the AI ​​agent module.

[0188] In some embodiments, the AI ​​wake-up command is triggered in ways including, but not limited to, the user pressing the AI ​​button module, or the user inputting wake-up information through a user interface (including a user interface in a local environment or a user interface in a cloud environment).

[0189] Step SL-1: The AI ​​agent module sends an AI wake-up command to the local AI assistant and wakes up the local AI assistant.

[0190] In some embodiments, when the computing power of the awakened AI module does not meet the computing power requirements of the AI ​​interaction command, it is determined whether the assistance of the unawakened local AI module is needed. If so, the AI ​​wake-up command is triggered.

[0191] In some embodiments, when the AI ​​agent module responds to the AI ​​wake-up command and determines that the network environment in which the user device is located is a local environment, it sends the AI ​​wake-up command to the local AI assistant and wakes up the local AI assistant.

[0192] In this embodiment, the design of the AI ​​button module simplifies the AI ​​module wake-up process. Users can quickly enjoy AI services without cumbersome voice wake-up or manual operation. At the same time, the AI ​​wake-up button settings can be configured through software, eliminating the need for users to purchase a customized keyboard that supports AI buttons, thus reducing user costs.

[0193] Figure 7 is a schematic diagram of the process of waking up an AI module located in a cloud environment according to an embodiment of the present disclosure. As shown in Figure 7, the process includes the following steps:

[0194] Step S-0: Trigger the AI ​​wake-up command and send the AI ​​wake-up command to the AI ​​agent module.

[0195] In some embodiments, the AI ​​wake-up command is triggered in ways including, but not limited to, the user pressing the AI ​​button module, or the user inputting wake-up information through a user interface (including a user interface in a local environment or a user interface in a cloud environment).

[0196] After performing step S-0 above, at least one of the following steps can be performed:

[0197] Step SC1-1: The AI ​​agent module sends an AI wake-up command to the cloud AI assistant 1 and wakes up the cloud AI assistant 1.

[0198] Step SC2-1: The AI ​​agent module sends an AI wake-up command to the cloud AI assistant 2 and wakes up the cloud AI assistant 2.

[0199] In some embodiments, when the AI ​​agent module responds to the AI ​​wake-up command and determines that the network environment in which the user device is located is a cloud environment, and the connected cloud host is not cloud host 1, it sends an AI wake-up command to the cloud AI assistant 1 and wakes up the cloud AI assistant 1.

[0200] In some embodiments, when the AI ​​agent module responds to the AI ​​wake-up command and determines that the network environment in which the user device is located is a cloud environment, and the connected cloud host is not cloud host 2, it sends an AI wake-up command to the cloud AI assistant 2 and wakes up the cloud AI assistant 2.

[0201] In some embodiments, when the computing power of the awakened AI module does not meet the computing power requirements of the AI ​​interaction command, it is determined whether the assistance of the unawakened cloud AI module is needed. If so, the AI ​​wake-up command is triggered.

[0202] In this embodiment, the AI ​​agent module wakes up the cloud AI assistant, so that the application of the AI ​​assistant is not limited to a single environment, such as the use of the AI ​​assistant in a single local environment or the use of the AI ​​assistant in a single cloud environment, thus enhancing the interaction between inside and outside the cloud.

[0203] Figure 8 is a flowchart illustrating the interaction of AI modules in a local environment according to an embodiment of this disclosure. As shown in Figure 8, the process includes the following steps:

[0204] Step T-0: Obtain the interactive information input by the user through the user interface;

[0205] Step T-1: The user interface converts the interactive information into digital signals and then forwards them to the AI ​​agent module for processing;

[0206] In step TL-2, the AI ​​agent module forwards the digital signal to the local AI assistant for processing;

[0207] Step TL-3: The local AI assistant identifies the digital signal, converts the digital signal into a system command that the system can recognize, and forwards it to the local AI service;

[0208] Step TL-4: The local AI service completes the relevant processing according to the system instructions and returns the response result to the local AI assistant;

[0209] Step TL-5: The local AI assistant returns the response result to the user interface located in the local environment;

[0210] In step TL-6, the local AI assistant forwards the digital signal to the AI ​​agent module.

[0211] In some embodiments, when the AI ​​agent module determines that the network environment in which the user device is located is a local environment, it forwards the digital signal to the local AI assistant for processing.

[0212] In some embodiments, step TL-6 is optional, meaning it can be omitted. In an exemplary embodiment, when the computing resources of the awakened AI module do not meet the computing power requirements of the AI ​​interaction instructions, it is determined whether the assistance of the awakened local AI module is needed. If so, step TL-6 is executed.

[0213] In this embodiment, digital signals and system commands are different manifestations of AI interaction commands.

[0214] In this embodiment, the local AI assistant is triggered by the AI ​​agent module, and the AI ​​service is used to process the user's input interaction information in a timely manner, which can ensure a smooth user experience.

[0215] Figure 9 is a schematic diagram of the interaction of AI modules in a cloud environment according to an embodiment of this disclosure. As shown in Figure 9, the process includes the following steps:

[0216] Step T-0: Obtain the interactive information input by the user through the user interface;

[0217] Step T-1: The user interface converts the interactive information into digital signals and then forwards them to the AI ​​agent module for processing;

[0218] In step TC1-2, the AI ​​agent module forwards the digital signal to the cloud AI assistant 1 for processing;

[0219] In steps TC1-3, the cloud AI assistant 1 identifies the digital signal, converts the digital signal into a system command that the system can recognize, and forwards it to the cloud AI service 1.

[0220] In steps TC1-4, Cloud AI Service 1 completes the relevant processing according to the system instructions and returns the response result to Cloud AI Assistant 1;

[0221] In steps TC1-5, Cloud AI Assistant 1 returns a response result to the AI ​​Agent Module;

[0222] In steps TC1-6, the AI ​​agent module forwards the response result to the local AI assistant;

[0223] In steps TC1-7, the local AI assistant returns the response result to the user interface located in the local environment.

[0224] In some embodiments, after step TC1-4, step TC1-9 can also be executed directly, and the cloud AI assistant 1 returns a response result to the user interface 1 located in the cloud environment.

[0225] In some embodiments, when the AI ​​agent module determines that the network environment in which the user device is located is a cloud environment and the connected host is cloud host 1, it forwards the digital signal to cloud AI assistant 1 for processing.

[0226] In some embodiments, when the AI ​​agent module determines that the network environment in which the user device is located is a cloud environment and the connected host is cloud host 1, it forwards the digital signal to cloud AI assistant 1 for processing.

[0227] In this embodiment, the cloud AI assistant is triggered through the AI ​​agent module, and the AI ​​service is used to process the interactive information input by the user in a timely manner. Seamless switching of the AI ​​module can ensure that the user experience is consistent both inside and outside the cloud.

[0228] Figure 10 is a schematic diagram of the interaction between an AI module located in a local environment and an AI module located in a cloud environment according to an embodiment of this disclosure. As shown in Figure 10, the process includes the following steps:

[0229] Step T-0: Obtain the interactive information input by the user through the user interface;

[0230] Step T-1: The user interface converts the interactive information into digital signals and then forwards them to the AI ​​agent module for processing;

[0231] In step TL-2, the AI ​​agent module forwards the digital signal to the local AI assistant for processing;

[0232] Step TL-3: The local AI assistant identifies the digital signal, converts the digital signal into a system command that the system can recognize, and forwards it to the local AI service;

[0233] Step TL-4: The local AI service completes the relevant processing according to the system instructions and returns the response result to the local AI assistant;

[0234] Step TL-5: The local AI assistant returns the response result to the user interface located in the local environment;

[0235] Step TL-6: The local AI assistant forwards the digital signal to the AI ​​agent module;

[0236] Step SC1-1: The AI ​​agent module sends an AI wake-up command to the cloud AI assistant 1 and wakes up the cloud AI assistant 1.

[0237] In step TC1-2, the AI ​​agent module forwards the digital signal to the cloud AI assistant 1 for processing;

[0238] In steps TC1-3, the cloud AI assistant 1 identifies the digital signal, converts the digital signal into a system command that the system can recognize, and forwards it to the cloud AI service 1.

[0239] In steps TC1-4, Cloud AI Service 1 completes the relevant processing according to the system instructions and returns the response result to Cloud AI Assistant 1;

[0240] In steps TC1-5, Cloud AI Assistant 1 returns a response result to the AI ​​Agent Module;

[0241] In steps TC1-6, the AI ​​agent module forwards the response result to the local AI assistant;

[0242] In steps TC1-7, the local AI assistant returns the response result to the user interface located in the local environment.

[0243] In some embodiments, after performing step T1, the AI ​​agent module may also initiate a service request based on the cloud host connection status, and then perform other steps. This disclosure does not impose any limitations on this.

[0244] In this embodiment, the local AI assistant has been activated (refer to the above embodiment for the specific activation process). The AI ​​agent module first forwards the interaction command to the local AI assistant for processing. At this time, the local AI assistant determines that the computing power resources of the local AI service do not meet the requirements of the AI ​​interaction command, and thus executes step TL-6 to forward the digital signal to the AI ​​agent module, which then activates other AI modules to assist. Further, the AI ​​agent module or the local AI assistant can determine that the cloud AI assistant 1, which meets the computing power requirements, can assist in processing or process the AI ​​interaction command independently. At this time, step SC1-1 and the subsequent cloud interaction process will be triggered to activate the cloud AI assistant 1 and process the interaction command through the cloud AI assistant.

[0245] In this embodiment, the local AI assistant forwards digital signals to the cloud AI assistant 1 for processing through the AI ​​agent module, realizing intelligent judgment and switching between inside and outside the cloud, strengthening the interaction between inside and outside the cloud, and ensuring that users get the same experience in different environments.

[0246] In this embodiment of the disclosure, the unified AI capabilities that integrate both in-cloud and out-of-cloud environments are fully utilized, which can simultaneously support multiple scenarios such as office work, entertainment, and education, and AI services can be accessed with a single click.

[0247] For example, in an office setting, a cloud AI assistant can act as a meeting assistant, helping users record key points, schedule appointments, and provide translation services, thereby improving work efficiency and collaboration experience. Furthermore, by continuously learning users' interaction habits and behavioral patterns, it can provide personalized recommendations and services, such as adjusting schedules based on users' work habits and pushing important product information based on users' interests.

[0248] For example, in an entertainment environment, a local AI assistant can recommend suitable entertainment content such as music, movies, and games based on the user's interests and preferences, thereby enhancing the user's entertainment experience.

[0249] For example, in a gaming environment, an AI assistant can act as a gaming companion, interacting with users, providing game guides and tips, and enhancing the fun and interactivity of the game.

[0250] For example, in educational settings, a local AI assistant can function as an intelligent tutoring system, providing students with personalized learning plans and tutoring services to help solve learning difficulties and improve learning efficiency.

[0251] The control methods in this disclosure will be explained in detail below in conjunction with specific application scenarios.

[0252] Figure 11 is a schematic diagram of a user device in a local environment performing remote work according to an embodiment of this disclosure. As shown in Figure 11, the process includes the following steps:

[0253] Step S1102: Wake up the local AI assistant;

[0254] Step S1104: Interact with the local AI assistant;

[0255] Step S1106: Perform relevant processing according to the AI ​​interaction instructions and return the results to the user interface located in the local environment;

[0256] Step S1108: Obtain cloud computing power support.

[0257] In some embodiments, step S1102 may further include the following steps:

[0258] Step S-0: The user presses the AI ​​button module to trigger the AI ​​wake-up command;

[0259] In step SL-1, the AI ​​agent module responds to the AI ​​wake-up command, determines that the network environment in which the user device is located is the local environment, sends the AI ​​wake-up command to the local AI assistant, and wakes up the local AI assistant.

[0260] In this embodiment, the AI ​​button module is used to trigger the AI ​​wake-up signal. If the current user device is in a local environment, the local AI assistant will be woken up immediately.

[0261] In some embodiments, step S1104 may further include the following steps:

[0262] Step T-0: Obtain the interactive information input by the user through the user interface;

[0263] Step T-1: The user interface converts the interactive information into digital signals and then forwards them to the AI ​​agent module for processing;

[0264] In step TL-2, the AI ​​agent module determines that the network environment of the user device is a local environment and forwards the digital signal to the local AI assistant for processing.

[0265] In some embodiments, the interactive information input by the user may be text entered via a keyboard, or it may be user voice, images, videos, or other information. This disclosure does not limit the type of interactive information.

[0266] In this embodiment, the user inputs interactive information to the user interface located in the local environment via voice commands. The voice commands instruct the local AI assistant to prepare meeting materials and enable the real-time translation function.

[0267] In some embodiments, step S1106 may further include the following steps:

[0268] Step TL-3: The local AI assistant identifies the digital signal, converts the digital signal into a system command that the system can recognize, and forwards it to the local AI service;

[0269] Step TL-4: The local AI service completes the relevant processing according to the system instructions and returns the response result to the local AI assistant;

[0270] In step TL-5, the local AI assistant returns the response result to the user interface located in the local environment.

[0271] In this embodiment, based on voice commands, the local AI assistant retrieves relevant meeting materials from the cloud computer's file system and organizes them into an easy-to-read format. Simultaneously, the local AI assistant connects to a cloud-based translation service to perform multilingual translation. During the meeting, the local AI assistant can also record key points in real time and instantly convert content requiring translation, displaying it to the user through the cloud computer's user interface. Furthermore, users can adjust their interaction with the local AI assistant via voice or text, for example, adjusting translation settings or adding notes.

[0272] In some embodiments, step S1108 may include the following steps:

[0273] Step TL-6: The local AI assistant forwards the digital signal to the AI ​​agent module;

[0274] Step SC1-1: The AI ​​agent module sends an AI wake-up command to the cloud AI assistant 1 and wakes up the cloud AI assistant 1.

[0275] In step TC1-2, the AI ​​agent module forwards the digital signal to the cloud AI assistant 1 for processing;

[0276] In steps TC1-3, the cloud AI assistant 1 identifies the digital signal, converts the digital signal into a system command that the system can recognize, and forwards it to the cloud AI service 1.

[0277] In steps TC1-4, Cloud AI Service 1 completes the relevant processing according to the system instructions and returns the response result to Cloud AI Assistant 1;

[0278] In steps TC1-5, Cloud AI Assistant 1 returns a response result to the AI ​​Agent Module;

[0279] In steps TC1-6, the AI ​​agent module forwards the response result to the local AI assistant;

[0280] In steps TC1-7, the local AI assistant returns the response result to the user interface located in the local environment.

[0281] In some embodiments, steps TC1-4 can also be directly executed, with the cloud AI assistant 1 returning a response result to the user interface 1 located in the cloud environment.

[0282] In other embodiments, the AI ​​agent module may also directly output the response results obtained from the cloud AI assistant to the local user interface, which is not a limitation of this disclosure.

[0283] In some embodiments, when the cloud computer's computing power is insufficient to process large amounts of data or perform complex calculations (such as real-time image recognition, sentiment analysis, etc.) during a meeting, the local AI assistant can wake up the cloud AI assistant with GPU capabilities through the AI ​​agent module, and call cloud AI services through interaction with the cloud AI assistant. The cloud AI services can then call cloud GPU computing power to improve processing speed and ensure the smooth running of the meeting.

[0284] Figure 12 is a schematic diagram of a process for remote office work performed by a cloud host located in a cloud environment according to an embodiment of this disclosure. As shown in Figure 12, the process includes the following steps:

[0285] Step S1202: Wake up the cloud AI assistant 1;

[0286] Step S1204: Interact with Cloud AI Assistant 1;

[0287] Step S1206: Obtain local computing power support;

[0288] Step S1208: Obtain additional computing power support.

[0289] In this embodiment, steps S1206 and S1208 are optional steps. When cloud host 1 cannot meet the AI ​​computing power requirements, it can obtain computing power support from local or other cloud hosts, and multiple cloud hosts or local user devices can cooperate with cloud hosts to complete the corresponding AI interaction.

[0290] In some embodiments, step S1202 may include the following steps:

[0291] Step S-0: The user presses the AI ​​button module to trigger the AI ​​wake-up command;

[0292] In step SC1-1A, the AI ​​agent module responds to the AI ​​wake-up command. When it determines that the network environment in which the user device is located is a cloud environment and the connected cloud host is not cloud host 1, it sends an AI wake-up command to cloud AI assistant 1 and wakes up cloud AI assistant 1.

[0293] In this embodiment, the AI ​​button module is used to trigger the AI ​​wake-up signal. If the current user device is in a cloud environment, the cloud AI assistant will be woken up immediately.

[0294] In some embodiments, step S1204 may further include the following steps:

[0295] Step T-0: Obtain the interactive information input by the user through the user interface;

[0296] Step T-1: The user interface converts the interactive information into digital signals and then forwards them to the AI ​​agent module for processing;

[0297] In step TC1-2, when the AI ​​agent module determines that the network environment of the user device is a cloud environment, it forwards the digital signal to the cloud AI assistant 1 for processing.

[0298] In steps TC1-3, the cloud AI assistant 1 identifies the digital signal, converts the digital signal into a system command that the system can recognize, and forwards it to the cloud AI service 1.

[0299] In steps TC1-4, Cloud AI Service 1 completes the relevant processing according to the system instructions and returns the response result to Cloud AI Assistant 1;

[0300] In steps TC1-5, Cloud AI Assistant 1 returns a response result to the AI ​​Agent Module;

[0301] In steps TC1-6, the AI ​​agent module forwards the response result to the local AI assistant;

[0302] In steps TC1-7, the local AI assistant returns the response result to the user interface located in the local environment.

[0303] In some embodiments, after steps TC1-4, the cloud AI assistant 1 can directly output the response result through the user interface 1.

[0304] In some embodiments, the interactive information obtained from user input may be text entered via keyboard, or it may be user voice, images, videos, or other information. This disclosure does not limit the type of interactive information.

[0305] In this embodiment, the user requests the cloud AI assistant to prepare meeting materials and enable the real-time translation function. This requires inputting interactive information through voice commands on the user interface located in the local environment. The user interface converts the interactive information into data signals and forwards them to the AI ​​agent module. The AI ​​agent module then transmits the data to the cloud AI assistant via a remote desktop protocol.

[0306] In this embodiment, based on voice commands, the cloud AI assistant 1 retrieves relevant meeting materials from the file system of the cloud host 1 on the cloud computer and organizes them into an easy-to-read format. Simultaneously, the cloud AI assistant connects to a cloud-based translation service to perform multilingual translation. During the meeting, the cloud AI assistant can also record key points in real time and instantly convert content requiring translation, displaying it to the user through the cloud host's user interface. Furthermore, users can adjust their interaction with the local AI assistant via voice or text, for example, adjusting translation settings or adding notes.

[0307] In some embodiments, step S1206 may include the following steps:

[0308] In steps TC1-8, the cloud AI assistant 1 forwards the digital signal to the AI ​​agent module;

[0309] In step TL-2, the AI ​​agent module forwards the digital signal to the local AI assistant for processing;

[0310] Step TL-3: The local AI assistant identifies the digital signal, converts the digital signal into a system command that the system can recognize, and forwards it to the local AI service;

[0311] Step TL-4: The local AI service completes the relevant processing according to the system instructions and returns the response result to the local AI assistant;

[0312] In step TL-5, the local AI assistant returns the response result to the user interface located in the local environment.

[0313] In this embodiment, when the cloud host has insufficient computing power to process large amounts of data or perform complex calculations (such as real-time image recognition and sentiment analysis) during a meeting, the cloud AI assistant can wake up the local AI assistant through the AI ​​agent module and call local AI services to improve processing speed and ensure the smooth running of the meeting.

[0314] In some embodiments, step S1208 may include the following steps:

[0315] Step TL-6: The local AI assistant forwards the digital signal to the AI ​​agent module;

[0316] In step TC2-2, the AI ​​agent module forwards the digital signal to the cloud AI assistant 2 for processing.

[0317] In step TC2-3, the cloud AI assistant 2 identifies the digital signal, converts the digital signal into a system command that the system can recognize, and forwards it to the cloud AI service 2;

[0318] In step TC2-4, Cloud AI Service 2 completes the relevant processing according to the system instructions and returns the response result to Cloud AI Assistant 2;

[0319] In step TC2-5, Cloud AI Assistant 2 returns a response result to the AI ​​Agent Module;

[0320] In step TC2-6, the AI ​​agent module forwards the response result to the local AI assistant;

[0321] In step TC2-7, the local AI assistant returns the response result to the user interface located in the local environment.

[0322] In some embodiments, after steps TC2-4, the cloud AI assistant 2 can directly output the response result through the user interface 2.

[0323] In this embodiment, if there is still insufficient computing power in the cloud host after executing step TL-5 (such as when the user's interaction information requests the generation of an image or video), the cloud AI assistant can be woken up through the AI ​​agent module, and the GPU capabilities of other cloud hosts can be called through the interaction with the cloud AI assistant to improve the processing speed and ensure the smooth progress of the meeting.

[0324] Figure 13 is a schematic diagram of a user device performing online learning in a local environment according to an embodiment of this disclosure. As shown in Figure 13, the process includes the following steps:

[0325] Step S1302: Wake up the local AI assistant;

[0326] Step S1304: Interact with the local AI assistant;

[0327] Step S1306: Obtain cloud computing power support.

[0328] In some embodiments, step S1302 may include the following steps:

[0329] Step S-0: The user presses the AI ​​button module to trigger the AI ​​wake-up command;

[0330] In step SL-1, the AI ​​agent module responds to the AI ​​wake-up command, determines that the network environment in which the user device is located is the local environment, sends the AI ​​wake-up command to the local AI assistant, and wakes up the local AI assistant.

[0331] In this embodiment, the use of the AI ​​button module makes the wake-up mechanism intuitive and effective. That is, once the AI ​​button module is pressed, the AI ​​wake-up command is triggered. If the current user device is in the local environment, the local AI assistant is immediately woken up.

[0332] In some embodiments, step S1304 may include the following steps:

[0333] Step T-0: Obtain the interactive information input by the user through the user interface;

[0334] Step T-1: The user interface converts the interactive information into digital signals and then forwards them to the AI ​​agent module for processing;

[0335] Step TL-2: The AI ​​agent module determines that the network environment of the user device is a local environment and forwards the digital signal to the local AI assistant for processing.

[0336] Step TL-3: The local AI assistant identifies the digital signal, converts the digital signal into a system command that the system can recognize, and forwards it to the local AI service;

[0337] Step TL-4: The local AI service completes the relevant processing according to the system instructions and returns the response result to the local AI assistant;

[0338] In step TL-5, the local AI assistant returns the response result to the user interface located in the local environment.

[0339] In some embodiments, the interactive information obtained from user input may be text entered via keyboard, or it may be user voice, images, videos, or other information. This disclosure does not limit the type of interactive information.

[0340] In this embodiment, the user inputs a mathematical problem into the user interface located locally via an image. The local AI assistant utilizes local AI services to connect to a large cloud-based model and perform in-depth analysis of the mathematical problem, generating detailed solution steps and explanations of the thought process. Simultaneously, the local AI assistant records the user's learning progress and habits, providing personalized learning suggestions and practice questions. This approach effectively combines the processing resources of both the cloud and local environments, improving the user experience.

[0341] In some embodiments, step S1306 may further include the following steps:

[0342] Step TL-6: The local AI assistant forwards the digital signal to the AI ​​agent module;

[0343] Step SC1-1: The AI ​​agent module sends an AI wake-up command to the cloud AI assistant 1 and wakes up the cloud AI assistant 1.

[0344] In step TC1-2, the AI ​​agent module forwards the digital signal to the cloud AI assistant 1 for processing;

[0345] In steps TC1-3, the cloud AI assistant 1 identifies the digital signal, converts the digital signal into a system command that the system can recognize, and forwards it to the cloud AI service 1.

[0346] In steps TC1-4, Cloud AI Service 1 completes the relevant processing according to the system instructions and returns the response result to Cloud AI Assistant 1;

[0347] In steps TC1-5, Cloud AI Assistant 1 returns a response result to the AI ​​Agent Module;

[0348] In steps TC1-6, the AI ​​agent module forwards the response result to the local AI assistant;

[0349] In steps TC1-7, the local AI assistant returns the response result to the user interface located in the local environment.

[0350] In some embodiments, after steps TC1-4, the cloud AI assistant 1 can directly output the response result to the user interface 1. In other embodiments, after steps TC1-5, the AI ​​agent module can also directly output the response result to the local user interface, and this disclosure does not impose any limitations on this.

[0351] In this embodiment, when the user device located in the local environment has insufficient computing power to perform complex graphics processing or simulation calculations during the problem-solving process, the local AI assistant can trigger an AI wake-up command through the AI ​​agent module and wake up the cloud AI assistant with GPU capabilities. Through interaction with the cloud AI assistant, the cloud AI service can be invoked to achieve seamless access to cloud GPU computing power.

[0352] Embodiments of this disclosure also provide a computer-readable storage medium storing a computer program configured to perform the steps in any of the above method embodiments when executed.

[0353] In one exemplary embodiment, the aforementioned computer-readable storage medium may include, but is not limited to, various media capable of storing computer programs, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard disk, magnetic disk, or optical disk.

[0354] Embodiments of this disclosure also provide an electronic device including a memory and a processor, the memory storing a computer program and the processor being configured to run the computer program to perform the steps in any of the above method embodiments.

[0355] In one exemplary embodiment, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor and the input / output device is connected to the processor.

[0356] Specific examples in this embodiment can be found in the examples described in the above embodiments and exemplary implementations, and will not be repeated here.

[0357] It is obvious to those skilled in the art that the modules or steps of this disclosure described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. They can be implemented using computer-executable program code, and thus can be stored in a storage device for execution by a computing device. In some cases, the steps shown or described can be performed in a different order than those presented herein, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, this disclosure is not limited to any particular combination of hardware and software.

[0358] The above description is merely an exemplary embodiment of this disclosure and is not intended to limit this disclosure. Various modifications and variations can be made to this disclosure by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A control method, comprising: Obtain AI interaction commands; The target AI module for processing the AI ​​interaction command is determined from multiple AI modules based on the network environment of the user device, wherein the multiple AI modules include: AI modules located in the cloud environment and AI modules located in the local environment.

2. The method according to claim 1, wherein, The acquisition of AI interaction commands includes: The AI ​​interaction command input through the AI ​​button module is obtained, wherein the AI ​​button module is a preset physical button or a virtual button.

3. The method according to claim 1, wherein, The acquisition of AI interaction commands includes: The AI ​​interaction command input through the interface interaction module is obtained, wherein the interface interaction module includes a user interface located in the cloud environment and / or a user interface located in the local environment.

4. The method according to claim 1, wherein, The step of determining the target AI module for processing the AI ​​interaction command from multiple AI modules based on the user device's network environment includes: In response to the AI ​​interaction command, determine the network environment of the user device; The first AI module is determined based on the network environment of the user equipment; The first AI module is identified as the target AI module.

5. The method according to claim 4, wherein, The step of determining the first AI module based on the network environment of the user equipment includes: When the network environment is the cloud environment, the AI ​​module located in the cloud environment is identified as the first AI module; or, When the network environment is the local environment, the AI ​​module located in the local environment is identified as the first AI module.

6. The method according to claim 5, wherein, When the network environment is the cloud environment, determining the AI ​​module located in the cloud environment as the first AI module includes: When the network environment is the cloud environment, a target cloud host connected to the user equipment is determined from multiple cloud hosts; The AI ​​module deployed on the target cloud host is identified as the first AI module.

7. The method according to claim 4, wherein, The step of determining the target AI module for processing the AI ​​interaction command from multiple AI modules based on the network environment of the user equipment further includes: In response to the AI ​​interaction command, determine the computing resources of the multiple AI modules; The second AI module is determined based on the computing power resources of the multiple AI modules and the AI ​​interaction instructions; The first AI module and the second AI module are identified as the target AI module.

8. The method according to claim 7, wherein, The step of determining the second AI module based on the computing resources of the plurality of AI modules and the AI ​​interaction instructions includes: In response to the fact that the computing power resources of the first AI module do not meet the computing power requirements of the AI ​​interaction command, the AI ​​module whose computing power resources meet the computing power requirements of the AI ​​interaction command is determined from the plurality of AI modules as the second AI module.

9. The method according to claim 1, wherein, The step of determining the target AI module for processing the AI ​​interaction command from multiple AI modules based on the user equipment's network environment includes: In response to the AI ​​interaction command, the network environment of the user device and the computing resources of the multiple AI modules are determined; The third AI module is determined from the multiple AI modules based on the network environment of the user equipment, the computing resources of the multiple AI modules, and the AI ​​interaction instructions. The third AI module is identified as the target AI module.

10. The method according to claim 1, wherein, The method further includes: The AI ​​interaction command is sent to the target AI module located in the cloud environment, and the response result of the AI ​​interaction command is received from the target AI module located in the cloud environment.

11. The method according to claim 1, wherein, The method further includes: The AI ​​interaction command is sent to the target AI module located in the cloud environment, and the response result of the AI ​​interaction command is output through the user interface of the cloud environment.

12. The method according to claim 1, wherein, The method further includes: The AI ​​interaction command is sent to the target AI module located in the local environment, and the response result of the AI ​​interaction command is received from the target AI module located in the local environment.

13. The method according to claim 1, wherein, The method further includes: The AI ​​interaction command is sent to the target AI module located in the local environment, and the response result of the AI ​​interaction command is output through the user interface of the local environment.

14. The method according to claim 1, wherein, The method further includes at least one of the following: The response result of the AI ​​interaction command is sent to the AI ​​module located in the local environment, so that the AI ​​module located in the local environment outputs the response result through the user interface located in the local environment; The response result of the AI ​​interaction command is sent to the AI ​​module located in the cloud environment, so that the AI ​​module located in the cloud environment outputs the response result through the user interface located in the cloud environment.

15. A user equipment, comprising: The AI ​​agent module is configured to acquire AI interaction commands and determine the target AI module for processing the AI ​​interaction commands from multiple AI modules based on the network environment of the user device. The multiple AI modules include AI modules located in a cloud environment and AI modules located in a local environment.

16. A control system, comprising: The system includes a user device located in a local environment and at least one cloud host located in a cloud environment. The user device includes an AI agent module and an AI module located in the local environment. The cloud host includes an AI module located in the cloud environment. The AI ​​agent module is configured to acquire AI interaction instructions and determine the target AI module for processing the AI ​​interaction instructions from multiple AI modules based on the network environment of the user device. The multiple AI modules include the AI ​​module located in the cloud environment and the AI ​​module located in the local environment.

17. A computer-readable storage medium, wherein, The storage medium stores a computer program, wherein the computer program, when executed by a processor, implements the steps of the method described in any one of claims 1 to 14.

18. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein, When the processor executes the computer program, it implements the steps of the method described in any one of claims 1 to 14.

19. A computer program product comprising a computer program that, when executed by a processor, implements the steps of the method described in any one of claims 1 to 14.

Images