Communication method and related apparatus

Abstract

A communication system, a communication method, and a related apparatus. The method comprises: receiving first information, wherein the first information comprises a user intent collected by a terminal device; acquiring media vector information of the terminal device from a second network element, wherein the media vector information is associated with user behavior information collected by the terminal device; and on the basis of the media vector information and the first information, invoking a large multimodal model to acquire a recognized user intent and / or a response related to the recognized user intent. A network element in a network side invokes the large multimodal model to acquire the user intent in the first information and / or the response related to the user intent, improving support for a media information interaction function, and further improving intent recognition accuracy.

Description

Communication method and related apparatus

[0001] This application claims priority from the Chinese patent application No. 202410837283.9 filed on June 25, 2024, and entitled "Communication method and related apparatus", the content of which is incorporated herein by reference in its entirety. TECHNICAL FIELD

[0002] The present application relates to the field of artificial intelligence (AI) technology, and in particular, relates to a communication method and related apparatus. BACKGROUND

[0003] With the development of speech recognition technology, a model corresponding to a voice interaction function is set in a terminal device (for example, a mobile phone, etc.). A user can wake up a voice assistant through a voice instruction, and realize voice interaction operations such as opening a target application, playing music, and querying the weather through the voice assistant. However, limited by the memory and computing power of the terminal device, the support of the model for the voice interaction function is also limited. SUMMARY

[0004] Embodiments of the present application disclose a communication method and related apparatus, which call a multi-modal large model in a network element to obtain a user intent of first information and / or a response related to the user intent, which can improve the support for media information interaction function, and also can improve the accuracy of intent recognition.

[0005] In a first aspect, embodiments of the present application disclose a communication system, comprising a first network element and a second network element, wherein: the second network element is configured to store media vector information of a terminal device, the media vector information being associated with behavior information of a user collected by the terminal device; the first network element is configured to receive first information, the first information comprising a user intent collected by the terminal device; the first network element is further configured to obtain the media vector information from the second network element; and the first network element is further configured to call a multi-modal large model according to the media vector information and the first information to obtain an identified user intent and / or a response related to the identified user intent. In this way, the multi-modal large model is called by a network element in the network side to obtain a user intent of first information and / or a response related to the user intent, which can improve the support for media information interaction function. And the media vector information associated with the behavior information of the user is used to identify the user intent, which can also improve the accuracy of intent recognition.

[0006] In the embodiments of the present application, the large multimodal model (LMM) can include a large language model (LLM), or can include a large video model (LVM), etc., which is not limited herein. The large multimodal model is used to generate a user portrait of a user and / or evaluation information of the user portrait based on behavior information of the user.

[0007] In a possible example, the system further includes a third network element configured to configure resources of the first network element and resources of the second network element for the terminal device. In this way, the first network element and the second network element can be prevented from processing data of multiple users, random data access and leakage caused by non-isolation of user data of a part of business system can be solved, user privacy can be protected, and requirements of regulations (such as general data protection regulation (GDPR)) can be met.

[0008] In a possible example, the third network element is further configured to receive a first registration request of the first network element and / or a second registration request of the second network element, the first registration request being used to register a resource identifier of the first network element, and the second registration request being used to register a resource identifier of the second network element; and the third network element is further configured to send, to the first network element, a confirmation message of the first registration request, and / or send, to the second network element, a confirmation message of the second registration request. In this way, in a case where the third network element has a registration function of the resource identifier, the resources of the first network element can be registered according to the resource identifier of the first network element, and the resources of the second network element can be registered according to the resource identifier of the second network element, and after successful registration, the corresponding confirmation messages are sent.

[0009] Optionally, the resource identifier can be a universal resource identifier (URI). The resource identifier of the first network element is different from the resource identifier of the second network element. The resource identifier of the first network element is used to determine the resources of the first network element, and the resource identifier of the second network element is used to determine the resources of the second network element.

[0010] Optionally, the resource identifier can be referred to as a service resource identifier, such as a service uri.

[0011] Optionally, the registration function of the resource identifier is referred to as a service registration function.

[0012] In a possible example, the third network element is further configured to send a first check message to the first network element, the first check message being used to confirm the resource identifier of the first network element; and / or send a second check message to the second network element, the second check message being used to confirm the resource identifier of the second network element; and receive a confirmation message of the first check message of the first network element; and / or receive a confirmation message of the second check message of the second network element. In this way, in the case that the third network element does not have the registration function of the resource identifier, the first network element can be checked according to the resource identifier of the first network element, and the second network element can be checked according to the resource identifier of the second network element, and a successful response, i.e., a confirmation message, is returned by the corresponding network element after the check is successful.

[0013] Optionally, the confirmation message of the first check message includes the resource identifier of the first network element, and the confirmation message of the second check message includes the resource identifier of the second network element.

[0014] In a possible example, the first network element is configured to invoke the multi-modal large model according to the media vector information and the first information to obtain the identified user intent and / or the response related to the identified user intent, including: obtaining a first prompt word according to the first information and the media vector information; sending a first request including the first prompt word to the multi-modal large model, the first request being used to request to identify the user intent; and receiving the identified user intent and / or the response related to the identified user intent from the multi-modal large model. In this way, the first network element can interact with the multi-modal large model, so as to invoke the multi-modal large model according to the media vector information and the first information to obtain the identified user intent and / or the response related to the identified user intent.

[0015] In a possible example, the first network element is further configured to receive second information, the second information including behavior information of the user collected by the terminal device; and invoke the multi-modal large model according to the second information to obtain the media vector information; and send a storage request including the media vector information to the second network element, the storage request being used to request to store the media vector information.

[0016] In a possible example, the first network element is further configured to invoke the multi-modal large model according to the second information to obtain the media vector information, including: obtaining a second prompt word according to the second information; sending a second request including the second prompt word to the multi-modal large model, the second request being used to request to obtain the media vector information; and receiving the media vector information from the multi-modal large model. In this way, it is beneficial for subsequent obtaining of the user intent and / or the response related to the user intent through the media vector information of the terminal device.

[0017] In a possible example, the media vector information includes a user portrait of the terminal device and evaluation information of the user portrait, and an evaluation dimension of the evaluation information includes at least one of importance, recency, and relevance. In this way, the long-term memory corresponding to the most important, most relevant, and most recent user portrait can be selected for intent recognition, which helps to improve the efficiency and accuracy of intent recognition.

[0018] In a possible example, the first network element is further configured to invoke the multi-modal large model to reflect on the media vector information. In this way, the media vector information is reflected, and then the user portrait and the evaluation information of the user portrait can be reflected, which helps to improve the accuracy of user intent recognition.

[0019] In a possible example, the first network element and the second network element are further configured to receive verification information of the terminal device, verify the verification information of the terminal device, and allow a user of the terminal device to access when the verification is successful. In this way, the processing logic and data of the user are isolated by the first network element and the second network element, and the user identity can be verified through the verification of the verification information. When the verification is successful, the user is allowed to access the data, which can solve the problem of random data access and leakage caused by non-isolation of user data in some business systems, thereby protecting the privacy of the user and meeting the requirements of regulations (such as GDPR).

[0020] In a possible example, the first network element is further configured to decrypt the received information and / or encrypt the transmitted information. In this way, the network element receiving the encrypted information can decrypt the received information. By encrypting and / or decrypting the transmission data, the security of the data can be improved.

[0021] In a possible example, the first network element is further configured to establish a media long connection with the terminal device. In this way, the media information collected by the terminal device or the media vector information obtained by the multi-modal large model can be processed by the first network element.

[0022] In a possible example, the first network element is further configured to start a media information interaction function, and the media information interaction function is configured to interact with the terminal device. In this way, after the media information interaction function is started, the media information collected by the terminal device or the media vector information obtained by the multi-modal large model can be processed by the first network element, which improves the flexibility of processing media information.

[0023] In a second aspect, the embodiments of the present application disclose a communication method applied to a first network element, for example, an artificial intelligence agent subscriber function (SAAF) network element. The method comprises the following steps: receiving first information, wherein the first information comprises user intention collected by a terminal device; obtaining media vector information of the terminal device from a second network element, wherein the media vector information is associated with behavior information of the user collected by the terminal device; and calling a multi-modal large model according to the media vector information and the first information to obtain identified user intention and / or a response related to the identified user intention.

[0024] In a possible example, the step of calling the multi-modal large model according to the media vector information and the first information to obtain the identified user intention and / or the response related to the identified user intention comprises the following steps: obtaining a first prompt word according to the first information and the media vector information; sending a first request comprising the first prompt word to the multi-modal large model, wherein the first request is used to request to identify the user intention; and receiving the identified user intention and / or the response related to the identified user intention from the multi-modal large model.

[0025] In a possible example, the method further comprises the following steps: receiving second information, wherein the second information comprises behavior information of the user collected by the terminal device; calling the multi-modal large model according to the second information to obtain the media vector information; and sending a storage request comprising the media vector information to the second network element, wherein the storage request is used to request to store the media vector information.

[0026] In a possible example, the step of calling the multi-modal large model according to the second information to obtain the media vector information comprises the following steps: obtaining a second prompt word according to the second information; sending a second request comprising the second prompt word to the multi-modal large model, wherein the second request is used to request to obtain the media vector information; and receiving the media vector information from the multi-modal large model.

[0027] In a possible example, the media vector information comprises a user portrait of the terminal device and evaluation information of the user portrait, and an evaluation dimension of the evaluation information comprises at least one of the following: importance, recency and relevance.

[0028] In a possible example, the method further comprises the following step: calling the multi-modal large model to reflect on the media vector information.

[0029] In a possible example, the resources of the first network element and the resources of the second network element are configured for the terminal device by a third network element.

[0030] In a possible example, the method further includes: sending a first registration request to the third network element, the first registration request being used to register the resource identifier of the first network element; and receiving a confirmation message of the first registration request from the third network element.

[0031] In a possible example, the method further includes: receiving a first check message from the third network element, the first check message being used to confirm the resource identifier of the first network element; and sending a confirmation message of the first resource identifier to the third network element.

[0032] In a possible example, the method further includes: receiving authentication information of the terminal device; and performing authentication on the authentication information of the terminal device, and allowing a user of the terminal device to access when the authentication is successful.

[0033] In a possible example, the method further includes: decrypting received information and / or encrypting sent information.

[0034] In a possible example, the method further includes: establishing a media long connection with the terminal device.

[0035] In a possible example, the method further includes: starting a media information interaction function, the media information interaction function being used to interact with the terminal device on media information.

[0036] It can be understood that the steps of the communication method corresponding to the second aspect are the steps performed by the first network element of the first aspect, and details are not described herein again.

[0037] In a third aspect, another communication method provided by the embodiments of the present application is applied to a second network element, for example, a subscriber vector database function (SVDF) network element. The method includes: receiving a storage request of a first network element, the storage request being used to request to store media vector information of a terminal device; and storing the media vector information of the terminal device. In this way, the media vector information of the terminal device is stored. The first network element and the second network element can avoid processing data of multiple users, and can solve the problem of random data access and leakage caused by non-isolation of user data of a part of business system, thereby protecting user privacy and meeting the requirements of regulations (such as GDPR).

[0038] In a possible example, the method further includes: receiving an information acquisition request of the first network element, the information acquisition request being used to acquire the media vector information of the terminal device; and sending the media vector information to the first network element. In this way, the media vector information of the terminal device is provided to the first network element in the case where the first network element needs to acquire user intent and / or a response of the user intent.

[0039] In a possible example, the resource of the second network element is configured by a third network element for the terminal device.

[0040] In a possible example, the method further includes: sending a second registration request to the third network element, the second registration request being used to register the resource identifier of the second network element; and receiving a confirmation message of the second registration request from the third network element.

[0041] In a possible example, the method further includes: receiving a second check message from the third network element, the second check message being used to confirm the resource identifier of the second network element; and sending a confirmation message of the second resource identifier to the third network element.

[0042] In a possible example, the method further includes: receiving authentication information of the terminal device; and verifying the authentication information of the terminal device, and allowing a user of the terminal device to access when the verification is successful.

[0043] In a possible example, the method further includes: decrypting the received information and / or encrypting the sent information.

[0044] It can be understood that the steps of the communication method corresponding to the third aspect are the steps performed by the second network element of the first aspect, and the beneficial effects of the first aspect can be referred to and will not be described here again.

[0045] In the fourth aspect, the embodiments of the present application disclose another communication method, which is applied to a third network element, for example, an artificial intelligence agent management function (AAMF) network element, and the like. The method includes: configuring a resource of a first network element and a resource of a second network element for a terminal device.

[0046] In a possible example, the method further includes: receiving a first registration request of the first network element and / or a second registration request of the second network element, the first registration request being used to register a resource identifier of the first network element, and the second registration request being used to register a resource identifier of the second network element; sending a confirmation message of the first registration request to the first network element; and / or sending a confirmation message of the second registration request to the second network element.

[0047] In a possible example, the method further includes: sending a first check message to the first network element, the first check message being used to confirm the resource identifier of the first network element; and / or sending a second check message to the second network element, the second check message being used to confirm the resource identifier of the second network element; receiving a confirmation message of the first check message of the first network element; and / or receiving a confirmation message of the second check message of the second network element.

[0048] It can be understood that the steps of the communication method corresponding to the fourth aspect are the steps performed by the third network element of the first aspect. For the beneficial effects of the first aspect, the description is not repeated here.

[0049] In the fifth aspect, the present application discloses a communication apparatus, which comprises units or modules or means for performing the steps of the method of the second aspect, the third aspect, or the fourth aspect, or any implementation thereof.

[0050] In the sixth aspect, the embodiments of the present application disclose another communication apparatus, which can be a terminal device or a network device. The communication apparatus can comprise a processor configured to cause the communication apparatus to perform the method in any of the aspects or possible examples described above by executing instructions in a memory or by a logic circuit.

[0051] In some possible examples, the communication apparatus further comprises one or more of a memory or a transceiver configured to transceive data and / or signaling.

[0052] In the seventh aspect, the present application discloses a seventh communication apparatus, comprising a processor, a memory connected to the processor, and a communication interface, the memory is configured to store one or more programs, and the processor is configured to execute the steps of any of the aspects described above.

[0053] In the eighth aspect, the embodiments of the present application disclose a communication system comprising the communication apparatus of any of the aspects described above.

[0054] In combination with the fifth aspect, the sixth aspect, the seventh aspect, or the eighth aspect, in one possible example, the communication apparatus can comprise a terminal device, a first network element, a second network element, a third network element, a control plane network element, a media plane network element, a multi-modal conversion device, a multi-modal large model, and other intelligent agents, or apparatuses therein.

[0055] In the embodiments of the present application, the multi-modal conversion device can comprise an automatic speech recognition (ASR) server and / or a natural language processing (NLP) server, etc. The ASR server is configured to convert sound into text. The NLP server is configured to understand and process text.

[0056] In the embodiments of the present application, the other intelligent agent can be understood as an intelligent agent other than the intelligent agent corresponding to the first network element, and the other intelligent agent is obtained based on the other multi-modal large model. The operator network of the multi-modal large model can be an operator network corresponding to a communication system, and the operator network of the other multi-modal large model can be an operator network corresponding to a third-party application. The other multi-modal large model can also be referred to as a vertical multi-modal large model, and the other intelligent agent can also be referred to as a vertical intelligent agent.

[0057] Optionally, the other intelligent agent communicates with the first network element. In this way, the efficiency and accuracy of intent recognition can be further improved.

[0058] In a ninth aspect, the embodiments of the present application disclose a computer readable storage medium, and the computer readable storage medium stores instructions, when the instructions are executed on a computer, the computer executes the method of any one of the above aspects.

[0059] In a tenth aspect, the embodiments of the present application disclose a computer program product for storing a computer program, when the computer program is executed on a computer, the computer executes the method of any one of the above aspects.

[0060] In an eleventh aspect, the embodiments of the present application disclose a first chip, including a processor and a memory, the processor is used to call and execute instructions stored in the memory, so that the device installed with the chip executes the method of any one of the above aspects.

[0061] In a twelfth aspect, the embodiments of the present application disclose a second chip, including: an input interface, an output interface and a processing circuit, the input interface, the output interface and the processing circuit are connected through an internal connection path, and the processing circuit is used to execute the method of any one of the above aspects.

[0062] In a thirteenth aspect, the embodiments of the present application disclose a third chip, including: an input interface, an output interface, a processor, and optionally, a memory, the input interface, the output interface, the processor and the memory are connected through an internal connection path, and the processor is used to execute the code in the memory, when the code is executed, the processor is used to execute the method in any one of the above aspects.

[0063] In a fourteenth aspect, the embodiments of the present application disclose a chip system, including at least one processor, a memory and an interface circuit, the memory, the transceiver and the at least one processor are interconnected through a line, and the at least one memory stores a computer program; the computer program is executed by the processor to execute the method in any one of the above aspects. BRIEF DESCRIPTION OF DRAWINGS

[0064] The following describes the drawings used in the embodiments of the present application.

[0065] FIG. 1 is a schematic diagram of an architecture of a communication system according to an embodiment of the present application;

[0066] FIG. 2 is a schematic diagram of an architecture of an IMS network according to an embodiment of the present application;

[0067] FIG. 3 and FIG. 4 are respectively an interaction diagram of a communication method according to an embodiment of the present application;

[0068] FIG. 5 is an interaction diagram of another communication method according to an embodiment of the present application;

[0069] FIG. 6 is an interaction diagram of another communication method according to an embodiment of the present application;

[0070] FIG. 7 is an interaction diagram of another communication method according to an embodiment of the present application;

[0071] FIG. 8 is an interaction diagram of another communication method according to an embodiment of the present application;

[0072] FIG. 9 is a schematic diagram of a structure of a communication apparatus according to an embodiment of the present application;

[0073] FIG. 10 is a schematic diagram of a structure of another communication apparatus according to an embodiment of the present application. DETAILED DESCRIPTION

[0074] The technical solutions of the embodiments of the present application can be applied to various communication systems, for example, a long term evolution (LTE) communication system, a new radio (NR) communication system, a long term evolution advanced (LTE-A) communication system, a device-to-device (D2D) communication system, a vehicle to everything (V2X) communication system, a machine to machine (M2M) communication system, an internet of things (IoT) communication system, a narrow band internet of thing (NB-IoT) communication system, a cognitive communication integrated system, a frequency division duplex (FDD) communication system, a time division duplex (TDD) communication system, a non-terrestrial network (NTN) communication system, a wireless projection communication system, an integrated access and backhaul (IAB) communication system, a public land mobile network (PLMN) communication system, a non-public network (NPN) communication system, and a communication system evolved after a 5G communication system (such as a 6G communication system), or a non-(3rd generation partnership project, 3GPP) communication system, without limitation.

[0075] In order to make the purposes, technical solutions, and advantages of the present application clearer, the technical solutions in the embodiments of the present application will be described below with reference to the drawings in the embodiments of the present application.

[0076] Please refer to FIG. 1, which is an architecture schematic diagram of a communication system provided by an embodiment of the present application. As shown in FIG. 1, the architecture of the communication system can include the following devices, network elements, and networks:

[0077] 1. A terminal device is an entity on a user side for receiving or transmitting signals, and can provide voice and / or data for a user. The terminal device can also be referred to as a terminal, a user equipment (UE), an access terminal, a UE unit, a UE station, a mobile device, a mobile station, a mobile station (MS), a mobile terminal, a mobile client, a mobile unit, a remote station, a remote terminal, a remote unit, a wireless unit, a wireless communication device, a user agent, or a user device, etc. The access terminal can be a cellular phone, a cordless phone, a session initiation protocol (SIP) phone, a wireless local loop (WLL) station, a personal digital assistant (PDA), a handheld device having wireless communication function, a computing device, or other processing device connected to a wireless modem, an in-vehicle device, a terminal in a future 5G communication system, or a terminal in a future evolved PLMN, or a terminal in a future NPN, etc.

[0078] By way of example and not limitation, in embodiments of the present application, the terminal device can also be a wearable device. The wearable device can also be referred to as a wearable smart device, which is a general term for devices that are designed and developed by applying wearable technology to daily wear, such as glasses, gloves, watches, clothes, accessories, and shoes, etc. The wearable device is a portable device that can be directly worn on the body or integrated into the clothes or accessories of the user. The wearable device is not only a hardware device, but also a device that can realize powerful functions through software support and data interaction and cloud interaction. The general wearable smart device includes devices with full functions and large sizes, which can realize complete or partial functions without relying on a smart phone, such as smart watches or smart glasses, etc., and devices that focus on a certain application function and need to cooperate with other devices such as smart phones, such as various smart wristbands and smart jewelry for monitoring vital signs, etc.

[0079] In embodiments of the present application, the apparatus for implementing the function of the terminal device can be a terminal device, or can be an apparatus capable of supporting the terminal device to implement the function, such as a chip system or a chip, which can be installed in the terminal device. The chip system can be composed of a chip, or can include a chip and other discrete devices.

[0080] The terminal device and the terminal device can communicate with each other using a certain air interface technology (such as NR or LTE technology, etc.). The terminal device and the network device can also communicate with each other using a certain air interface technology (such as NR or LTE technology, etc.).

[0081] In the embodiments of the present application, the network device can include at least one of (radio) access network ((R)AN) device, core network (CN) device, data network (DN) device, and the like, or a network element therein, or can also include a stand-alone network element.

[0082] 2. A network element, which can also be referred to as a functional network element, a functional entity, a node, a device, and the like. The network element can be a network element implemented on a dedicated hardware, a software instance running on a dedicated hardware, or an instance of a virtualized function on a suitable platform, for example, the virtualized platform can be a cloud platform. In future communication systems, the above network element can have other names, which are not limited in the present application.

[0083] In the embodiments of the present application, the network element can include a control plane (CP) functional network element and a media plane (MF) functional network element. Among them, the control plane functional network element can be referred to as a control plane network element or a CPF, and the media plane functional network element can be referred to as a media plane network element or a MF. The following takes an internet protocol (IP) multimedia subsystem (IMS) network including a control plane network element and a media plane network element as an example.

[0084] The IMS network is a subsystem composed of various core network devices providing multimedia services such as voice, video, text, chat, and the like, and delivering these services together over a packet switch (PS) domain. Simply speaking, the IMS is a subsystem for providing mobile multimedia services over a PS domain (which can be an evolved packet core (EPC) or a 5G core (5GC) + 4 / 5G access network).

[0085] The significant feature of the IMS is that it adopts a session initiation protocol (SIP) system, is independent of communication and access, can have multiple media service control functions and bearer capacity separation, call and session separation, application and service separation, service and network separation, and mobile network and Internet service integration, and the like. The proposal of the IMS conforms to the trend of the development of the communication network convergence. The IMS is not a specific product, but a network architecture, and is a general term for network elements for controlling IP multimedia services.

[0086] Please refer to FIG. 2, which is a schematic diagram of an architecture of an IMS network provided by the present application. FIG. 2 takes a terminal device as UE (such as a calling-side terminal device UE_O and a called-side terminal device UE_T), a control plane network element as a proxy-call session control function (PCSCF) network element (such as a calling-side proxy-call session control function PCSCF_O and a called-side proxy-call session control function PCSCF_T), and an application service (AS) network element (such as a calling-side application service AS_O and a called-side application service AS_T), and a media plane network element as a MF network element (such as MF_1, MF_2 and MF_3) as an example for illustration. It can be understood that other network elements capable of implementing the functions of the control plane network element and the functions of the media plane network element are also within the protection scope of the present application.

[0087] In an embodiment of the present application, a serving-call session control function (CSCF) network element is a functional entity inside an IMS network and is the core of the entire IMS network, and is mainly responsible for processing signaling control in a multimedia call session process. In a 5G communication system, the call control function network element can be a serving-call session control function (CSCF) network element. In a future communication system, the call session control function network element can still be a CSCF network element, or can also have other names, and the embodiments of the present application do not limit this.

[0088] The CSCF network element can be further divided into a serving-call session control function (SCSCF) network element, an interrogating-call session control function (ICSCF) network element and a PCSCF network element according to functions. Alternatively, the CSCF network element is referred to as a CSCF, the SCSCF network element is referred to as a SCSCF, the ICSCF network element is referred to as an ICSCF, and the ICSCF network element is referred to as an ICSCF.

[0089] SCSCF is a service processing node of the IMS network, responsible for the IMS network registration of the terminal device and related media service processing. The PCSCF is located in the visited network and is an edge network node of the IMS network. It can also be an entry node for the user to access the IMS network. The PCSCF plays a role similar to performing proxy services in the IMS network. Whether the information is from the terminal device or sent to the terminal device, it needs to be forwarded through the P-CSCF. For example, the PCSCF can be responsible for forwarding SIP signaling between the SIP user and the home network. The ICSCF is located in the home network and is a unified entry point of the home network. The ICSCF can connect the SCSCF and the PCSCF, and is used to provide an entry to the home network for the terminal device. When the terminal device roams to other networks, the ICSCF sends a message to the PCSCF. The PCSCF can forward the message from the terminal device to the ICSCF, and send the message from the terminal device to the SCSCF through the ICSCF. The ICSCF can be responsible for allocating or querying the SCSCF serving the user. In addition, the ICSCF / SCSCF is connected with the PCSCF in the home network through the Mw interface.

[0090] Compared with the media plane network element providing media services, the PCSCF, the SCSCF, and the ICSCF belong to the control plane network element. In terms of function, the PCSCF can be considered as an access type control plane network element, and the ICSCF can be considered as an interworking type control plane access network element. It can be understood that the PCSCF, the SCSCF, and the ICSCF can be independently configured in different entities, or can be integrated in the same entity. Hereinafter, for the convenience of understanding and description, the PCSCF, the SCSCF, and the ICSCF are collectively referred to as the CSCF.

[0091] 3. Control plane network element, responsible for managing the transmission of user plane data and QoS control, traffic statistics and other functions, and can perform user data packet forwarding according to the routing rules of the session management network element, such as sending uplink data to the data network or other user plane network elements, and forwarding downlink data to other user plane network elements or (R)AN network elements.

[0092] In the embodiments of the present application, the control plane network element can include the CSCF and / or the AS described above, or can include a policy control function (PCF) network element, a unified data management (UDM) network element, an application function (AF) network element, an access and mobility management function (AMF) network element, a session management function (SMF) network element, a location management function (LMF) network element, etc., mainly to complete functions such as access authentication, security encryption, and location registration of the terminal device, and to complete functions such as establishment, release, and change of a user plane transmission path, and the specific implementation can be determined according to the actual application scenario, which is not limited herein.

[0093] The media plane network element can be connected with the call control function network element through a media gateway control network element. The media gateway control network element can be understood as a gateway enabling communication between IMS users and client users. Call control signaling from the client user is all directed to the media gateway control network element, which is mainly responsible for protocol conversion between an integrated services digital network (ISDN) user part or a bearer-independent call control and an SIP protocol, and forwarding a session to the IMS. In the 5G communication system, the media gateway control network element can be a media gateway control function (MGCF) network element. In future communication systems, the media gateway control function network element can still be the MGCF, or can have other names, and the embodiments of the present application are not limited thereto. The MGCF is connected with the PCSCF / ICSCF network element / SCSCF network element in the network.

[0094] The media plane network element can also be connected with the call control function network element through an interconnection border point control network element. The interconnection border point control network element is a border function entity of the control plane between two operator networks. In the 5G communication system, the interconnection border point control network element can be an interconnection border control function (IBCF) network element. In the future communication system, the interconnection border point control network element can still be an IBCF network element, or can also have other names, which are not limited by the embodiments of the present application. The IBCF network element or the media plane (transition gateway (TrGW)) of the IBCF network element is deployed between different IMS core networks or between the IMS core network and other IP networks, and the networks are connected by IP relay. The IBCF is connected with the PCSCF / ICSCF / SCSCF in the network, the signaling plane of the IBCF is connected with other networks through an Ici interface, and the media plane of the IBCF network element is connected with other networks through an Izi interface. Each operator can select an interworking node according to the nature of the calling user. For example, when the calling user is an IMS user, the interworking through the IBCF network element / TrGW is adopted.

[0095] The multimedia system entry network element is an access point for terminal devices to access the IMS, and is used to complete user registration, quality of service control and security management, etc. For example, in the 5G communication system, the multimedia system entry network element can be a PCSCF, or have other names. The PCSCF serves as a SIP proxy function, and proxies the SIP transaction from the visited access network to the S-CSCF of the home network (according to the information recorded during registration). Specifically, the P-CSCF is responsible for IP bearer (or other granularity such as a session, a data flow, etc.) resource authentication of the IMS local access network, network address translation (NAT) and quality of service (QoS) management functions, and provides end-to-end service assurance for the IMS. In the future communication, the multimedia system entry network element can still be a PCSCF, or have other names, which are not limited by the embodiments of the present application.

[0096] 4. The media plane network element refers to a function network element capable of providing media plane services. The media plane network element can use its own media capabilities to provide media stream transmission services for terminal devices. The media plane network element exposes its own media capabilities to other network entities (such as control plane network elements) through a service-based interface (SBI), so as to indirectly enable other network entities to provide media stream transmission services for terminal devices.

[0097] For example, the media plane network element can include an access side media processing network element, a center side media processing network element, an interworking side media processing network element, and the like. The access side media processing network element can implement user media access proxy and forwarding, NAT traversal, and audio codec conversion, and the like. For example, the access side media processing network element includes an IMS access gateway (IMS-AGW). The interworking side media processing network element can implement media interworking between the IMS and other networks. For example, the interworking side media processing network element can include a TrGW, which is used to implement media interworking between the IMS and other IP networks. For another example, the interworking side media processing network element can include an IP multimedia media gateway (IM-MGW), which can implement media interworking between the IMS and other non-IP networks. The center side media processing network element can provide media resources for media plane services, such as announcement (playing audio content to a terminal device) resources, digit collection (receiving a number input by a terminal device, such as a menu number) resources, voice conference resources, video conference resources, data channel (DC) resources, extended reality management (XRM) resources, and the like. For example, the center side media processing network element can include a multimedia resource function processor (MRFP). It can be understood that the MRFP is deployed to the center side as an example. In some embodiments, the MRFP can also be deployed to the access side. In other words, the access side media processing network element can also include the MRFP. It can be understood that the media plane network element can be a physical network element or a logical network element (or a virtual network element).

[0098] According to the types of media plane services that can be provided, the media plane network element can be divided into a converged media plane network element and a non-converged media plane network element. The former refers to a media plane network element that can provide multiple media plane services, such as a media plane network element that can provide both access type services and resource type services. The latter refers to a network element that provides a single media plane service, such as a media plane service that only provides access type services.

[0099] In embodiments of the present application, the media plane services that can be provided by different MFs can be different or the same. For example, the media plane services that can be provided by MF_1 include conference services and voice recognition services; the media plane services that can be provided by MF_2 include network address translation services, announcement services, and media processing services; and the media plane services that can be provided by MF_3 include network address translation services and announcement services.

[0100] As shown in FIG. 2, the MF can provide media streaming service for the UE_O, and can also provide media streaming service for the UE_T. During the call service process, the UE_O and the UE_T and the MF, the PCSCF_O, the PCSCF_T and other network elements in the network interact, so that the necessary media transmission channel (or connection) between the MF and the UE_O and / or the UE_T is established, so that the MF provides media streaming service to the UE_O and / or the UE_T.

[0101] In the embodiment of the present application, the media endpoint of the MF interfacing with the UE and the media endpoint of the MF interfacing with other MFs are connected by default. It can also be understood that, for a MF, the media endpoint interfacing with the UE receives information by default and forwards the information to the media endpoint of the MF interfacing with other MFs. For example, the media endpoint interfacing with the UE receives information and broadcasts the information. Similarly, for a MF, the media endpoint interfacing with other MFs receives information by default and forwards the information to the media endpoint of the MF interfacing with the UE. For example, the media endpoint interfacing with other MFs receives information and broadcasts the information.

[0102] The application service network element can be used to obtain the information of the media plane network element accessed initially, or can be used to request the media plane network element to provide media plane service. In the 5G communication system, the application service network element can be an AS network element. In the future communication system, the application service network element can still be AS, or can also have other names, which are not limited in the embodiment of the present application. The AS is an upper layer application layer device in the IMS system. The AS interacts with the CSCF through the standard SIP protocol, and then realizes the triggering and execution of various network services, and provides various service logic control functions.

[0103] It can be understood that FIG. 2 is only an example. The IMS network to which the embodiment of the present application is applicable is not limited to only containing the network elements shown in FIG. 2, but can also contain other devices not represented in FIG. 2, such as SCSCF network elements, etc., which are not listed one by one here. And the distribution form of each network element in FIG. 2 is only exemplary, and the distribution form of each network element is not limited in the embodiment of the present application. The interface between each network element in FIG. 2 can refer to the existing interface protocol definition, which is not described here.

[0104] In addition, FIG. 2 is only an example, and the present application can be applied to a communication system corresponding to an IMS network, and can also be applied to other communication systems (including control plane network elements and media plane network elements, or other network elements capable of realizing the functions of control plane network elements and media plane network elements), such as the 6G real time communication (RTC) network which is still in the definition stage. The 6G RTC network is an operator's network, which can be used for real-time transmission of media information or media vector information.

[0105] In the embodiments of the present application, the types of media information or media vector information include text, pictures, videos, voice, location, files, etc., which are not limited herein.

[0106] 5、The first network element is a network element that implements the logical function of the user artificial intelligence agent (AI Agent). In the embodiments of the present application, the first network element can be a subscription function SAAF network element for the artificial intelligence agent.

[0107] There is no unified definition of artificial intelligence agent in academia and industry, and the mainstream definition is an artificial intelligence agent that can autonomously complete a task (driven by a task / goal / intent). The AI Agent is driven by a task, recognizes user needs through a natural language instead of a graphical interface, generates a reasonable plan and interface, helps users complete tasks / achieve goals, that is, interacts with people using natural language. And the AI Agent autonomously decides actions and reviews results through a multi-modal large model LMM to complete the task. The user only needs to demand, and most of the time is completed by the AI itself, which can realize intelligent reasoning, decompose sub-tasks according to the task, reason out the solution, achieve the task effect while reflecting and iterating, self-learning, and has reflection ability, and can work in an unpredictable environment with a lot of new information. In addition, the AI Agent can use tools (such as calendar tools, calculators, search functions, etc.) by calling external application programming interfaces (APIs). For example, they can browse web pages, read and write files, pay, control computers, etc. The AI Agent can also interact with the real environment, code environment, etc. by injecting environmental information through tools or external functions calling APIs.

[0108] The essence of the AI Agent is to use the LMM to command the brain, and use tools as hands and feet to execute tasks by calling external APIs. Among them, the reasoning ability of the LMM restricts the upper limit of the intelligence of the Agent. The calling ability of the Tools restricts the upper limit of the action of the Agent, and how much potential can be exerted depends on the prompt engineering. The LMM needs to provide sufficient context as a prompt word input to get an ideal result.

[0109] An artificial intelligence agent can be simply referred to as an agent. There are many definitions of an agent, and in the embodiments of the present application, an agent is an entity that perceives the state of the environment according to the instructions given by the user or the target set, retrieves built-in knowledge and perception experience memory, defines, decomposes and plans the task, forms and executes the action strategy, and feeds back to the target environment. The agent has the following characteristics:

[0110] (1)Autonomy: the ability of an agent to adjust its behavior and state automatically according to changes in the external environment, rather than merely passively accepting external stimuli, and to have the ability of self-management and self-regulation.

[0111] (2) Reactivity: the ability to react to external stimuli.

[0112] (3) Proactivity: the ability of an agent to take initiative to change the external environment.

[0113] (4) Sociality: the ability of an agent to cooperate with other agents or people, and different agents can interact with each other according to their own intentions to achieve the purpose of solving problems.

[0114] (5) Evolution: the ability of an agent to accumulate or learn experience and knowledge, and to modify its behavior to adapt to a new environment.

[0115] In the embodiments of the present application, the first network element can be understood as an intelligent agent of a user of a terminal device, to complete a Plan-Action-Observation (ReAct) process of planning and executing the user's user intention, and to realize one personalized assistant logic for one user. ReAct is essentially a paradigm that combines reasoning and acting, and is a common Agent iteration method. It can be understood that the first network element can collect the user's behavior information through the terminal device, and then implement the RAG mechanism to continuously reflect and refine the user's behavior information, thereby continuously deepening the accurate understanding of the user, and can obtain the "most important and most relevant memories" for the current session, and improve the user's user portrait. The user portrait can be understood as the user's personal data or profile, etc.

[0116] In the embodiments of the present application, the information (for example, the user's behavior information, etc.) collected by the terminal device can be understood as short-term memory (cached data), which provides a session-level context. The information obtained through the ReACT process can be understood as long-term memory, which provides a user-level context.

[0117] The user's behavior information can be operation information collected by the terminal device when the user operates the terminal device, or can be an action log obtained by the terminal device summarizing the user's behavior. Taking a user Xiaoming as an example, the user's behavior information includes:

[0118] I1: Xiaoming met a colleague and chatted about today's research goal.

[0119] I2: Xiaoming read a paper this morning, "6G use case and key tech" photo conversion of the abstract.

[0120] I3: Xiaoming's schedule for the day, 3 network architecture seminars.

[0121] Among them, I1, I2 and I3 are respectively the identification of different behavior information.

[0122] In the case where the third-party application allows the terminal device to collect information, the terminal device can also collect the user's third-party information, optionally.

[0123] Among them, the third-party application includes an application (application, APP) or a small program in the application, etc., which refers to the client software on the terminal device. The third-party information refers to the user's information on the terminal device or the information related to the information. For example:

[0124] I4: Xiaoming's shopping information.

[0125] I5: Xiaoming's express information.

[0126] I6: Xiaoming's navigation information.

[0127] Among them, I1, I2 and I3 are respectively the identification of different third-party information. I4 can be information on a shopping APP, and I6 can be information on a navigation APP. I5 can be information on a shopping APP, or it can be information on a delivery mini-program.

[0128] It can be understood that the third-party information can reflect the user's behavior information. Optionally, the user's behavior information is associated with the user's third-party information. In this way, the user's behavior information can be obtained according to the third-party information, and the user's intention can be obtained according to the user's behavior.

[0129] 6、Second network element, for storing media vector information of terminal device.

[0130] Among them, the media vector information is a kind of vector data, and the type of the vector data can include text, picture, video, voice, location, file, etc., which is not limited here. The second network element can be regarded as a kind of vector database (vector database, VDB). In the embodiment of the present application, the second network element can be a database subscription function SVDF network element. The media vector information of the terminal device can include the user's behavior information collected by the terminal device, and can also include the user's user portrait and / or evaluation information of the user portrait obtained by reflecting the user's behavior information.

[0131] The user profile can include a plurality of long term memories (LTM), such as LTM1, LTM2, and the like. The long term memories are different from short term memories, which are used to reflect the context at the session level, while the long term memories are used to reflect the context at the user level, thereby improving the accuracy of the user profile. Each long term memory can also include relevant evaluation information. In some possible examples, the evaluation information can include at least one of the following: importance, recency, and relevance.

[0132] For example, the media vector information of the terminal device is as follows:

[0133] Master profile1: LTM1, importance level 1

[0134] Master profile1: LTM2, importance level 2

[0135] Master profile1: LTM3, importance level 3

[0136] ...

[0137] Master profile1: LTMx, importance level x

[0138] Wherein, Master profile1 represents the user profile of the terminal device. LTM1, LTM2, LTM3, …, LTMx represent x different long term memories, and importance level 1, 2, 3, …, and x represent x different importance levels.

[0139] 7. A third network element configured to configure the resources of the first network element and the resources of the second network element for the terminal device. In the embodiments of the present application, the third network element can be an artificial intelligence agent management function (AAMF) network element. It can be understood that the third network element is configured to configure the resources of the first network element and the resources of the second network element for the terminal device. In this way, the first network element and the second network element can be prevented from processing data of multiple users, and random data access and leakage caused by non-isolation of user data of a part of business system can be solved, thereby protecting user privacy and meeting the requirements of regulations (such as the general data protection regulation (GDPR)).

[0140] The process of the third network element configuring the first network element and the second network element of the terminal device can refer to the process of the virtual machine management system creating a virtual machine and / or the container management system creating a container, for example, first determining the configuration file of the first network element and the second network element, the configuration file containing resource information (such as the memory size occupied by the hard disk file), and then starting the instance of the first network element and the second network element, so that the first network element and the second network element can implement corresponding functions, and the like, which will not be described here.

[0141] For example, taking the first network element as SAAF, the second network element as SVDF, and the resource identifier as a uniform resource identifier (URI) as an example, the URI of the first network element can be http: / / saaf-baseuri.example.com / A / instanceid-uuid / , and the URI of the second network element can be http: / / svdf-baseuri.example.com / A / instanceid-uuid / . Wherein, instanceid-uuid should be filled according to the resource identifier of the instance generated by the network element. In this way, the first network element can be interacted with through the URI of the first network element, and the second network element can be interacted with through the URI of the second network element.

[0142] The application does not limit the timing of the third network element configuring the resources of the first network element and the resources of the second network element of the terminal device, which can be after the terminal device wakes up the voice assistant, or after the terminal device opens the media information interaction function, or after the terminal device starts receiving the registration message or the service request, and the like, which will not be limited here.

[0143] In the embodiment of the application, the third network element configuring the resources of the first network element and the resources of the second network element of the terminal device can be understood as the third network element creating the resources of the first network element and the resources of the second network element. Alternatively, the third network element can also be used to start the resources of the first network element and the resources of the second network element of the terminal device. The third network element can also be used to close the resources of the first network element and the resources of the second network element of the terminal device. The third network element can also be used to delete the resources of the first network element and the resources of the second network element of the terminal device. In this way, the life cycle of the first network element and the second network element of the terminal device can be managed by creating, starting, closing, and deleting.

[0144] Alternatively, the first network element determines the resource identifier of the first network element according to the resources of the first network element, and the second network element determines the resource identifier of the second network element according to the resources of the second network element.

[0145] The resource identifier can be a URI. The resource identifier of the first network element is different from the resource identifier of the second network element. The resource identifier of the first network element is used to determine the resource of the first network element, and the resource identifier of the second network element is used to determine the resource of the second network element. Optionally, the resource identifier can be referred to as a service resource identifier, such as a service URI.

[0146] Optionally, the registration function of the resource identifier is referred to as a service registration function.

[0147] In some possible examples, the third network element is further configured to receive the first registration request of the first network element and / or the second registration request of the second network element, and send a confirmation message of the first registration request to the first network element and / or a confirmation message of the second registration request to the second network element.

[0148] Correspondingly, the first network element is further configured to send the first registration request to the third network element, and the second network element is further configured to send the second registration request to the third network element. The first network element is further configured to receive a confirmation message of the first registration request from the third network element, and the second network element is further configured to receive a confirmation message of the second registration request from the third network element.

[0149] The first registration request includes the resource identifier of the first network element, and the second registration request includes the resource identifier of the second network element. The third network element can store a correspondence between the user identifier of the terminal device and the resource identifier of the first network element, and a correspondence between the user identifier of the terminal device and the resource identifier of the second network element. Optionally, the confirmation message of the first registration request includes the resource identifier of the first network element, and the confirmation message of the second registration request includes the resource identifier of the second network element.

[0150] It can be understood that, in the case that the third network element has the registration function of the resource identifier, the resource of the first network element can be registered according to the resource identifier of the first network element, and the resource of the second network element can be registered according to the resource identifier of the second network element, and after the registration is successful, the corresponding confirmation message is sent.

[0151] In other possible examples, the third network element is further configured to send a first check message to the first network element and / or a second check message to the second network element, and receive a confirmation message of the first check message of the first network element and / or a confirmation message of the second check message of the second network element.

[0152] Correspondingly, the first network element is further configured to receive the first check message from the third network element, and send a confirmation message of the first check message to the third network element. And / or the second network element is further configured to receive the second check message from the third network element, and send a confirmation message of the second check message to the third network element.

[0153] The first check message is used to confirm the resource identifier of the first network element, and the second check message is used to confirm the resource identifier of the second network element. That is, the third network element can determine the resource of the first network element based on the resource identifier of the first network element, and determine the resource of the second network element based on the resource identifier of the second network element.

[0154] It can be understood that, in the case that the third network element does not have the registration function of the resource identifier, it can be checked whether the resource of the first network element corresponds to the first network element according to the resource identifier of the first network element, and whether the resource of the second network element corresponds to the second network element according to the resource identifier of the second network element, and a success response, that is, a confirmation message, is returned by the corresponding network element after the check is successful. Optionally, the confirmation message of the first check message includes the resource identifier of the first network element, and the confirmation message of the second check message includes the resource identifier of the second network element.

[0155] 8、Multi-modal large model, refers to an artificial intelligence model that can process and understand multiple different types of input data (such as text, images, audio, and other media information). These models can perform a series of complex tasks by learning and integrating features of different modalities (i.e. data types), such as: image annotation: identifying objects in an image and generating descriptive text.

[0156] In the embodiments of the present application, the multi-modal large model LMM can include a language large model LLM, or can include a large video model LVM, etc., which is not limited here. The multi-modal large model is used to generate a user portrait of a user and / or evaluation information of the user portrait based on behavior information of the user.

[0157] 9、Multi-modal conversion device, used to convert the type of input media information into another type of media information, and can output another type of media information. In the embodiments of the present application, the multi-modal conversion device can include an automatic speech recognition ASR server and / or a natural language processing NLP server, etc. The ASR server is used to convert sound into text. The NLP server is used to understand and process text.

[0158] 10、Other intelligent agent, which can be understood as an intelligent agent other than the intelligent agent corresponding to the first network element. The other intelligent agent is obtained based on the other multi-modal large model, and the other multi-modal large model is also a multi-modal large model, but is deployed in a different operator network from the aforementioned multi-modal large model. In the embodiments of the present application, the operator network of the multi-modal large model can be the operator network corresponding to the communication system, and the operator network of the other multi-modal large model can be the operator network corresponding to the third-party application program. The other multi-modal large model can also be referred to as a vertical multi-modal large model, and the other intelligent agent can also be referred to as a vertical intelligent agent.

[0159] The other intelligent agent can be obtained by the other multi-modal large model based on third-party information of the user in a third-party application. The application does not limit the data type of the other intelligent agent determined by the other multi-modal large model for the user, for example, the other multi-modal large model can obtain third-party information (such as travel data) of the user in a travel application, and then obtain the other intelligent agent of the user in travel. For another example, the other multi-modal large model can obtain third-party information (such as shopping data) of the user in a shopping application, and then obtain the other intelligent agent of the user in shopping, and the like.

[0160] Optionally, the other intelligent agent is used for communication with the first network element. It can be understood that when the other intelligent agent communicates with the intelligent agent of the first network element, the information of the user in the third-party application can be obtained from the other intelligent agent, and / or the other intelligent agent is called to identify the user's intention, which is beneficial to further improve the efficiency and accuracy of intention identification.

[0161] In some embodiments, the terminal device, network device, and various network elements described above can be referred to as communication apparatuses, which can be general-purpose devices or special-purpose devices, and the embodiments of the present application do not make specific limitations thereto.

[0162] The network architecture and business scenarios described in the embodiments of the present application are used to more clearly illustrate the technical solutions of the embodiments of the present application, and do not constitute a limitation on the technical solutions provided by the embodiments of the present application. It can be known by those skilled in the art that, with the evolution of network architecture and the appearance of new business scenarios, the technical solutions provided by the embodiments of the present application are also applicable to similar technical problems.

[0163] The application provides a communication method, which calls a multi-modal large model through a network element in a network side to obtain a user's intention and / or a response related to the user's intention, so as to improve the support for media information interaction function. The user's intention is identified according to the media vector information related to the user's behavior information, so as to improve the accuracy of intention identification.

[0164] In order to better understand the scheme described in the present application, the following introduces the application scenarios to which the present application can be applied.

[0165] In a first application scenario, the terminal device is in a communication state, and the terminal device establishes a media long connection with the first network element.

[0166] The communication state refers to a state in which the terminal device can communicate with other terminal devices or network devices. The communication state is usually a state after the terminal device is successfully registered in the network device after being turned on. The terminal device can establish a media long connection with the first network element after being turned on, or can establish a media long connection with the first network element after starting a media information interaction function. The media information interaction function can be started or stopped in the setting interface of the terminal device, or can be awakened by a preset awakening condition. The awakening condition can include an awakening instruction, a timing time, or a long time of not responding to media information of other users, which is not limited herein. When the awakening condition is met, the media information interaction function is started to enable the terminal device to establish a media long connection with the first network element. Otherwise, the media information interaction function is not started to prevent the terminal device from transmitting data to the network side in real time, which can save energy.

[0167] Optionally, before the terminal device establishes the media long connection with the first network element, the terminal device can further send a media long connection request to the first network element, and the media long connection request is used to establish the media long connection.

[0168] Correspondingly, the first network element receives the media long connection request from the terminal device. Specifically, the terminal device sends a first media long connection request to the control plane network element, and the control plane network element sends a second media long connection request to the first network element. The first media long connection request is used for the terminal device to request to establish a media long connection with the control plane network element. The second media long connection request is used to request the control plane network element to request to establish a media long connection with the first network element.

[0169] It should be noted that the terminal device establishes a media long connection with the first network element through the media long connection request, which is different from the traditional session request, which is released after the session ends, while the media long connection established by the terminal device and the first network element in the present application will be continuously maintained after the connection is successful.

[0170] Exemplarily, please refer to FIG. 3 or FIG. 4, which are respectively an interaction schematic diagram of a communication method provided by an embodiment of the present application. In FIG. 3 and FIG. 4, the terminal device establishes a media long connection with the first network element after being turned on. In FIG. 3, the third network element has a registration function of resource identification. In FIG. 4, the third network element does not have a registration function of resource identification. The communication devices involved in the method can include a terminal device, a first network element, a second network element, a third network element, a control plane network element, a media plane network element, and a multi-modal large model, etc. These communication devices can refer to the description of FIG. 1 and FIG. 2, which will not be repeated herein. The method includes but is not limited to the following steps S301-S323, wherein:

[0171] S301, the terminal device is turned on.

[0172] S302, the terminal device sends a registration message to the control plane network element.

[0173] Accordingly, the control plane network element receives a registration message from the terminal device.

[0174] The registration message can be, for example, a Register message. The registration message can be a session message, such as a SIP message. The registration request can also be a non-access stratum (NAS) or other type of network registration message, which is not limited herein.

[0175] S303. The control plane network element sends a third-party registration message to a third network element.

[0176] Accordingly, the third network element receives a third-party registration message from the control plane network element, for example, a 3 rd Party Register message.

[0177] Optionally, the third-party registration message includes subscription information of a user of the terminal device.

[0178] S304. The third network element configures resources of the first network element and resources of the second network element for the terminal device.

[0179] The step of the third network element configuring resources of the first network element and resources of the second network element for the terminal device can be understood as a resource application, for example, a resource apply. In this way, the resources of the first network element and the resources of the second network element correspond to the terminal device, which can avoid the first network element and the second network element processing data of multiple users, solve the problem of random data access and leakage caused by non-isolation of user data of some business system, and further protect user privacy and meet regulatory requirements (such as GDPR).

[0180] The first network element can invoke a multi-modal large model to implement corresponding processing logic, so that the first network element of the terminal device can be understood as a sandbox instance (such as a SAAF sandbox instance) of a user of the terminal device, and the second network element of the terminal device can be understood as another sandbox instance (such as a SVDF sandbox instance) of the user of the terminal device.

[0181] In some feasible examples, the first network element receives verification information of the terminal device, verifies the verification information, and allows the user of the terminal device to access information in the first network element when the verification is successful. The second network element receives verification information of the terminal device, verifies the verification information, and allows the user of the terminal device to access information in the second network element when the verification is successful.

[0182] The verification information of the terminal device can be a fixed key or a dynamically changing key. For example, the verification information is a verification code. The verification information can be a password text or a password pattern, or can be biological verification data, such as a fingerprint image, a face image, a voiceprint voice, etc. The type of the verification information is not limited in the present application. The verification information received by the first network element can be the same as or different from the verification information received by the second network element.

[0183] Optionally, the verification of the verification information can include matching the verification information with preset verification information, and determining that the verification of the verification information is successful when the matching is successful.

[0184] The preset verification information can be information set by a user or a network element. For example, in the process of configuring resources of the SAAF and resources of the SVDF for the terminal device by the AAMF, a verification setting request is sent to the terminal device to enable the user to set the preset verification information of the SAAF and the SVDF, such as a fingerprint image. Then the AAMF sends the preset verification information to the SAAF and the SVDF. When the terminal device sends an access request to the SAAF, the SAAF requires the terminal device to obtain and transmit the verification information. The SAAF matches the verification information received from the terminal device with the preset verification information. When the matching is successful, the user of the terminal device is allowed to access data in the SAAF. When the matching fails, the user of the terminal device is not allowed to access data in the SAAF.

[0185] It can be understood that the processing logic and data of the user are isolated by the first network element and the second network element, and the identity of the user is verified through the verification of the verification information. When the verification is successful, the user is allowed to access the data, which can solve the problem of random data access and leakage caused by non-isolation of user data in some business systems, thereby protecting the privacy of the user and meeting the requirements of regulations (such as GDPR).

[0186] In some possible examples, the first network element decrypts the received information and / or encrypts the transmitted information.

[0187] In the embodiments of the present application, the first network element can receive the information of the terminal device from the media plane network element, or can receive the information of the terminal device processed by the multi-modal conversion device from the media plane network element, or can receive the media vector information of the terminal device from the second network element, or can receive the media vector information obtained from the behavior information or the media vector information of the user from the multi-modal large model, etc. The first network element can send the user intent and / or the response of the user intent to the terminal device through the media plane network element, or can send the media vector information to be stored to the second network element, or can send the information of the to-be-identified intent and / or the prompt word of the combination of the media vector information to the multi-modal large model, etc. The present application does not limit the type of information received by the first network element and the type of information sent.

[0188] Optionally, the key for encryption by the first network element can be paired with the key for decryption by the first network element, that is, the decryption key is the encryption key. The encryption or decryption key can be a fixed key or a dynamically changing key, which is not limited here. The user can be unaware of the key, which is encrypted or decrypted by the network side.

[0189] It can be understood that the first network element can encrypt before data transmission, so that the network element receiving the encrypted data can decrypt it. Similarly, other network elements can encrypt before data transmission, so that the network element receiving the encrypted data can decrypt it. In this way, the network element receiving the encrypted information can decrypt the received information. By encrypting and / or decrypting the transmission data, the security of the data can be improved.

[0190] The third network element configures the resources of the first network element and the resources of the second network element for the terminal device, which can be referred to as the foregoing, and will not be described here. Optionally, the method further comprises: the third network element determines the timing time of the media vector information of the terminal device.

[0191] The timing time can be understood as the start time of the timer or the timer. The timer can be understood as the start timer of the step of reflecting the media vector information of the terminal device, and the timer can be understood as the start timer of the step of ending the execution of the reflection of the media vector information of the terminal device. In this way, when the timing time arrives, the first network element is triggered to reflect the media vector information of the terminal device to perfect the user portrait of the user, which is beneficial to improve the accuracy of intent recognition.

[0192] S305, the second network element is initialized.

[0193] The second network element performs initialization, for example, SVDF sandbox instance initialization. The initialization process of the second network element can include determining the resource identifier of the second network element.

[0194] S306. The second network element sends a second registration request to the third network element, and the second registration request is used to register the resource identifier of the second network element.

[0195] Correspondingly, the third network element receives the second registration request from the second network element. For example, the SVDF sends a registration message (such as SVDF register to AAMF) to the AAMF. The second registration request can include the identification information of the terminal device, for example, a subscribe id. In this way, after receiving the second registration request, the third network element can register the resource or (service) function of the second network element according to the resource identifier of the second network element, and save the identification information of the terminal device and the resource identifier of the second network element. The third network element can also provide the resource identifier of the second network element of the terminal device to the first network element of the terminal device and the device (such as a control plane network element) on the network side.

[0196] S307. The third network element sends a second registration request confirmation message to the second network element.

[0197] Correspondingly, the second network element receives the second registration request confirmation message from the third network element. For example, the second registration request confirmation message can include a registration response such as (rsp to register).

[0198] S308. The first network element performs initialization.

[0199] The first network element performs initialization, for example, SAAF sandbox instance initialization. The initialization process of the first network element includes determining the resource identifier of the first network element, which can refer to the description of the initialization of the second network element, and will not be repeated here.

[0200] Optionally, the first network element can also include starting a timer or a timer during initialization. The timer or timer can be periodic, for example, the timer is started three times at 01:00, 13:00, and 18:00 respectively in a day. In this way, when the timing time arrives, the first network element triggers the reflection of the media vector information of the terminal device to perfect the user portrait of the user.

[0201] S309. The first network element sends a first registration request to the third network element, and the first registration request is used to register the resource identifier of the first network element.

[0202] Accordingly, the third network element receives the first registration request from the first network element. For example, the SAAF sends a registration message (e.g., SAAF register to AAMF) to the AAMF. The first registration request can include the identification information of the terminal device. In this way, after receiving the first registration request, the third network element can register the resource or (service) function of the first network element according to the resource identifier of the first network element, and save the identification information of the terminal device and the resource identifier of the first network element. The third network element can also provide the resource identifier of the first network element of the terminal device to the device (e.g., a control plane network element) on the network side.

[0203] S310, the third network element sends a first registration request confirmation message to the first network element.

[0204] Accordingly, the first network element receives the first registration request confirmation message from the third network element. For example, the first registration request confirmation message can include a registration response (e.g., rsp to register).

[0205] Optionally, the first registration request confirmation message includes the resource identifier of the first network element, and the second registration request confirmation message includes the resource identifier of the second network element.

[0206] It can be understood that in the case where the third network element has a registration function of the resource identifier, the resource of the first network element can be registered according to the resource identifier of the first network element, and the resource of the second network element can be registered according to the resource identifier of the second network element, and after the registration is successful, the corresponding confirmation message is sent.

[0207] S311, the first network element sends a first resource acquisition request to the third network element, and the first resource acquisition request is used to acquire the resource identifier of the second network element.

[0208] Accordingly, the third network element receives the first resource acquisition request of the first network element.

[0209] Optionally, the first resource acquisition request includes the identification information of the terminal device. In this way, the resource identifier of the second network element of the terminal device can be sent to the first network element of the terminal device based on the identification information of the terminal device.

[0210] S312, the third network element sends a response of the first resource acquisition request to the first network element, and the response of the first resource acquisition request includes the resource identifier of the second network element.

[0211] Accordingly, the first network receives the response of the first resource acquisition request from the third network element.

[0212] S313, the first network element stores the resource identifier of the second network element.

[0213] S314, the third network element sends a third-party registration message confirmation message to the control plane network element, the third-party registration message confirmation message including the resource identifier of the first network element and the resource identifier of the second network element.

[0214] Correspondingly, the control plane network element receives the third-party registration message confirmation message from the third network element.

[0215] S315, the control plane network element stores the resource identifier of the first network element and the resource identifier of the second network element.

[0216] S316, the control plane network element sends a registration message confirmation message to the terminal device.

[0217] Correspondingly, the terminal device receives the registration message confirmation message from the control plane network element.

[0218] The third-party registration message confirmation message and the registration message confirmation message can be 200 messages.

[0219] S317, the terminal device sends a first session request to the control plane network element, the first session request being used to establish a media long connection with the control plane network element.

[0220] Correspondingly, the control plane network element receives the first session request from the terminal device.

[0221] The first session request includes a SIP request (SIP INVITE) or other session control messages, etc., which are not limited herein. The first session request can include a user identifier of the terminal device. The service information in the first session request includes at least one of a data network name, a network slice identifier, a service identifier, a service type identifier, and a service provider identifier. It should be understood that the first session request can be understood as a media long connection request (such as the first media long connection request described above) of the terminal device, which is different from a traditional session request that is released after a session ends, and the first session request is continuously maintained after the media long connection established by the terminal device and the first network element is successfully connected, i.e., the media long connection is maintained after step S323.

[0222] S318, the media plane network element sends a second resource acquisition request to the control plane network element, the second resource acquisition request being used to acquire the resource identifier of the first network element and the resource identifier of the second network element.

[0223] Correspondingly, the control plane network element receives the second resource acquisition request from the media plane network element.

[0224] S319, the control plane network element sends a response to the second resource acquisition request to the media plane network element, the response to the second resource acquisition request including the resource identifier of the first network element and the resource identifier of the second network element.

[0225] Correspondingly, the media plane network element receives the resource identifier of the first network element and the resource identifier of the second network element from the control plane network element.

[0226] S320, the media plane network element stores the resource identifier of the first network element and the resource identifier of the second network element.

[0227] S321, the control plane network element sends a second session request to the first network element, and the second session request is used to establish a media long connection with the first network element.

[0228] Correspondingly, the first network element receives the second session request from the control plane network element. It can be understood as the aforementioned second media long connection request by referring to the description of the first session request.

[0229] S322, the first network element sends an acknowledgement message of the second session request to the control plane network element.

[0230] Correspondingly, the control plane network element receives the acknowledgement message of the second session request from the first network element.

[0231] S323, the control plane network element sends an acknowledgement message of the first session request to the terminal device.

[0232] Correspondingly, the terminal device receives the acknowledgement message of the first session request from the control plane network element.

[0233] The acknowledgement message of the first session request and the acknowledgement message of the second session request can be a 200 message, which is used to indicate that the media long connection has been established. In this way, the terminal device can communicate with the first network element through the control plane network element and realize media information interaction.

[0234] In the method shown in FIG. 3, after the terminal device is powered on, the terminal device is configured with the resource of the first network element and the resource of the second network element through the third network element, so that the third network element registers the resource identifier of the first network element and the resource identifier of the second network element, and the terminal device is registered in the network. After successful registration, the terminal device establishes a media long connection with the first network element, so that the terminal device can process the media information collected by the terminal device or the media vector information obtained by the multi-modal large model through the first network element.

[0235] Please refer to FIG. 4, the communication method provided by FIG. 4 involves communication devices, which can include a terminal device, a first network element, a second network element, a third network element, a control plane network element, a media plane network element, and a multi-modal large model. These communication devices can be referred to the description of FIG. 1 and FIG. 2, which will not be described here. The method includes but is not limited to the following steps S401 to S423, wherein:

[0236] S401, the terminal device is powered on.

[0237] S402, the terminal device sends a registration message to the control plane network element.

[0238] Accordingly, the control plane network element receives the registration message from the terminal device.

[0239] S403, the control plane network element sends a third-party registration message to the third network element.

[0240] Accordingly, the third network element receives the third-party registration message from the control plane network element.

[0241] S404, the third network element configures the resources of the first network element and the resources of the second network element for the terminal device.

[0242] S405, the second network element is initialized.

[0243] Among them, steps S401 to S405 can refer to the description of steps S301 to S305, which will not be repeated here.

[0244] S406, the third network element sends a second check message to the second network element, and the second check message is used to confirm the resource identifier of the second network element.

[0245] Accordingly, the second network element receives the second check message from the third network element.

[0246] S407, the second network element sends a confirmation message of the second check message to the third network element.

[0247] Accordingly, the third network element receives the confirmation message of the second check message from the second network element.

[0248] S408, the first network element is initialized.

[0249] S409, the third network element sends a first check message to the first network element, and the first check message is used to confirm the resource identifier of the first network element.

[0250] Accordingly, the first network element receives the first check message from the third network element.

[0251] S410, the first network element sends a confirmation message of the first check message to the third network element.

[0252] Accordingly, the third network element receives the confirmation message of the first check message from the first network element.

[0253] S411, the first network element sends a first resource acquisition request to the third network element, and the first resource acquisition request is used to acquire the resource identifier of the second network element.

[0254] Accordingly, the third network element receives the first resource acquisition request of the first network element.

[0255] S412, the third network element sends a response of the first resource acquisition request to the first network element, and the response of the first resource acquisition request includes the resource identifier of the second network element.

[0256] S413, the first network element stores the resource identifier of the second network element.

[0257] S414, the third network element sends a confirmation message of the third-party registration message to the control plane network element, and the confirmation message of the third-party registration message includes the resource identifier of the first network element and the resource identifier of the second network element.

[0258] Correspondingly, the control plane network element receives the confirmation message of the third-party registration message from the third network element.

[0259] S415, the control plane network element stores the resource identifier of the first network element and the resource identifier of the second network element.

[0260] S416, the control plane network element sends a confirmation message of the registration message to the terminal device.

[0261] Correspondingly, the terminal device receives the confirmation message of the registration message from the control plane network element.

[0262] S417, the terminal device sends a first session request to the control plane network element, and the first session request is used to establish a media long connection with the control plane network element.

[0263] Correspondingly, the control plane network element receives the first session request from the terminal device.

[0264] S418, the media plane network element sends a second resource acquisition request to the control plane network element, and the second resource acquisition request is used to acquire the resource identifier of the first network element and the resource identifier of the second network element.

[0265] Correspondingly, the control plane network element receives the second resource acquisition request from the media plane network element.

[0266] S419, the control plane network element sends a response of the second resource acquisition request to the media plane network element, and the response of the second resource acquisition request includes the resource identifier of the first network element and the resource identifier of the second network element.

[0267] Correspondingly, the media plane network element receives the resource identifier of the first network element and the resource identifier of the second network element from the control plane network element.

[0268] S420, the media plane network element stores the resource identifier of the first network element and the resource identifier of the second network element.

[0269] S421, the control plane network element sends a second session request to the first network element, and the second session request is used to establish a media long connection with the first network element.

[0270] Correspondingly, the first network element receives the initial session request from the control plane network element.

[0271] S422, the first network element sends a second session request confirmation message to the control plane network element.

[0272] Correspondingly, the control plane network element receives the second session request confirmation message from the first network element.

[0273] S423, the control plane network element sends a first session request confirmation message to the terminal device.

[0274] Correspondingly, the terminal device receives the first session request confirmation message from the control plane network element.

[0275] Wherein, steps S411 to S423 can refer to the description of steps S311 to S323, which will not be repeated here.

[0276] In the method shown in FIG. 4, after the terminal device is powered on, the terminal device is configured with the resources of the first network element and the resources of the second network element by the third network element, so that the third network element checks the resource identifier of the first network element and the resource identifier of the second network element, and the terminal device is registered in the network. After successful registration, the terminal device establishes a media long connection with the first network element, so that the terminal device can process the media information collected by the terminal device or the media vector information obtained by the multi-modal large model through the first network element.

[0277] It should be noted that the two methods of establishing a media long connection are established after the terminal device is powered on. In fact, other cases can also be included, such as after the media information interaction function is started.

[0278] It can be understood that in the case that the terminal device is in a communication state and the terminal device establishes a media long connection with the first network element, the communication method provided by the present application can be realized. In the case that the terminal device is in a communication state and the terminal device does not establish a media long connection with the first network element, the communication method provided by the present application is not executed.

[0279] The second application scenario is that the terminal device is in a non-communication state, and the media information interaction function is started when the first network element receives a trigger request from another terminal device.

[0280] Wherein, the non-communication state refers to a state in which the terminal device cannot communicate with other terminal devices or network devices. The non-communication state usually refers to a state in which the terminal device is in a power-off state, or the terminal device has closed the network connection function, for example, the cellular network is closed. The non-communication state can also include the case of communication conflict, for example, the terminal device is in a call, and does not support third-party call, etc.

[0281] The trigger request is used to instruct the first network element to start the media information interaction function, so as to help realize the media information interaction between the other terminal device and the terminal device. The first network element starting the media information interaction function can be understood as the first network element waking up a multi-modal assistant (personal assistant or intelligent agent AI agent), for example, a voice assistant.

[0282] It can be understood that when the terminal device is in a non-communication state, it cannot communicate with other terminal devices or network devices. Then there can be an interaction request of the other terminal device sending media information to the terminal device, so when the terminal device is in a non-communication state, if the first network element receives the trigger request, the media information interaction function is started to realize the communication method provided in the present application. In the case that the terminal device is in a non-communication state and the first network element does not receive the trigger request, the communication method provided in the present application is not executed.

[0283] Optionally, in the case that the terminal device is in a non-communication state and receives a call request of the other terminal device, the trigger request sent by the control plane network element to the first network element.

[0284] Correspondingly, the first network element receives the trigger request from the control plane network element.

[0285] Exemplarily, please refer to FIG. 5, which is an interaction schematic diagram of another communication method provided in an embodiment of the present application. In FIG. 5, the terminal device is turned off as an example. The communication devices involved in the method can include a terminal device, a first network element, a second network element, a third network element, a control plane network element and a media plane network element, and a multi-modal large model, etc. These communication devices can refer to the description of FIG. 1 and FIG. 2, which will not be described here. The method includes but is not limited to the following steps S501 to S505, wherein:

[0286] S501, the terminal device is turned off.

[0287] S502, the other terminal device sends a call request to the control plane network element, and the call request is used to call the terminal device.

[0288] Correspondingly, the control plane network element receives the call request from the other terminal device.

[0289] The call request includes the identification information of the other terminal device and the identification information of the terminal device. The call request can also include the call content, for example, the other terminal device makes an appointment with the user of the terminal device for an afternoon architecture seminar, or an annual procurement plan, etc.

[0290] Since the terminal device is powered off, the terminal device cannot receive the call request at this time. The control plane network element obtains the call request, and optionally determines that the terminal device is in a non-communication state according to the communication state and the subscription information of the terminal device. The subscription information can determine the network information accessed by the terminal device before, so as to determine that the terminal device does not access the communication network at present, and further determine that the terminal device is in a non-communication state.

[0291] S503, the control plane network element starts an unreachable forward flow.

[0292] It should be understood that the control plane network element can also start a flow other than the unreachable forward flow to achieve the same function, and the name of the flow is not limited in the present application.

[0293] S504, the control plane network element sends a trigger request to the first network element.

[0294] Correspondingly, the first network element receives the trigger request of the control plane network element. Specifically, the control plane network element sends the trigger request to the first network element based on the resource identifier of the first network element. In the embodiment of the present application, the trigger request is used to start the media information interaction function.

[0295] S505, the first network element starts the media information interaction function.

[0296] It can be understood that when the terminal device is powered off, real-time communication with other terminal devices or network devices cannot be performed. The call request sent by the other terminal device to the terminal device is received by the control plane network element, and when the control plane network element determines that the terminal device is in a non-communication state, a trigger request is sent to the first network element to start the media information interaction function, so that the communication method provided in the present application is realized through the network side, and the flexibility of processing media information is improved.

[0297] It should be noted that FIG. 5 is exemplified by the terminal device being powered off. In fact, there can be other non-communication state conditions of the terminal device.

[0298] The above two application scenarios are examples of executing the communication method of the present application. In fact, there can be other application scenarios.

[0299] The communication method provided in the present application will be analyzed in detail below.

[0300] Please refer to FIG. 6, which is an interaction diagram of another communication method provided in the embodiments of the present application. The communication devices involved in the method can include terminal equipment, a first network element, a second network element, a third network element, a control plane network element and a media plane network element, as well as a multi-modal large model, a multi-modal conversion device and other intelligent agents, etc. These communication devices can refer to the descriptions of FIG. 1 and FIG. 2, which will not be repeated here. The method includes, but is not limited to, the following steps S601-S603, wherein step S601 can be executed after step S323 or S423 or S505 described above.

[0301] S601, the media plane network element sends first information to the first network element, and the first information includes the user intent collected by the terminal equipment.

[0302] Correspondingly, the first network element receives the first information from the media plane network element. Specifically, the media plane network element sends the first information to the first network element based on the resource identifier of the first network element.

[0303] In the embodiments of the present application, the media plane network element can obtain the resource identifier of the first network element from the control plane network element. The control plane network element can obtain the resource identifier of the first network element from the third network element, as shown in FIG. 3 and FIG. 4, and the confirmation message of the third-party registration message sent by the third network element can include the resource identifier of the first network element. In this way, the control plane network element can communicate with the first network element.

[0304] Optionally, the terminal equipment transmits the first information to the first network element through the media plane network element. The device for transmission can include the control plane network element.

[0305] Optionally, the terminal equipment transmits the user intent collected by the terminal equipment to the first network element through the media plane network element; the media plane network element sends the user intent to the multi-modal conversion device; the multi-modal conversion device converts the user intent to obtain the first information; the multi-modal conversion device sends the first information to the media plane network element; and the media plane network element sends the first information to the first network element. In this way, the type of the user intent collected by the terminal equipment is converted by the multi-modal conversion device, so that the first information after conversion can be processed by the first network element, which helps to improve the effectiveness of media information processing.

[0306] In the embodiments of the present application, the user intent of the first information can be the intent of the user of the terminal equipment. For example, the user of the terminal equipment sends a voice instruction to the terminal equipment, and the text corresponding to the voice instruction is "Is there a three-party meeting today?". The user intent of the first information can also be the intent of the user of the other terminal equipment received by the terminal equipment. For example, the user of the other terminal equipment sends a mail to the terminal equipment through the other terminal equipment, and the body of the mail is "Hello, can we discuss the technical solution at 2 pm?".

[0307] S602, the first network element obtains media vector information of the terminal device from the second network element, and the media vector information is associated with the behavior information of the user collected by the terminal device.

[0308] Correspondingly, the second network element receives the media vector information of the terminal device from the first network element. Specifically, the first network element sends an information obtaining request to the second network element based on the resource identifier of the second network element, and the second network element sends the media vector information of the terminal device to the first network element.

[0309] In the embodiments of the present application, the media vector information includes the user portrait obtained by continuously reflecting the behavior information of the user. The media vector information in step S602 is the user portrait and the evaluation information of the user portrait, and the user portrait can include multiple long-term memories and can also include evaluation information related to each long-term memory. The present application does not limit the method of obtaining media vector information and the process of reflection, and specific descriptions can be referred to the description of FIG. 7 described later.

[0310] S603, the first network element calls a multi-modal large model according to the media vector information and the first information to obtain the identified user intent and / or the response related to the identified user intent.

[0311] In the embodiments of the present application, the response related to the identified user intent can be understood as the feedback corresponding to the user intent, and the feedback can be certain information or uncertain information. For example, in the case that the user intent of the first information is the intent of the user of the terminal device, and the text corresponding to the user intent is “What is the work arrangement this afternoon?”, the response can be certain information such as “There is a meeting in the afternoon” or “There is no meeting in the afternoon”. The response can be uncertain information such as “The meeting in the afternoon depends on the time of the leader”.

[0312] For another example, in the case that the user intent of the first information is the intent of the user of the other terminal device received by the terminal device, and the text corresponding to the user intent is “Hello, can we discuss the technical solution at 2 pm?”, the response can be a certain information “Yes” or “No”, or an uncertain opinion such as the suggestion “Hello, about the problem you mentioned, I need to consult the opinion of the master, and I will inform you of the further information”.

[0313] It can be understood that different responses can be given according to different user intents, including certain information and uncertain information. These two kinds of information can be completely different replies, and when the multi-modal large model can give the user of the terminal device two different replies, it is considered that the behavior information of the user has been reflected and the user portrait has been generated.

[0314] In some feasible examples, step S603 can include: the first network element obtaining a first prompt word according to the first information and the media vector information; the first network element sending a first request including the first prompt word to the multi-modal large model, the first request being used to request to identify a user intent; and the first network element receiving the identified user intent and / or a response related to the identified user from the multi-modal large model.

[0315] Correspondingly, the multi-modal large model receives the first request from the first network element; and the multi-modal large model sends the identified user intent and / or the response related to the identified user to the first network element.

[0316] The first prompt word can be a combined word of the first information and the media vector information. For example, the first information is S1 and the media vector information is I1, and the first prompt word can be S1+I1. The first prompt word can also be information that is a combined word of the first information and the media vector information after format conversion. For example, the first network element can convert the combined word of the first information and the media vector information into the first prompt word according to an embedding parameter required by the multi-modal large model. In this way, the first prompt word input to the multi-modal large model can be directly used by the multi-modal large model, thereby improving the effectiveness of data processing.

[0317] The media vector information can include a plurality of long-time memories. For example, the first prompt word is as follows:

[0318] Caller: A

[0319] Content: T1

[0320] Master status: Offline

[0321] Master profile1: LTM1, importance level 1

[0322] Master profile1: LTM2, importance level 2

[0323] Master profile1: LTM3, importance level 3

[0324] …

[0325] Master profile1: LTMx, importance level x

[0326] Question: According to A’s request, how to answer?

[0327] Wherein, the Master profile, the LTM and the importance level can refer to the foregoing, and will not be described here again. A represents a caller, which can be understood as a user of another terminal device. T1 represents input content, which is the first information in the method shown in FIG. 6. The Master status represents the master status of the terminal device, which is offline in this example, and can be understood as the non-communication state described above. The Question represents an intention, which is how to respond according to the request of A in this example.

[0328] It can be understood that the first network element can interact with the multi-modal large model, so as to call the multi-modal large model according to the media vector information and the first information to obtain the identified user intention and / or the response related to the identified user intention.

[0329] In the method shown in FIG. 6, after the first network element receives the first information including the user intention collected by the terminal device, the media vector information of the terminal device is obtained from the second network element, so as to call the multi-modal large model according to the media vector information and the first information to obtain the identified user intention and / or the response related to the identified user intention. That is, the multi-modal large model is called by the network element in the network side to obtain the user intention of the first information and / or the response related to the user intention, which can improve the support for the media information interaction function. And the media vector information associated with the behavior information of the user is used to identify the user intention, which can also improve the accuracy of intention identification.

[0330] Optionally, the identified user intention and / or the response related to the identified user obtained by the multi-modal large model can not meet the expectation. The method can further include: the first network element determines that the identified user intention and / or the response related to the identified user as a return result does not meet a target condition; the first network element obtains a second prompt word according to the first information, the media vector information and the first result; the first network element sends a first request including the second prompt word to the multi-modal large model, the first request being used to request to identify the user intention; and the first network element receives the identified user intention and / or the response related to the identified user from the multi-modal large model.

[0331] Wherein, the target condition is used to determine whether the return result meets the expectation, and the content of the target condition is not limited in the present application. The second prompt word can be a combination word of the first information, the media vector information and the first result.

[0332] It can be understood that when the returned result does not meet the target condition, the first network element can plan the processing logic again to call the multi-modal large model to obtain the identified user intent and / or the response related to the identified user according to the second prompt word. If the identified user intent and / or the response related to the identified user still does not meet the target condition, the method can be executed again to obtain a returned result that meets the target condition.

[0333] Optionally, when the returned result does not meet the target condition and the receiving number of the returned result is equal to the preset number, the method is no longer executed. In this way, the method can be prevented from being executed for a long time to prevent timeout.

[0334] Optionally, after obtaining the returned result that meets the target condition, the first network element sends the returned result to the terminal device.

[0335] Correspondingly, the terminal device receives the returned result from the first network element. Specifically, the first network element sends the returned result to the terminal device through the media plane network element. The returned result can also be obtained by conversion through the multi-modal conversion device. In this way, the terminal device can receive the returned result and display or play the returned result.

[0336] Exemplarily, assuming that the first information is S1, the media vector information is I1, and the returned result obtained by the first identification is GT1. In the case that GTI does not meet the target condition, the first network element can plan the processing logic again to obtain the first prompt word PG1, and generate the second prompt word according to PG1, GT1, SI and I1. The first network element sends the second prompt word to the multi-modal large model to obtain the returned result obtained by the second identification, such as GT2. In the case that GT2 does not meet the target condition, the processing logic can be planned again to obtain a new first prompt word PG1, and a new second prompt word can be generated according to PG1, GT2, SI and I1. The first network element sends the new second prompt word to the multi-modal large model to obtain the returned result obtained by the third identification, such as GT3. In the case that GT3 does not meet the target condition, if the preset number is 3 times, the first network element sends GT3 to the media plane network element. In the case that GT3 meets the target condition, the first network element can also send GT3 to the media plane network element.

[0337] Optionally, the method can further include that the first network element communicates with other intelligent agents. For example, after the first network element determines that the returned result of the multi-modal large model, such as GT2, does not meet the target condition, the first network element communicates with other intelligent agents to reflect GT2. In this way, the first network element can improve the accuracy of intent recognition by communicating with other intelligent agents.

[0338] The method of obtaining (embedding) the media vector information is described below.

[0339] Please refer to FIG. 7, which is an interaction diagram of another communication method provided by the embodiments of the present application. The communication devices involved in the method can include terminal equipment, a first network element, a second network element, a third network element, a control plane network element and a media plane network element, as well as a multi-modal large model and a multi-modal conversion device, etc. These communication devices can refer to the descriptions of FIG. 1 and FIG. 2, which will not be repeated here. The method includes, but is not limited to, the following steps S701-S703, wherein step S701 can be executed after step S323 or step S423 or step S505 described above, and step S601 can be executed after step S703.

[0340] S701, the media plane network element sends second information to the first network element, and the second information includes the behavior information of the user collected by the terminal equipment.

[0341] Correspondingly, the first network element receives the second information from the media plane network element. Specifically, the media plane network element sends the second information to the first network element based on the resource identifier of the first network element.

[0342] Wherein, the behavior information of the user can refer to the foregoing, which will not be repeated here. The second information can be one or more, such as I1, I2, I3, I4 and I5 described above, and each second information executes the subsequent steps S702 and S703 one by one.

[0343] Optionally, the terminal equipment transmits the second information to the first network element through the media plane network element. Wherein, the device for transmitting can include a control plane network element.

[0344] Optionally, the terminal equipment transmits the behavior information of the user collected by the terminal equipment to the first network element through the media plane network element; the media plane network element sends the behavior information to the multi-modal conversion device; the multi-modal conversion device converts the behavior information to obtain the second information; the multi-modal conversion device sends the second information to the media plane network element; and the media plane network element sends the second information to the first network element. In this way, the type of the behavior information of the user collected by the terminal equipment is converted by the multi-modal conversion device, so that the second information after conversion can be processed by the first network element, which is beneficial to improve the effectiveness of media information processing.

[0345] S702, the first network element calls the multi-modal large model according to the second information to obtain media vector information of the terminal equipment.

[0346] In some feasible examples, step S703 can include: the first network element obtains a second prompt word according to the second information; the first network element sends a second request including the second prompt word to the multi-modal large model, the second request being used to request to obtain the media vector information; and the first network element receives the media vector information from the multi-modal large model.

[0347] Correspondingly, the multi-modal large model receives a second request from the first network element; the multi-modal large model sends media vector information to the first network element.

[0348] The second prompt word can be information for format conversion of the second information. For example, the first network element converts the second information into the second prompt word according to embedding parameters required by the multi-modal large model. In this way, the second prompt word input into the multi-modal large model can be directly used by the multi-modal large model, improving the effectiveness of data processing.

[0349] It can be understood that the first network element can interact with the multi-modal large model, and thus call the multi-modal large model to obtain the media vector information according to the second information.

[0350] In some feasible examples, the media vector information includes a user portrait of the terminal device and evaluation information of the user portrait. In this way, the user portrait can be selected according to the evaluation information of the user portrait, which is beneficial to improve the efficiency and accuracy of intent recognition.

[0351] In some feasible examples, the evaluation dimension of the evaluation information of the user portrait includes at least one of the following: importance, recency, relevance. In this way, the long-term memory corresponding to the most important, most relevant, and most recent user portrait can be selected for intent recognition, which is beneficial to improve the efficiency and accuracy of intent recognition.

[0352] Optionally, the first network element sends a second request to the multi-modal large model if no new second information is received within a first time length.

[0353] The second request includes one or more second information. The present application does not limit the first time length. It can be understood that if no new second information is received within the first time length, the previously received second information that has not been transmitted can be sent to the multi-modal large model, and the multi-modal large model is requested to obtain the media vector information. In this way, the media vector information corresponding to at least one second information can be obtained, avoiding processing each second information as media vector information. Processing multiple second information in a time period is beneficial to improve the accuracy of obtaining user-level context.

[0354] S703, the first network element sends a storage request including the media vector information to the second network element, and the storage request is used to request to store the media vector information.

[0355] Correspondingly, the second network element receives the storage request from the first network element. Specifically, the first network element sends the storage request including the media vector information to the second network element based on the resource identifier of the second network element.

[0356] It can be understood that in the method shown in FIG. 7, after the first network element receives the behavior information of the user collected by the terminal device, the multimodal large model can be called according to the information to obtain the media vector information of the terminal device. Then the media vector information of the terminal device is stored through the second network element, thereby facilitating subsequent acquisition of the user intention and / or the response related to the user intention through the media vector information of the terminal device.

[0357] In some feasible examples, the method can further include: when the timing time arrives, the first network element calls the multimodal large model to reflect the media vector information.

[0358] Exemplarily, please refer to FIG. 8, which is an interaction schematic diagram of another communication method provided by the embodiments of the present application. The communication devices involved in the method can include a terminal device, a first network element, a second network element, a third network element, a control plane network element and a media plane network element, and a multimodal large model and a multimodal conversion device, etc. These communication devices can refer to the description of FIGS. 1 and 2, which will not be described here. The method includes but is not limited to the following steps S801 to S803, wherein step S801 can be executed after step S703 described above, and step S601 can be executed after step S803.

[0359] S801, when the timing time arrives, the first network element sends an information acquisition request to the second network element, and the information acquisition request is used to acquire the media vector information of the terminal device.

[0360] Correspondingly, the second network element receives the information acquisition request from the first network element. Specifically, the first network element sends the information acquisition request to the second network element based on the resource identifier of the second network element.

[0361] S802, the second network element sends the media vector information to the first network element.

[0362] Correspondingly, the first network element receives the media vector information from the second network element.

[0363] S803, the first network element generates a third prompt word according to the media vector information.

[0364] The third prompt word can be information for format conversion of the media vector information. For example, optionally, the first network element converts the media vector information into the third prompt word according to the embedding parameters required by the multimodal large model. In this way, the third prompt word input to the multimodal large model can be directly used by the multimodal large model, improving the effectiveness of data processing.

[0365] S804, the first network element sends a third request including the third prompt word to the multimodal large model, and the third request is used to request to reflect the media vector information.

[0366] S805, the multi-modal large model sends the reflected media vector information to the first network element.

[0367] Correspondingly, the first network element receives the reflected media vector information from the multi-modal large model.

[0368] S806, the first network element sends a storage request including the media vector information to the second network element, and the storage request is used to request storage of the media vector information.

[0369] Correspondingly, the second network element receives the storage request from the first network element. Specifically, the first network element sends the storage request including the media vector information to the second network element based on the resource identifier of the second network element.

[0370] S807, the second network element stores the media vector information.

[0371] It can be understood that in the method shown in FIG. 8, by reflecting the media vector information in time, the user portrait and the evaluation information of the user portrait can be reflected, which is beneficial to improve the accuracy of user intent recognition.

[0372] Exemplarily, it is assumed that the first network element has sent the behavior information of the aforementioned I1, I2, I3, I4 and I5 to the multi-modal large model one by one, so that the multi-modal large model obtains the media vector information corresponding to I1, I2, I3, I4 and I5 respectively. Through the implementation of this example, the first network element can convert the media vector information corresponding to I1, I2, I3, I4 and I5 respectively into a third prompt word such as PG3 according to the embedding parameters required by the multi-modal large model. Then the first network element sends a third request including the third prompt word to the multi-modal large model, and the multi-modal large model reflects the media vector information according to the third prompt word, thereby obtaining the reflected media vector information, and can obtain more perfect user portrait and evaluation information of the user portrait, which is beneficial to improve the accuracy of user intent recognition.

[0373] The above describes the method of the embodiment of the application in detail, and the device of the embodiment of the application is provided below.

[0374] Please refer to FIG. 9, which is a structural schematic diagram of a communication device provided by an embodiment of the application. The communication device can include a transceiver unit 901 and a processing unit 902. The transceiver unit 901 can be a device with input (reception) or output (transmission) of signals, used for signal transmission with other network devices or other devices in the device.

[0375] The processing unit 902 can be a device with processing functions, and can include one or more processors. The processor can be a general-purpose processor or a special-purpose processor, etc. The processor can be a baseband processor or a central processing unit. The baseband processor can be used to process communication protocols and communication data, and the central processing unit can be used to control the device (such as a host node, a relay node, or a chip, etc.), execute software programs, and process data of the software programs.

[0376] The communication device can include a terminal device, a first network element, a second network element, a third network element, a control plane network element, a media plane network element, a multi-modal conversion device, a multi-modal large model, and other intelligent agents, or devices therein. The communication device can be applied to the two application scenarios described above.

[0377] When the communication device is a first network element, it includes:

[0378] The transceiver unit 901 is configured to receive first information, the first information including user intent collected by a terminal device;

[0379] The transceiver unit 901 is further configured to obtain media vector information of the terminal device from a second network element, the media vector information being associated with behavior information of a user collected by the terminal device;

[0380] The processing unit 902 is configured to invoke a multi-modal large model according to the media vector information and the first information to obtain an identified user intent and / or a response related to the identified user intent.

[0381] In some possible examples, the processing unit 902 is configured to obtain a first prompt word according to the first information and the media vector information; the transceiver unit 901 is configured to send a first request including the first prompt word to a multi-modal large model, the first request being used to request identification of a user intent; and the transceiver unit 901 is further configured to receive, from the multi-modal large model, an identified user intent and / or a response related to the identified user intent.

[0382] In some possible examples, the transceiver unit 901 is further configured to receive second information, the second information including behavior information of a user collected by the terminal device; the processing unit 902 is further configured to invoke the multi-modal large model according to the second information to obtain the media vector information; and the transceiver unit 901 is further configured to send a storage request including the media vector information to the second network element, the storage request being used to request storage of the media vector information.

[0383] In some possible examples, the processing unit 902 is configured to acquire a second prompt word according to the second information; the transceiver unit 901 is further configured to send, to the multi-modal large model, a second request including the second prompt word, the second request being used to request to acquire the media vector information; and the transceiver unit 901 is further configured to receive, from the multi-modal large model, the media vector information.

[0384] In some possible examples, the media vector information includes a user portrait of the terminal device and evaluation information of the user portrait, and an evaluation dimension of the evaluation information includes at least one of importance, recency, and relevance.

[0385] In some possible examples, the processing unit 902 is further configured to invoke the multi-modal large model to reflect on the media vector information.

[0386] In some possible examples, the resource of the first network element and the resource of the second network element are configured for the terminal device by a third network element.

[0387] In some possible examples, the transceiver unit 901 is further configured to send, to the third network element, a first registration request, the first registration request being used to register a resource identifier of the first network element; and the transceiver unit 901 is further configured to receive, from the third network element, a confirmation message of the first registration request.

[0388] In some possible examples, the transceiver unit 901 is further configured to receive, from the third network element, a first check message, the first check message being used to confirm the resource identifier of the first network element; and the transceiver unit 901 is further configured to send, to the third network element, a confirmation message of the first resource identifier.

[0389] In some possible examples, the transceiver unit 901 is further configured to receive verification information of the terminal device; and the processing unit 902 is further configured to verify the verification information of the terminal device, and allow a user of the terminal device to access when the verification is successful.

[0390] In some possible examples, the processing unit 902 is further configured to decrypt received information and / or encrypt sent information.

[0391] In some possible examples, the transceiver unit 901 is further configured to establish a media long connection with the terminal device.

[0392] In some possible examples, the processing unit 902 is further configured to start a media information interaction function, the media information interaction function being used to interact with the terminal device on media information.

[0393] When the communication apparatus is a second network element, comprising:

[0394] The transceiver unit 901 is configured to receive a storage request of the first network element, the storage request being used to request to store the media vector information of the terminal device.

[0395] The processing unit 902 is configured to store the media vector information of the terminal device.

[0396] In a possible example, the transceiver unit 901 is further configured to receive an information acquisition request of the first network element, the information acquisition request being used to acquire the media vector information of the terminal device; and send the media vector information to the first network element.

[0397] In a possible example, the resource of the second network element is configured by a third network element for the terminal device.

[0398] In a possible example, the transceiver unit 901 is further configured to send a second registration request to the third network element, the second registration request being used to register the resource identifier of the second network element; and receive a confirmation message of the second registration request from the third network element.

[0399] In a possible example, the transceiver unit 901 is further configured to receive a second check message from the third network element, the second check message being used to confirm the resource identifier of the second network element; and send a confirmation message of the second resource identifier to the third network element.

[0400] In a possible example, the transceiver unit 901 is further configured to receive verification information of the terminal device; and the processing unit 902 is further configured to verify the verification information of the terminal device, and allow a user of the terminal device to access when the verification is successful.

[0401] In a possible example, the processing unit 902 is further configured to decrypt the received information and / or encrypt the sent information.

[0402] When the communication apparatus is the third network element, comprising:

[0403] The processing unit 902 is configured to configure the resource of the first network element and the resource of the second network element for the terminal device.

[0404] In a possible example, the transceiver unit 901 is configured to receive a first registration request of the first network element and / or a second registration request of the second network element, the first registration request being used to register the resource identifier of the first network element, and the second registration request being used to register the resource identifier of the second network element; and the transceiver unit 901 is further configured to send a confirmation message of the first registration request to the first network element; and / or send a confirmation message of the second registration request to the second network element.

[0405] In a possible example, the transceiver unit 901 is configured to send a first check message to the first network element, the first check message being used to confirm the resource identity of the first network element; and / or send a second check message to the second network element, the second check message being used to confirm the resource identity of the second network element; and receive a confirmation message of the first check message of the first network element; and / or receive a confirmation message of the second check message of the second network element.

[0406] It should be noted that the implementation of each unit can also correspond to the description of the corresponding method embodiments in any of FIGS. 3 to 8.

[0407] Referring to FIG. 10, FIG. 10 is a structural schematic diagram of another communication apparatus provided in an embodiment of the present application. The communication apparatus can be a terminal device, a first network element, a second network element, a third network element, a control plane network element, a media plane network element, a multi-modal conversion device, a multi-modal large model, and other intelligent agents, or an apparatus therein, to implement the method described in the method embodiments.

[0408] As shown in FIG. 10, the communication apparatus can include a processor 111 and a storage medium 112. The processor 111 can also be referred to as a processing unit, and can implement certain control functions. The storage medium 112 can also be referred to as a storage unit, or a memory. The storage medium 112 has instructions 114 stored thereon. The instructions 114 can be run on the processor 111, so that the communication apparatus performs any of the methods described in FIGS. 3 to 8 of the embodiments of the present application.

[0409] Optionally, the processor 111 can include instructions 113, which can be run on the processor 111, so that the communication apparatus performs any of the methods described in FIGS. 3 to 8 of the embodiments of the present application.

[0410] The communication apparatus described in the above embodiments can be a first device or a second device, but the scope of the apparatus described in the present application is not limited thereto. The communication apparatus can be a standalone device or can be part of a larger device. For example, the communication apparatus can be:

[0411] (1) a standalone integrated circuit (IC), or a chip, or a chip system or subsystem;

[0412] (2) a set of one or more ICs, optionally including storage components for storing data and / or instructions;

[0413] (3) an ASIC, such as a modem;

[0414] (4) a module that can be embedded within other devices;

[0415] The embodiment of the present application further provides a computer readable storage medium, which stores a computer program. The computer program is executed by a processor to implement the related processes in the communication method provided by the method embodiment.

[0416] The embodiment of the present application further provides a computer program product for storing a computer program. When the computer program is run on a computer (or a processor), the computer is caused to execute one or more steps in any of the above communication methods. The constituent modules of the above-mentioned devices, if realized in the form of software function units and sold or used as independent products, can be stored in a computer readable storage medium.

[0417] The embodiment of the present application provides a chip, comprising a processor, configured to call and run instructions stored in a memory, so that a communication device installed with the chip executes any of the above methods.

[0418] The embodiment of the present application further provides another chip, comprising an input interface, an output interface and a processing circuit. The input interface, the output interface and the circuit are connected through an internal connection path. The processing circuit is configured to execute any of the above methods. Optionally, the chip further comprises a memory. The input interface, the output interface, the processor and the memory are connected through the internal connection path. The processor is configured to execute the code in the memory. When the code is executed, the processor is configured to execute any of the above methods.

[0419] The embodiment of the present application further provides a chip system, comprising at least one processor and a communication interface. The communication interface and the at least one processor are connected through a line. The at least one processor is configured to run a computer program or instructions to execute any of the above methods. The chip system can be composed of a chip, or can include a chip and other discrete devices.

[0420] The embodiment of the present application further provides a communication system, which comprises a terminal device, a first network element, a second network element, a third network element, a control plane network element, a media plane network element, a multi-modal conversion device, a multi-modal large model and other intelligent agents, or devices therein. The specific description can be referred to the above-mentioned any method.

[0421] It should be appreciated that the memory mentioned in the embodiments of the present application can be a volatile memory or a non-volatile memory, or can include both volatile and non-volatile memories. Among them, the non-volatile memory can be a hard disk drive (HDD), a solid-state drive (SSD), a ROM, a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically EPROM (EEPROM), or a flash memory. The volatile memory can be a RAM used as an external cache. The memory can be any other medium capable of carrying or storing desired program codes in the form of instructions or data structures and capable of being accessed by a computer, but is not limited thereto. The memory in the embodiments of the present application can also be a circuit or any other device capable of realizing a storage function, used for storing program instructions and / or data.

[0422] It should also be understood that the processor mentioned in the embodiments of the present application can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor or can be any conventional processor.

[0423] It should be noted that when the processor is a general-purpose processor, a DSP, an ASIC, an FPGA or other programmable logic device, a discrete gate or transistor logic device, a discrete hardware component, the memory (storage module) is integrated in the processor.

[0424] It should be noted that the memory described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0425] Those skilled in the art can appreciate that units and algorithm steps of each example described in combination with the embodiments provided herein can be realized by electronic hardware, or a combination of computer software and electronic hardware. Whether the functions are performed in hardware or software depends on the specific application and design constraints of the technical solutions. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0426] In several embodiments provided in the present application, it should be understood that the disclosed system, device and method can be implemented in other ways. For example, the device embodiments described above are only schematic. The division of units is only a logical function division. In actual implementation, there can be another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some interface, device or unit, which can be electrical, mechanical or other forms.

[0427] The units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, that is, they can be located in one place, or can be distributed on a plurality of network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.

[0428] In addition, the functional units in each embodiment of the present application can be integrated in one processing unit, or can be physically present as individual units, or two or more units can be integrated in one unit.

[0429] The steps in the method embodiments of the present application can be adjusted, combined and deleted in sequence according to actual needs. The steps of each embodiment can be partially executed (for example, the terminal device can not execute the steps executed by the terminal device in the above embodiments). The execution order of different steps can be changed. The embodiments described herein can be combined with other embodiments, and different steps of different embodiments in the present application can be combined.

[0430] The modules / units in the device embodiments of the present application can be combined, divided and deleted according to actual needs.

[0431] In the present application, "embodiment" means that the specific features, structures or characteristics described in combination with the embodiments can be included in at least one embodiment of the present application. The phrase appears at various places in the specification does not necessarily refer to the same embodiment, and is not an independent or alternative embodiment that is not mutually exclusive with other embodiments.

[0432] In the present application, a communication protocol or specification, such as a 3GPP communication protocol, can be referred to.

[0433] In the present application, the terms "first", "second", "third", "fourth" and the like (if any) are used to distinguish similar objects, and do not necessarily have to describe a specific order or sequence.

[0434] In the present application, "includes" can be a containing relationship, or can be an equal relationship. For example, A includes B, which can be A containing B in addition to other content, or A and B are the same content.

[0435] In the description of the present application, unless otherwise specified, " / " represents that the objects before and after are in an "or" relationship, for example, A / B can represent A or B; "and / or" in the present application is only a description of the relationship between the associated objects, which means that there can be three kinds of relationships, for example, A and / or B, which can represent: A exists alone, A and B exist together, and B exists alone, of which A and B can be singular or plural. And in the description of the present application, unless otherwise specified, "multiple" means two or more than two. "At least one of the following" or the like refers to any combination of these items, including any combination of single or multiple items. For example, at least one of a, b, or c, can represent: a, b, c, a-b, a-c, b-c, or a-b-c, where a, b, and c can be single or multiple.

[0436] It should be understood that in various embodiments of the present application, the size of the sequence number of the above-mentioned processes does not mean the order of execution, and the execution order of the processes should be determined according to its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

Claims

1. A communication system, characterized by The system comprises a first network element and a second network element, wherein: The second network element is configured to store media vector information of a terminal device, the media vector information being associated with behavior information of a user collected by the terminal device; The first network element is configured to receive first information, the first information comprising an intention of the user collected by the terminal device; The first network element is further configured to obtain the media vector information from the second network element; The first network element is further configured to invoke a multi-modal large model according to the media vector information and the first information to obtain an identified intention of the user and / or a response related to the identified intention of the user.

2. The system of claim 1, wherein, The system further comprises a third network element configured to configure resources of the first network element and resources of the second network element for the terminal device.

3. The system of claim 2, wherein: The third network element is further configured to receive a first registration request of the first network element and / or a second registration request of the second network element, the first registration request being configured to register a resource identifier of the first network element, and the second registration request being configured to register a resource identifier of the second network element; The third network element is further configured to send a confirmation message of the first registration request to the first network element and / or a confirmation message of the second registration request to the second network element.

4. The system of claim 2, wherein: The third network element is further configured to send a first check message to the first network element, the first check message being configured to confirm the resource identifier of the first network element, and / or send a second check message to the second network element, the second check message being configured to confirm the resource identifier of the second network element; The third network element is further configured to receive a confirmation message of the first check message of the first network element and / or a confirmation message of the second check message of the second network element.

5. The system of any one of claims 1 to 4, wherein, The first network element is configured to invoke a multi-modal large model according to the media vector information and the first information to obtain an identified intention of the user and / or a response related to the identified intention of the user, comprising: obtaining a first prompt word according to the first information and the media vector information; sending a first request comprising the first prompt word to the multi-modal large model, the first request being configured to request to identify the intention of the user; receiving the identified intention of the user and / or the response related to the identified intention of the user from the multi-modal large model.

6. The system of any one of claims 1 to 5, wherein: The first network element is further configured to receive second information, the second information comprising behavior information of the user collected by the terminal device; The first network element is further configured to invoke the multi-modal large model according to the second information to obtain the media vector information; The first network element is further configured to send a storage request comprising the media vector information to the second network element, the storage request being configured to request to store the media vector information.

7. The system of claim 6, wherein, The first network element is further configured to invoke the multi-modal large model according to the second information to obtain the media vector information, comprising: obtaining a second prompt word according to the second information; sending, to a multi-modal large model, a second request including the second prompt word, the second request being used to request to obtain the media vector information; receiving, from the multi-modal large model, the media vector information.

8. The system of any one of claims 1 to 7, wherein, The media vector information includes a user portrait of the terminal device and evaluation information of the user portrait, and an evaluation dimension of the evaluation information includes at least one of importance, recency, and relevance.

9. The system of any one of claims 1 to 8, wherein, The first network element is further configured to invoke the multi-modal large model to reflect on the media vector information.

10. The system of any one of claims 1 to 9, wherein, The first network element and the second network element are further configured to receive verification information of the terminal device, verify the verification information of the terminal device, and allow a user of the terminal device to access when the verification is successful.

11. The system of any one of claims 1 to 10, wherein, The first network element is further configured to decrypt received information and / or encrypt sent information.

12. The system of any of claims 1-11, wherein The first network element is further configured to establish a media long connection with the terminal device; or The first network element is further configured to start a media information interaction function, and the media information interaction function is used to interact with the terminal device.

13. A method of communication, comprising: The method is applied to a first network element, and the method includes: receiving first information, the first information including user intent collected by a terminal device; obtaining, from a second network element, media vector information of the terminal device, the media vector information being associated with behavior information of a user collected by the terminal device; invoking a multi-modal large model according to the media vector information and the first information to obtain identified user intent and / or a response related to the identified user intent.

14. The method of claim 13, wherein, The invoking of the multi-modal large model according to the media vector information and the first information to obtain the identified user intent and / or the response related to the identified user intent includes: obtaining a first prompt word according to the first information and the media vector information; sending, to a multi-modal large model, a first request including the first prompt word, the first request being used to request to identify user intent; receiving, from the multi-modal large model, the identified user intent and / or the response related to the identified user intent.

15. The method according to claim 13 or 14, characterized in that, The method further includes: receiving second information, the second information including behavior information of a user collected by the terminal device; invoking the multi-modal large model according to the second information to obtain the media vector information; sending, to the second network element, a storage request including the media vector information, the storage request being used to request to store the media vector information.

16. The method of claim 15, wherein, The invoking of the multi-modal large model according to the second information to obtain the media vector information includes: obtaining a second prompt word according to the second information; sending, to a multi-modal large model, a second request including the second prompt word, the second request being used to request to obtain the media vector information; receiving, from the multi-modal large model, the media vector information.

17. The method according to any one of claims 13 to 15, characterized in that, The media vector information includes a user portrait of the terminal device and evaluation information of the user portrait, and an evaluation dimension of the evaluation information includes at least one of importance, recency, and relevance.

18. The method according to any one of claims 13 to 17, characterized in that, The method further includes: At the time of the timing time, the multi-modal large model is called to reflect on the media vector information.

19. The method according to any one of claims 13 to 18, characterized in that, The resource of the first network element and the resource of the second network element are configured by a third network element for a terminal device.

20. The method of claim 19, wherein, Further comprising: sending a first registration request to the third network element, the first registration request being used to register the resource identifier of the first network element; receiving a confirmation message of the first registration request from the third network element.

21. The method of claim 19, wherein, Further comprising: receiving a first check message from the third network element, the first check message being used to confirm the resource identifier of the first network element; sending a confirmation message of the first resource identifier to the third network element.

22. The method of any one of claims 13-21, wherein, Further comprising: receiving authentication information of the terminal device; verifying the authentication information of the terminal device, and allowing a user of the terminal device to access when the verification is successful.

23. The method according to any one of claims 13 to 22, characterized in that, Further comprising: decrypting the received information and / or encrypting the sent information.

24. The method of any one of claims 13-23, wherein, Further comprising: establishing a media long connection with the terminal device; or starting a media information interaction function, the media information interaction function being used to interact with the terminal device on media information.

25. A communications device, characterized by The method comprises the steps of: receiving a first registration request from a terminal device, the first registration request being used to register a resource identifier of a first network element; sending a first check message to a third network element, the first check message being used to confirm the resource identifier of the first network element; receiving a confirmation message of the first resource identifier from the third network element; and sending a second registration request to the third network element, the second registration request being used to register a resource identifier of a second network element.

26. A communications device, characterized by The communication device comprises a processor and a storage medium, the storage medium storing instructions, when the instructions are run by the processor, the method according to any one of claims 13 to 24 is executed.

27. A computer readable storage medium, characterized in that, The computer readable storage medium comprises instructions, when the instructions are run by the processor, the method according to any one of claims 13 to 24 is implemented.

28. A chip, characterized by The processor is used to call and run the instructions stored in the memory, so that the communication device installed with the chip executes the method according to any one of claims 13 to 24.

Images