Delivery content generation method and device, storage medium and electronic equipment

Abstract

The invention discloses a launch content generation method and device, a storage medium and electronic equipment, and the method comprises the steps: determining a to-be-launched launch booth and reference launch data corresponding to the launch booth, and determining a launch label set according to the reference launch data and the launch booth, and calling a delivery text generation large model to generate a delivery text according to the delivery label set and the reference delivery data, determining a delivery picture according to the delivery label set, and generating target delivery content according to the delivery picture, the delivery text and the delivery label set. Therefore, according to the method of the embodiment of the invention, the delivery text is automatically and intelligently generated, the efficiency and convenience of generating the delivery text are improved, then the target delivery content is generated according to the delivery text and the delivery picture, and the target delivery content is automatically and intelligently generated, so that the efficiency and convenience of generating the target delivery content are improved, and the user experience is improved. The problem of low efficiency caused by the fact that the delivery content is generated depending on artificial experience is avoided.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a method, device, storage medium and electronic device for generating delivery content. Background Art

[0002] With the development of Internet finance, more and more financial institutions are providing users with financial services at their fingertips with the help of Internet financial platforms, which has greatly improved the popularity and convenience of financial services. Using technologies such as big data and artificial intelligence, Internet financial platforms can deliver different types of content to users according to different placement booths to enhance the user experience. For example, in some placement booths, users can receive copywriting related to financial products, and in other placement booths, users can receive copywriting related to user rights, etc.

[0003] In the related art, before an Internet financial platform places a booth to place content to users, it is usually necessary to develop a content placement strategy so as to place corresponding content according to the content placement strategy. Summary of the invention

[0004] This specification provides a method, device, storage medium and electronic device for generating delivery content. The technical solution is as follows:

[0005] In a first aspect, this specification provides a method for generating delivery content, the method comprising:

[0006] Determine a placement booth to be placed, obtain reference placement information for the placement booth, and determine a placement tag set based on the reference placement information and the placement booth;

[0007] Based on the delivery tag set and the reference delivery data, a delivery text generation model is called to perform delivery text generation processing to obtain a delivery text;

[0008] A delivery image is determined based on the delivery tag set, and target delivery content is generated based on the delivery tag set, the delivery text and the delivery image.

[0009] In a second aspect, this specification provides a delivery content generation device, the device comprising:

[0010] A label determination module is used to determine a placement booth to be placed, obtain reference placement information for the placement booth, and determine a placement label set based on the reference placement information and the placement booth;

[0011] A text generation module, used to call a large model for generating delivery text based on the delivery tag set and the reference delivery data to perform delivery text generation processing to obtain delivery text;

[0012] The content generation module is used to determine the delivery image based on the delivery tag set, and generate target delivery content based on the delivery tag set, the delivery text and the delivery image.

[0013] In a third aspect, the present specification provides a computer storage medium, wherein the computer storage medium has a plurality of instructions, wherein the instructions are suitable for being loaded by a processor and executing the above-mentioned method.

[0014] In a fourth aspect, the present specification provides a computer program product, wherein the computer program product stores at least one instruction, and the at least one instruction is loaded by a processor to execute the above method.

[0015] In a fifth aspect, the present specification provides an electronic device, which may include: a memory and a processor; wherein the memory stores a computer program, and the computer program is suitable for being loaded by the memory and executing the above method.

[0016] The beneficial effects of the technical solution provided in this specification include at least:

[0017] The delivery content generation method provided in the embodiment of this specification first determines the delivery booth to be delivered and the reference delivery materials corresponding to the delivery booth, and determines the delivery tag set according to the reference delivery materials and the delivery booth, and then calls the delivery text generation model to generate the delivery text according to the delivery tag set and the reference delivery materials, and then determines the delivery image according to the delivery tag set, and then generates the target delivery content according to the delivery image, delivery text and delivery tag set. Therefore, by determining the delivery tag set, and then calling the delivery text generation model to generate the delivery text according to the delivery tag set and the reference delivery materials, the delivery text is automatically and intelligently generated, which improves the efficiency and convenience of generating the delivery text. Subsequently, the target delivery content is generated according to the delivery text and the delivery image, which realizes the automatic and intelligent generation of the target delivery content, thereby improving the efficiency and convenience of generating the target delivery content, and avoiding the problem of inefficiency caused by relying on manual experience to generate delivery content. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without paying any creative work.

[0019] Figure 1 This is a schematic diagram of a scenario of a method for generating delivery content provided by an embodiment of this specification;

[0020] Figure 2It is a flowchart of a method for generating delivery content provided by an embodiment of this specification;

[0021] Figure 3 It is a flowchart of another method for generating delivery content provided by an embodiment of this specification;

[0022] Figure 4 It is a text diagram of a summary prompt word of a delivery label provided in an embodiment of this specification;

[0023] Figure 5 It is a text diagram of an indicator extraction prompt word provided in an embodiment of this specification;

[0024] Figure 6 It is a structural diagram of a delivery content generation device provided in an embodiment of this specification;

[0025] Figure 7 It is a structural schematic diagram of an electronic device provided in an embodiment of this specification. DETAILED DESCRIPTION

[0026] In order to make the purpose, features, and advantages of the embodiments of this specification more obvious and understandable, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this specification, not all of them. Based on the embodiments in this specification, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this specification.

[0027] In the description of this specification, it should be understood that the terms "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance. In the description of this specification, it should be noted that, unless otherwise clearly specified and limited, "including" and "having" and any of their variations are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally also includes steps or units that are not listed, or optionally also includes other steps or units inherent to these processes, methods, products or devices. For those of ordinary skill in the art, the specific meanings of the above terms in this specification can be understood in specific circumstances. In addition, in the description of this specification, unless otherwise specified, "multiple" refers to two or more. "And / or" describes the association relationship of associated objects, indicating that there can be three relationships, for example, A and / or B, which can represent: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the associated objects before and after are an "or" relationship.

[0028] In related technologies, Internet financial platforms can use technologies such as big data and artificial intelligence to develop content delivery strategies, and then deliver financial-related content to users at delivery booths based on the content delivery strategies. In the process of formulating content delivery strategies for Internet financial platforms, it is necessary to manually propose a hypothesis based on historical experience, then conduct a certain degree of verification based on statistical effects, and then develop strategies based on the proposed needs. In the above strategy development process, relying on manual experience to guide the production strategy, there is a problem of low strategy output efficiency, which leads to the problem of low output efficiency of delivery content. Therefore, how to improve the efficiency of Internet financial platforms in generating delivery content is a technical problem that needs to be solved urgently.

[0029] In order to solve the above technical problems, this specification is introduced and explained in conjunction with specific embodiments below.

[0030] In the following method embodiments, for the sake of convenience, only the electronic device is used as the execution subject of each step for introduction and description.

[0031] See also Figure 1 , is a schematic diagram of a scenario of a method for generating delivery content provided in an embodiment of this specification. Figure 1 The scenario shown may include at least a client cluster and a service platform 100 .

[0032] The client cluster may include at least one client, such as Figure 1 As shown, it specifically includes client 1 corresponding to user 1, client 2 corresponding to user 2, ..., client n corresponding to user n, where n is an integer greater than 0.

[0033] Each client in the client cluster may be an electronic device with communication function, including but not limited to: wearable device, handheld device, personal computer, tablet computer, vehicle-mounted device, smart phone, computing device or other processing device connected to a wireless modem, etc. Electronic devices may be called different names in different networks, such as: user equipment, access terminal, user unit, user station, mobile station, mobile station, remote station, remote terminal, mobile device, user terminal, terminal, wireless communication device, user agent or user device, cellular phone, cordless phone, personal digital assistant (PDA), electronic device in 5G network or future evolution network, etc.

[0034] The service platform 100 can be a separate server device, such as a rack-mounted, blade, tower, or cabinet-mounted server device, or a workstation, mainframe computer, or other hardware device with strong computing power; it can also be a server cluster composed of multiple servers, and the servers in the service cluster can be composed in a symmetrical manner, wherein each server has equivalent functions and status in the transaction link, and each server can provide services to the outside independently, and the independent service can be understood as not requiring the assistance of other servers.

[0035] In one or more embodiments of the present specification, the service platform 100 may establish a communication connection with at least one client in the client cluster, and complete data interaction during the delivery of target delivery content based on the communication connection, wherein the target delivery content is the delivery content obtained by the delivery content generation method provided in the embodiments of the present specification.

[0036] For example, the service platform 100 may determine a placement booth to be placed, obtain reference placement data for the placement booth, determine a placement tag set based on the reference placement data and the placement booth, call a placement text generation model based on the placement tag set and the reference placement data to perform placement text generation processing to obtain a placement text, perform image generation processing based on the placement tag set to obtain a placement image, and generate target placement content based on the placement tag set, the placement text, and the placement image;

[0037] Furthermore, after obtaining the target delivery content, the service platform 100 may send the target delivery content to any client in the client cluster, so as to deliver the target delivery content to the user through the client.

[0038] It should be noted that the service platform 100 and at least one client in the client cluster establish a communication connection through a network for interactive communication, wherein the network can be a wireless network or a wired network, the wireless network includes but is not limited to a cellular network, a wireless local area network, an infrared network or a Bluetooth network, and the wired network includes but is not limited to Ethernet, a universal serial bus (USB) or a controller local area network. In one or more embodiments of the specification, technologies and / or formats including Hyper Text Mark-up Language (HTML), Extensible Markup Language (XML), etc. are used to represent data (such as a target compressed package) exchanged through a network. In addition, conventional encryption technologies such as Secure Socket Layer (SSL), Transport Layer Security (TLS), Virtual Private Network (VPN), Internet Protocol Security (IPsec), etc. can also be used to encrypt all or some links. In other embodiments, customized and / or dedicated data communication technologies can also be used to replace or supplement the above data communication technologies.

[0039] In one or more embodiments of the specification, the execution subject of the delivery content generation method involved may be an electronic device, and the electronic device may be the above service platform 100. The specific implementation process of the delivery content generation method can be found in the following method embodiment, which will not be repeated here.

[0040] See also Figure 2 , is a flow chart of a method for generating delivery content provided in an embodiment of this specification. Figure 2 As shown, the method of the embodiment of this specification may include the following steps:

[0041] S202, determining a placement booth to be placed, obtaining reference placement information for the placement booth, and determining a placement tag set based on the reference placement information and the placement booth.

[0042] Among them, the placement booth refers to a specific space or location in the user interface for displaying and placing specific content.

[0043] The reference delivery materials refer to the production materials required to generate the delivery content to be displayed in the delivery booth. Specifically, the reference delivery materials may include at least one type of delivery materials among text type, picture type, and chart type.

[0044] A delivery tag set refers to a set of tags used to characterize various attributes of delivery content. Delivery content refers to delivery content displayed and placed at a delivery booth. The various attributes of delivery content may include attributes determined based on reference delivery materials and attributes determined based on the delivery booth. The delivery tag set can be used to generate delivery text, and the delivery text can be used to generate delivery content.

[0045] Specifically, the delivery tag set may include tags for representing the purpose of transaction delivery, the subject of transaction delivery, and the form of expression of transaction delivery. Among them, the purpose of transaction delivery refers to the purpose of the transaction to be achieved by delivering the delivery content in the delivery booth. For example, the purpose of transaction delivery may include but is not limited to conversion, activeness, etc. Conversion refers to, and activeness refers to, the tags used to represent the purpose of transaction delivery may include but are not limited to conversion tags, active tags, etc., and conversion tags may specifically include transaction tags, activation tags, etc., and active tags may specifically include subscription tags, browsing tags, comment tags, etc. The subject of transaction delivery refers to the subject of the transaction to be delivered through the delivery of content. For example, the subject of transaction delivery may include but is not limited to rights, products, professional content, services and other subjects. The labels used to characterize the subject of transaction delivery may include but are not limited to rights and interests labels, product labels, professional content labels, service labels, etc., and rights and interests labels may specifically include prize labels, coupon labels, task labels, activity labels, etc.; products may include various financial products, and product labels may include but are not limited to fund labels, stock labels, insurance labels, etc.; professional content may include information, live broadcast, professionally generated content (PGC for short), etc., and professional content labels may include but are not limited to information labels, live broadcast labels, professionally generated content labels; services may include financial management tools, investment advisory tools, simulation tools, content tools, etc., and service labels may include financial management tool labels, investment advisory tool labels, simulation tool labels, content tool labels, etc. The expression form of transaction delivery refers to the expression form of the delivery content. For example, the expression form of transaction delivery may include but is not limited to text, illustrations, layout templates, action points and the like. The tags used to characterize the expression form of transaction delivery may include but are not limited to text tags, illustration tags, layout template tags, action point tags and the like. Text tags may specifically include subject tags, user tags and macro tags. Illustration tags may include entity words. Layout template tags may include slot attribute tags, chart and animation tags and text data tags. Action points may include action tags, action adverb tags and action expression form tags.

[0046] In some embodiments, determining the placement booth to be placed may specifically include: determining the placement platform of the placement booth to be placed, determining the location information of the placement booth to be placed in the placement platform, and determining the booth form information of the placement booth to be placed. Among them, the placement platform may specifically include at least one of a website, a mobile application, and a social media; the location information may refer to the specific location of the placement booth on the placement platform, for example, in a specific area of ​​a specific page of the website, or, in a specific area of ​​a specific interface in the application, or, in a specific display area of ​​social media; the booth form information may include the area size occupied by the placement booth and the display state of the placement booth, for example, the area size is aa*bb, and the display state is dynamic display or static display. Specifically, a placement booth configuration request input by a user may be received, and the placement platform of the placement booth, the location information of the booth to be placed in the placement platform, and the booth form information of the placement booth may be determined based on the placement booth configuration request. It can be understood that the user can directly enter the placement booth configuration request in the electronic device, and the user can also enter the placement booth configuration request in the user terminal and the user terminal sends the placement booth configuration request to the electronic device.

[0047] Acquiring reference delivery materials for the placement booth may specifically include: acquiring reference delivery materials for the placement booth based on a placement booth configuration request for the placement booth. It is understandable that the placement booth configuration request may carry reference delivery materials for the placement booth; or, the placement booth configuration request may indicate obtaining a download link for the reference delivery materials, so as to obtain the download link from the placement booth configuration request, and obtain the reference delivery materials based on the download link.

[0048] Determining a set of delivery labels based on reference delivery materials and delivery booths may specifically include: obtaining a delivery label mapping relationship, and querying a delivery label set corresponding to the reference delivery materials and the delivery booth in the delivery label mapping relationship. Among them, the delivery label mapping relationship may include at least one mapping relationship between a reference delivery combination consisting of reference delivery materials and delivery booths and a reference delivery label set corresponding to the reference delivery combination. The reference delivery combination and the reference delivery label set corresponding to the reference delivery combination may be pre-configured based on expert experience.

[0049] S204, based on the delivery tag set and reference delivery data, a delivery text generation model is called to perform delivery text generation processing to obtain the delivery text.

[0050] Among them, the delivery text generation large model can be obtained after model training for delivery text generation scenarios based on the basic large model. The basic large model (Large Language Model, LLM for short) is an artificial intelligence content generation model designed to understand and generate human language. LLM is trained on a large amount of data and can perform a wide range of tasks, including text summarization, translation, sentiment analysis, etc. Optionally, the basic large model can adopt the Tongyi Qianwen large model, the Bailing large model, the GPT system large model, etc.

[0051] The placement text refers to the text in the placement content placed in the placement booth. For example, the placement booth is used to place a financial product, and the reference placement materials include relevant materials describing the financial product, and the placement text is the text describing the specific content of the financial product.

[0052] In some embodiments, a trained basic large model can be obtained, and the basic large model can be adapted to the delivery text generation scenario to obtain the delivery text generation large model. Specifically, the basic large model can be obtained first, and an initial delivery text generation large model for the delivery text generation scenario can be created, and sample data in the delivery text generation scenario can be obtained. Since the basic large model is usually a trained artificial intelligence generated content (AIGC) model with content generation capabilities, based on this, the initial delivery text generation large model can be adapted to the delivery text generation scenario. Specifically, the sample data in the delivery text generation scenario can be used to perform model fine-tuning training on the initial delivery text generation large model, and after the model fine-tuning training is completed, a delivery text generation large model adapted to the delivery text generation scenario can be obtained.

[0053] In some embodiments, the implementation process of step S204 may be: determining the target text delivery mode for the delivery tag set, determining the delivery text generation prompt words based on the target text delivery mode, reference delivery materials and delivery tag set, inputting the delivery text generation prompt words into the delivery text generation big model for delivery text generation processing, and obtaining the delivery text. For different target text delivery modes, the generated delivery text generation prompt words are also different. Among them, the target text delivery mode may include a text generation mode based on generated materials and a text generation mode based on no generated materials. The text generation mode based on generated materials refers to a text generation mode that provides production materials to the delivery text generation big model, and the text generation mode based on no production materials refers to a text generation mode that does not provide production materials to the delivery text generation big model.

[0054] S206, determining a delivery image based on the delivery tag set, and generating target delivery content based on the delivery tag set, the delivery text, and the delivery image.

[0055] The placement image refers to an image in the placement content placed in the placement booth. For example, the placement booth is used to place a financial product, the reference placement materials include relevant materials describing the financial product, and the placement image can be an image used to represent a certain feature of the financial product.

[0056] Target delivery content refers to the delivery content displayed at the delivery booth. The target delivery content can be composed of delivery text and delivery pictures.

[0057] In some embodiments, one way to determine the delivery image may be: determine the target entity word based on the delivery tag set, perform text-image processing on the target entity word, and obtain the delivery image; another way to determine the delivery image may be: perform image retrieval adaptation processing on the delivery tag set based on a preset image database to obtain the delivery image. Furthermore, the text layout information may be determined from the delivery tag set, and the delivery text and delivery image may be combined according to the text layout information to obtain the target delivery content.

[0058] The delivery content generation method provided in the embodiment of this specification first determines the delivery booth to be delivered and the reference delivery materials corresponding to the delivery booth, and determines the delivery tag set according to the reference delivery materials and the delivery booth, and then calls the delivery text generation model to generate the delivery text according to the delivery tag set and the reference delivery materials, and then determines the delivery image according to the delivery tag set, and then generates the target delivery content according to the delivery image, delivery text and delivery tag set. Therefore, by determining the delivery tag set, and then calling the delivery text generation model to generate the delivery text according to the delivery tag set and the reference delivery materials, the delivery text is automatically and intelligently generated, which improves the efficiency and convenience of generating the delivery text. Subsequently, the target delivery content is generated according to the delivery text and the delivery image, which realizes the automatic and intelligent generation of the target delivery content, thereby improving the efficiency and convenience of generating the target delivery content, and avoiding the problem of inefficiency caused by relying on manual experience to generate delivery content.

[0059] See also Figure 3 , is a flow chart of a method for generating delivery content provided in an embodiment of this specification. Figure 3 As shown, the method of the embodiment of this specification may include the following steps:

[0060] S302: Determine the booth to be deployed, and obtain reference deployment information for the booth.

[0061] For details, see Figure 2 The relevant description of step S202 in the illustrated embodiment will not be repeated here.

[0062] S304, obtaining a placement tag mapping relationship, and searching the placement tag set corresponding to the reference placement data and the placement booth in the placement tag mapping relationship.

[0063] In some embodiments, the delivery tag mapping relationship can be pre-configured and stored in a data table. After obtaining the reference delivery information and delivery booth, the delivery tag mapping relationship is obtained from the data table, and then the delivery tag set corresponding to the reference delivery information and delivery booth is queried in the delivery tag mapping relationship.

[0064] It is understandable that the delivery tag mapping relationship can be configured before executing step S304, or can be configured before executing step S302. Specifically, the process of configuring the delivery tag mapping relationship can refer to the following steps:

[0065] A2, obtaining historical placement booths and historical placement data corresponding to the historical placement booths, and obtaining a historical placement tag set for the historical placement data and the historical placement booths;

[0066] The historical placement booth refers to a specific space or location in a user interface that has been placed and used for displaying and placing specific content. The historical placement booth can be a specific space or location in a user interface that has been placed and used within a specified time period, and the specified time period can be within the past year, the past six months, the past three months, etc.

[0067] Historical delivery data may refer to historical production data required to generate delivery content that has been displayed at a historical delivery booth.

[0068] The historical delivery tag set may refer to a set of tags used to characterize various attributes of delivery content displayed at a historical delivery booth.

[0069] Since historical booths are booths that have been deployed and used, it is easy to obtain historical booths and historical deployment information from the database.

[0070] Regarding the acquisition method of the historical delivery label set, one method can be obtained by manual labeling, and the other method can be obtained by large model labeling. In the process of obtaining the historical delivery label set by manual labeling, the historical delivery label set for the historical delivery booth and the historical delivery data can be labeled according to the preset label tree. In the process of obtaining the historical delivery label set through the large model, the label labeling prompt words can be designed, and the label labeling prompt words can be input into the label generation large model for label labeling processing to obtain the historical delivery label set for the historical delivery booth and the historical delivery data, wherein the label labeling prompt words can include prompt words indicating the labeling label and the label constraint conditions, label tree knowledge, label set output samples, and label set input content. The prompt words indicating the labeling label and the label constraint conditions are used to indicate that the label generation large model performs label labeling processing based on the label tree knowledge, the label set output samples, and the label set input content to generate the historical delivery label set. The label tree knowledge includes the preset label tree, and the label set output sample is a sample of a historical delivery label set. The label set input content includes the historical delivery booth and the historical delivery data.

[0071] It can be understood that the above-mentioned preset label tree refers to a label tree constructed according to the purpose of business placement, the subject of business placement, and the expression form of business placement involved in the placement of booths. The preset label tree can specifically include labels for representing the purpose of business placement, the subject of business placement, and the expression form of business placement. For details, please refer to Figure 2 The description of the relevant part in step S202 in the illustrated embodiment will not be repeated here.

[0072] A4, determine the placement purpose corresponding to the historical placement booths, and determine the key indicators corresponding to the placement purpose;

[0073] The delivery purpose refers to the purpose of the content delivered in the historical delivery booth. The key indicator refers to the important indicator used to evaluate whether the delivery purpose has achieved the expected target value. For example, if the delivery purpose is conversion, the key indicator corresponding to conversion can be transaction amount.

[0074] Specifically, since historical booths are booths that have been used, the key indicators are real indicator data obtained from the content placed at the historical booths during the historical time period. The placement purposes and key indicators corresponding to the placement purposes of the historical booths can be queried from the transaction database. A large amount of relevant data on historical booths can be stored in the transaction database.

[0075] A6, based on the key indicators, adjust the historical delivery label set to obtain a reference delivery label set;

[0076] It is understandable that in the process of obtaining the historical delivery label set through manual labeling or large model labeling in step A2, since the labeling is not combined with the actual delivery effects corresponding to the historical delivery booths, and the historical delivery label set can be used as a reference for generating delivery content corresponding to the delivery booths in the future, the historical delivery label set may not be the delivery label set that can achieve the best delivery effect. Therefore, the historical delivery label set can be labeled and adjusted according to the key indicators corresponding to the historical delivery booths to obtain the adjusted delivery label set that can achieve the best delivery effect, that is, the reference delivery label set.

[0077] Specifically, executing step A6 specifically includes the following steps: determining the historical delivery effect data corresponding to each historical delivery label set based on the key indicators; determining a homogeneous delivery label set in all historical delivery label sets, and determining a first delivery label set and a second delivery label set in the homogeneous delivery label set, wherein the first historical delivery effect data of the first delivery label set is greater than the second historical delivery effect data of the second delivery label set; in all historical delivery label sets, performing label set adjustment processing on the second delivery label set based on the first delivery label set to obtain a reference delivery label set.

[0078] The historical delivery effect data refers to the data used to characterize the historical delivery effect achieved by the historical delivery booth corresponding to the historical delivery tag set. The homogeneous delivery tag set refers to the historical delivery tag set with some or all the same tags corresponding to the same historical delivery booth.

[0079] The method for determining the historical delivery effect data can be: based on the indicator data of the key indicators corresponding to the historical delivery booths, determine the historical delivery effect scores corresponding to the historical delivery booths, and use the historical delivery effect scores as the historical delivery effect data corresponding to the historical delivery label set, and the historical delivery label set corresponds to the historical delivery booths. It can be understood that for key indicators that evaluate delivery effects based on the numerical value of the indicator data, when the indicator data value of the key indicator is larger, the corresponding historical delivery effect score is also higher, and when the indicator data value of the key indicator is smaller, the corresponding historical delivery effect score is also lower. It can also be understood that for different key indicators, the judgment rules for judging the corresponding historical delivery effect data may be different.

[0080] The method for determining the reference delivery label set may be: among all historical delivery label sets, the second delivery label set is adjusted to be the same delivery label set as the first delivery label set, while other historical delivery label sets remain unchanged, to obtain the reference delivery label set. For example, all historical delivery label sets include set A, set B, set C, set D, set E, set F, and set G, set C and set F are homogeneous delivery label sets, set F is the first delivery label set, set C is the second delivery label set, set C is adjusted to be the same set as set F, and other sets remain unchanged, to obtain set A, set B, adjusted set C (the same as set F), set D, set E, set F, and set G as the reference delivery label set.

[0081] Therefore, by adjusting the historical delivery label set through the historical delivery effect data corresponding to the historical delivery label set, the historical delivery label set with poor historical delivery effects can be adjusted to a homogeneous historical delivery label set with good historical delivery effects, so that the delivery content corresponding to the delivery booth generated based on the adjusted delivery label set in the future can achieve better delivery effects.

[0082] A8, based on the reference delivery tag set, the reference delivery data corresponding to the reference delivery tag set, and the reference delivery booth, a delivery tag mapping relationship is established.

[0083] Specifically, since the reference delivery label set is obtained based on the historical delivery label set, the reference delivery data corresponding to the reference delivery label set is the historical delivery data corresponding to the historical delivery label set, and the reference delivery booth corresponding to the reference delivery label set is the historical delivery booth corresponding to the historical delivery label set. After obtaining the reference delivery label set, a delivery label mapping relationship can be established based on the reference delivery label set, the reference delivery data corresponding to the reference delivery label set, and the reference delivery booth corresponding to the reference delivery label set. In the delivery label mapping relationship, a reference delivery label set has a mapping relationship with its corresponding reference delivery booth and reference delivery data.

[0084] S306, determining the target text generation mode corresponding to the delivery tag set.

[0085] The target text generation mode may include a first text generation mode and a second text generation mode, wherein the first text generation mode refers to a text generation mode based on production materials, and the second text generation mode refers to a text generation mode based on no production materials. Optionally, the production materials may include a delivery subject landing page, delivery subject associated text, delivery subject database, and the like.

[0086] In some embodiments, the reference delivery materials may be subjected to production data detection processing. If it is detected that production data exists in the reference delivery materials, the target text generation mode corresponding to the delivery tag set is determined to be the first text generation mode. If it is detected that production data does not exist in the reference delivery materials, the target text generation mode corresponding to the delivery tag set is determined to be the second text generation mode. For example, when the reference delivery materials include the delivery subject landing page, it can be determined that production data exists in the reference delivery materials; for another example, when the reference delivery materials include access information to the delivery subject database, it can be determined that production data exists in the reference delivery materials; for another example, when the reference delivery materials include delivery subject-associated text or an access link to delivery subject-associated text, it can be determined that production data exists in the reference delivery materials.

[0087] In some other embodiments, a text generation mode selection operation input by the user can be obtained. If the text generation mode selection operation indicates a first text generation mode, the target text generation mode corresponding to the delivery tag set is determined to be the first text generation mode. If the text generation mode selection operation indicates a second text generation mode, the target text generation mode corresponding to the delivery tag set is determined to be the second text generation mode.

[0088] Therefore, by determining the target text generation mode, it is convenient to subsequently generate the delivery text according to the target text generation mode, which can improve the efficiency of generating the delivery text.

[0089] S308, based on the target text generation mode, reference delivery materials and delivery tag set, a delivery text generation model is called to perform delivery text generation processing to obtain delivery text.

[0090] In some embodiments, executing step S308 may specifically include the following steps:

[0091] B2, if the target text generation mode is the first text generation mode, then based on the reference delivery data and the delivery tag set, speech constraint content generation processing is performed to obtain speech generation prompt words, and the delivery text generation big model is called to perform speech generation processing based on the speech generation prompt words to obtain the target speech text;

[0092] Among them, the speech generation prompt words may include speech generation instructions, speech generation constraints, speech generation reference knowledge, and speech output samples. The speech generation instructions can be used to instruct the delivery text generation model to generate a target speech text. The speech generation constraints refer to the constraints that the delivery text generation model must follow to generate the target speech text. The speech generation reference knowledge is the reference knowledge that needs to be learned by the delivery text generation model to generate the target speech text. The speech output sample refers to an exemplary speech text.

[0093] When executing the step of generating speech constraint content based on reference delivery data and delivery tag set to obtain speech generation prompt words, it can specifically include: determining reference production data based on reference delivery data, determining speech generation reference knowledge based on reference production data and delivery tag set, and generating speech generation prompt words based on speech generation reference knowledge. Among them, reference production data can include but are not limited to delivery subject landing page, delivery subject associated text, delivery subject database. For different types of reference production data, the method of determining speech generation reference knowledge is also different.

[0094] Specifically, the method for determining the reference production data can be: detecting whether there is a delivery subject landing page in the reference delivery data, and if there is a delivery subject landing page, using the delivery subject landing page as the reference production data; if there is no delivery subject landing page, detecting whether there is delivery subject associated text in the reference delivery data, or detecting whether there is an access link to the delivery subject associated text in the reference delivery data, and if there is delivery subject associated text or an access link to the delivery subject associated text, using the delivery subject associated text as the reference production data, or using the delivery subject associated text obtained from the access link as the reference production data; if it is detected that there is no delivery subject associated text in the reference delivery data, and there is no access link to the delivery subject associated text, detecting whether there is access information to the delivery subject database in the reference delivery data; if there is access information to the delivery subject database, using the delivery subject database as the reference production data.

[0095] Specifically, based on the reference production data and the delivery tag set, the reference knowledge for speech generation is determined. The implementation method of this step can be: if the reference production data is the delivery subject landing page, then the indicator extraction processing is performed based on the reference production data to obtain the specified indicator data, the delivery tag summary prompt words are determined based on the specified indicator data and the delivery tag set, the delivery text generation large model is called to perform delivery tag summary processing based on the delivery tag summary prompt words to obtain the delivery tag summary text, and the delivery tag summary text is determined as the reference knowledge for speech generation;

[0096] If the reference production material is the text associated with the delivery subject, then the indicator extraction processing is performed based on the reference production material to obtain the specified indicator data, the delivery label summary prompt words are determined based on the specified indicator data and the delivery label set, the delivery text generation big model is called to perform delivery label summary processing based on the delivery label summary prompt words to obtain the delivery label summary text, the delivery label summary text is determined as the indicator extraction reference knowledge, the indicator extraction prompt words are determined based on the indicator extraction reference knowledge, and the delivery text generation big model is called to generate the speech generation reference knowledge based on the indicator extraction prompt words;

[0097] If the reference production material is the delivery subject database, then the associated indicator screening prompt words are determined based on the reference production material and the delivery tag set, and the delivery text generation large model is called to perform associated indicator screening processing based on the associated indicator screening prompt words to obtain the indicator screening result, and the indicator screening result is determined as the reference knowledge for speech generation.

[0098] It can be understood that the delivery subject landing page is a page that includes information related to the delivery subject. The delivery subject landing page can be understood as a page that jumps to the display after the user clicks on the promotion information. When the reference means of production are the delivery subject landing page, the indicator data of the dynamic indicators can be extracted from the delivery subject landing page to obtain the specified indicator data. The specified indicator data is the indicator data of the dynamic indicators. The dynamic indicators refer to the dynamic indicators related to the delivery subject. For example, taking the delivery subject as a financial product manager (such as a fund manager), the dynamic indicators can include performance, awards, and so on. Further, the label summary generation instruction can be determined, the label summary generation constraint conditions can be determined, the delivery label set can be used as the label summary generation reference knowledge, the specified indicator data can be used as the label summary input content, and the delivery label summary prompt words including the label summary generation instruction, the label summary generation constraint conditions, the label summary generation reference knowledge, and the label summary input content can be generated, wherein the label summary generation instruction can instruct the delivery text generation large model to generate the delivery label summary text based on the label summary generation constraint conditions, the label summary generation reference knowledge, and the label summary input content, the label summary generation constraint conditions are the constraints that the delivery text generation large model must follow to generate the delivery label summary text, the label summary generation reference knowledge is the reference knowledge that the delivery text generation large model must learn to generate the delivery label summary text, and the label summary input content is the input content required to obtain the output content (delivery label summary text). Further, the delivery label summary prompt words are input into the delivery text generation large model, so that the delivery text generation large model performs delivery label summary processing to obtain the delivery label summary text. Further, the delivery summary text output by the delivery text generation large model is used as the speech generation reference knowledge.

[0099] It is understandable that the text associated with the investment subject may refer to articles or information including information related to the investment subject. When the reference production materials are the text associated with the investment subject, the indicator data of the dynamic indicator can be extracted from the text associated with the investment subject to obtain the designated indicator data, and the designated indicator data is the indicator data of the dynamic indicator. The dynamic indicator refers to the dynamic indicator related to the investment subject. For example, taking the investment subject as a financial product manager (such as a fund manager), the dynamic indicators may include performance, awards, etc. Further, the label summary generation instruction can be determined, the label summary generation constraint conditions can be determined, the delivery label set can be used as the label summary generation reference knowledge, the specified indicator data can be used as the label summary input content, and the delivery label summary prompt words including the label summary generation instruction, the label summary generation constraint conditions, the label summary generation reference knowledge, and the label summary input content can be generated, wherein the label summary generation instruction can instruct the delivery text generation large model to generate the delivery label summary text based on the label summary generation constraint conditions, the label summary generation reference knowledge, and the label summary input content, the label summary generation constraint conditions are the constraints that the delivery text generation large model must follow to generate the delivery label summary text, the label summary generation reference knowledge is the reference knowledge that the delivery text generation large model must learn to generate the delivery label summary text, and the label summary input content is the input content required to obtain the output content (delivery label summary text). Further, the delivery label summary prompt words are input into the delivery text generation large model, so that the delivery text generation large model performs delivery label summary processing to obtain the delivery label summary text. Further, the delivery label summary text output by the delivery text generation model is determined as the index refinement input content, the index refinement instruction is determined, the delivery label set corresponding label mapping data in the indicator database is used as the index refinement reference knowledge, and the index refinement prompt words including the index refinement instruction, the index refinement reference knowledge, and the index refinement input content are generated, wherein the index refinement instruction is used to instruct the delivery text generation model to generate the speech reference knowledge based on the index refinement reference knowledge and the index refinement input content. Further, the indicator refinement prompt words are input into the delivery text generation model so that the delivery text generation model performs the indicator refinement processing to generate the speech generation reference knowledge.

[0100] For example, taking the investment subject as a financial product manager (such as a fund manager), please refer to Figure 4 The schematic diagram of a text of a delivery label summary prompt word is shown in FIG. The following is an explanation of the delivery label summary prompt word in a scenario where the reference production data is the delivery subject associated text: Figure 4As shown, "You are a financial assessment expert. Please evaluate the fund manager according to the following requirements based on the reported articles and existing knowledge descriptions, and give your summary" can generate instructions for the label summary; "1. Evaluate from several directions described in the given label knowledge base; 2. Each direction is 8-15 words; 3. Exaggeration and imprecise assumptions are not allowed" can generate constraints for the label summary; "--------" can represent a separator; "The existing knowledge description is as follows: Performance: Performance is ****, which can examine historical returns and future expectations, and can be compared with other fund managers or market averages; Peer evaluation: Peer evaluation is ****, in addition to the status in the professional field, awards can also be examined; Hot industries: Hot industries represent ****, which can examine whether the fund has historically or is currently managing the corresponding industry in the current market, and how the industry is performing" can generate reference knowledge for the label summary; "Then the reported articles are as follows: ******" can be used to input content for the label summary.

[0101] For example, taking the investment subject as a financial product manager (such as a fund manager), please refer to Figure 5 The following is a text diagram of an indicator extraction prompt word, and the indicator extraction prompt word in the scenario where the reference production materials are the associated texts of the delivery subject is explained: Figure 5 As shown, "You are an investment research expert. Based on the summary, please find the available key indicators and corresponding expressions within the given indicator knowledge base" is the indicator extraction instruction; "The summary is as follows: Performance: historical returns, losses of x funds in xxxx years, and a total loss of xxxxxxx yuan for managed funds. Evaluation by peers: awards won, management scale, market position. Popular industries: xxxx stock holding ratio, consumption recovery, platform economy, education, medical and other industry development" is the indicator extraction input content; "Indicator knowledge base: rate of return: ***; number of Golden Bull Awards: ***; Sharpe ratio: ***; management scale: ***; maximum drawdown: ***" is the indicator extraction reference knowledge; "--------" can represent a separator.

[0102] It is understandable that the delivery subject database can store indicator data related to the delivery subject. When the reference production material is the delivery subject database, the delivery subject database indicated by the reference production material is accessed, the existing indicator data associated with the delivery subject is extracted from the delivery subject database, the tag-related indicator is extracted from the delivery tag set, the related indicator screening instruction is determined, the related indicator screening constraint condition is determined, the tag-related indicator is used as the related indicator screening reference knowledge, the existing indicator data is used as the related indicator screening input content, and the related indicator screening prompt words including the related indicator screening instruction, the related indicator screening constraint condition, the related indicator screening reference knowledge, and the related indicator screening input content are generated; wherein, the related indicator screening instruction is used to instruct the delivery text generation large model to generate the indicator screening result based on the related indicator screening constraint condition, the related indicator screening reference knowledge, and the related indicator screening input content, the related indicator screening constraint condition refers to the constraint condition that the delivery text generation large model needs to follow to generate the indicator screening result, the related indicator screening reference knowledge refers to the reference knowledge that needs to be learned to generate the indicator screening result of the delivery text generation large model, and the related indicator screening input content refers to the input content required to obtain the output content (indicator screening result). Furthermore, the associated index screening prompt words are input into the delivery text generation model, so that the delivery text generation model performs the associated index screening process to obtain the index screening result. Furthermore, the index screening result is determined as the reference knowledge for speech generation.

[0103] Specifically, based on speech generation reference knowledge, speech generation prompt words are generated, and the implementation method thereof can be: determining speech generation instructions, determining speech generation constraints, determining speech examples, and generating speech generation prompt words including speech generation instructions, speech generation constraints, speech generation reference knowledge, and speech examples. Among them, the speech generation instructions are used to instruct the delivery text generation model to generate the target speech text based on the speech generation constraints, speech generation reference knowledge, and speech examples. The speech generation constraints refer to the constraints that the delivery text generation model needs to follow to generate the target speech text. The speech generation reference knowledge refers to the reference knowledge that the delivery text generation model needs to learn. The speech example is an exemplary speech text.

[0104] Therefore, in the scenario where there are reference delivery materials, by designing different series of prompt words for different types of materials, and then obtaining the target speech text through the series of prompt words, that is, instructing the delivery text big model to perform continuous text processing with strong correlation logic according to continuous prompt words one after another to obtain the target speech text, the delivery text big model can accurately generate the target speech text that matches the demand.

[0105] B4, screening layout elements based on the delivery tag set to obtain layout reference knowledge;

[0106] Specifically, the delivery tag set may include a delivery expression form tag, which may include a layout template tag, which may include a slot attribute tag and a text data tag. The layout template tag may be screened out from the delivery tag set, and the layout reference knowledge may be determined based on the text layout tag indicated by the layout template tag, the delivery subject tag associated with the layout template tag, and the delivery purpose tag associated with the layout template tag. Among them, the layout reference knowledge may include the number of text fillings for each slot (determined by the text layout tag), and the display purpose or subject of each slot (determined by the delivery subject tag associated with the layout template tag or the delivery purpose tag associated with the layout template tag). For example, taking the card placement booth as an example, the layout reference knowledge can be: Corner mark: 3-8 words are required to describe the purpose or positioning of this card; Main title: 6-12 words are required to express the most prominent features of this card; Description 1: 8-13 words are required to further express the features, and the sentences need to be fluent and have arguments; Description 2: 12-15 words are required to further supplement and express from other angles; Button: 3-5 words are required to indicate what the user can do next.

[0107] B6, generates layout adaptation prompt words based on the target speech text and layout reference knowledge, and calls the delivery text generation model to perform layout adaptation processing based on the layout adaptation prompt words to obtain the delivery text.

[0108] Specifically, the layout adaptation instruction can be determined, the layout adaptation constraint conditions can be determined, and the target speech text can be used as the reference knowledge for text generation to generate layout adaptation prompt words including the layout adaptation instruction, the layout adaptation constraint conditions, the text generation input content, and the layout reference knowledge generation, wherein the layout adaptation instruction is used to instruct the delivery text generation model to perform layout adaptation processing based on the layout adaptation constraint conditions, the text generation input knowledge, and the layout reference knowledge to obtain the delivery text, the layout adaptation constraint conditions refer to the constraint conditions that the delivery text generation model needs to follow to generate the delivery text, and the text generation input content refers to the input content required for the delivery text generation model to obtain the output content (delivery text). Furthermore, the layout adaptation prompt words can be input into the delivery text generation model so that the delivery text generation model generates the delivery text.

[0109] In some other embodiments, executing step S308 may specifically include the following steps:

[0110] C2: if the target text generation mode is the second text generation mode, then the layout elements are screened based on the delivery tag set to obtain layout reference knowledge;

[0111] It can be understood that the second text generation mode refers to a text generation mode based on no production materials, that is, in the second text generation mode, no production materials are provided to the delivery text generation model, so that the delivery text generation model automatically generates reference text content without production materials. Specifically, the delivery tag set may include tags of delivery expression forms, and the tags of delivery expression forms may include layout template tags, and the layout template tags may include slot attribute tags and text data tags. The layout template tags can be screened out from the delivery tag set, and the text layout information mapped by the layout template tags can be used as layout reference knowledge.

[0112] C4, based on the layout reference knowledge and reference delivery materials, determines the text content thinking prompt words, and calls the delivery text generation large model to generate reference text content based on the text content thinking prompt words;

[0113] Specifically, the text thinking instructions and text thinking constraints can be determined based on the reference delivery data, and text content thinking prompts including text thinking instructions, text thinking constraints, and layout reference knowledge can be generated, wherein the text thinking instructions can be used to instruct the delivery text generation model to generate reference text content based on the text thinking constraints and layout reference knowledge, the text thinking constraints refer to the constraints that the delivery text generation model must follow to generate the reference text content, and the layout reference knowledge refers to the layout knowledge that the delivery text generation model refers to when generating the reference text content. Furthermore, the text content thinking prompts can be input into the delivery text generation model to enable the delivery text generation model to generate reference text content.

[0114] C6, determining the rewriting and refining prompt words based on the reference text content, calling the delivery text generation large model to perform text rewriting and refining processing on the reference text content based on the rewriting and refining prompt words, and obtaining the delivery text.

[0115] Specifically, a rewrite and refinement instruction can be determined, and rewrite and refinement constraints can be determined. The reference text content can be used as the rewrite and refinement input content to generate a rewrite and refinement prompt word including the rewrite and refinement instruction, the rewrite and refinement constraints, and the rewrite and refinement input content. The rewrite and refinement instruction can instruct the delivery text generation big model to perform text rewrite and refinement processing on the rewrite and refinement input content based on the rewrite and refinement constraints to obtain the delivery text. The rewrite and refinement constraints are the constraints that the delivery text generation big model must follow to generate the delivery text. The rewrite and refinement input content is the input content required for the delivery text generation big model to obtain the output content (delivery text). Furthermore, the rewrite and refinement prompt word can be input into the delivery text generation big model so that the delivery text generation big model generates the delivery text.

[0116] S310, determining a target entity word based on the delivery tag set, performing text-based graph processing on the target entity word, and obtaining a delivery picture.

[0117] Among them, target entity words refer to words used to refer to specific things, places, people or concepts.

[0118] The displayed pictures refer to the pictures displayed at the display booth.

[0119] In some embodiments, entity word label screening processing can be performed on the delivery label set to obtain entity word labels and generate target entity words corresponding to the entity word labels. In the process of performing entity word label screening processing on the delivery label set, entity word labels with associated characteristics with the delivery text can be screened out with reference to the delivery text. Furthermore, the target entity word can be processed with a text graph large model to obtain a delivery picture.

[0120] S312, performing image retrieval and adaptation processing on the delivery tag set based on a preset image database to obtain delivery images.

[0121] The preset picture database may be a business-related picture database configured by professionals.

[0122] In some embodiments, entity word tags can be screened out from the delivery tag set, and delivery pictures that match the entity word tags can be retrieved from a preset picture database. In the process of screening out entity word tags from the delivery tag set, entity word tags that have associated characteristics with the delivery text can be screened out with reference to the delivery text.

[0123] It can be understood that step S310 and step S312 are two ways to obtain the delivery image. When executing the delivery content generation method provided in the embodiment of this specification, you can choose to execute any one of step S310 and step S312 to obtain the delivery image.

[0124] S314, determining the copy layout information based on the delivery tag set, and performing graphic-text combination processing on the delivery text and delivery image based on the copy layout information to obtain the target delivery content.

[0125] It can be understood that the delivery tag set may include delivery expression form tags, which may specifically include layout template tags and action point tags, layout template tags may include slot attribute tags, text data tags, icons and animation tags, and action point tags may include action tags and action expression form tags.

[0126] The copy layout information refers to the overall layout information including the placement text and the placement image. For example, the copy layout information may include the position of displaying the placement text in the placement booth, the position of displaying the placement image in the placement booth, and the like.

[0127] Specifically, layout template tags and action point tags can be extracted from the delivery tag set, and the copy layout information can be determined based on the layout template tags and the action point tags. Then, the delivery text and delivery image can be combined according to the copy layout information to obtain the target delivery content.

[0128] Optionally, after executing step S314, the target delivery content can also be delivered to the user at the delivery booth. After the delivery time of the target delivery content exceeds the preset observation time, the measured delivery effect data of the key indicators corresponding to the delivery booth within the preset observation time are determined, and the measured delivery effect improvement data corresponding to the delivery booth is determined; the estimated delivery effect improvement data corresponding to the delivery booth is obtained; if the difference between the measured delivery effect improvement data and the estimated delivery effect improvement data is less than or equal to the preset threshold, the delivery label set corresponding to the delivery booth and the reference delivery material is kept unchanged in the delivery label mapping relationship; if the difference between the measured delivery effect improvement data and the estimated delivery effect improvement data is greater than the preset threshold, the delivery label set corresponding to the delivery booth and the reference delivery material is subjected to label adjustment processing in the delivery label mapping relationship. Among them, the measured delivery effect improvement data is the difference between the measured delivery effect data and the historical delivery effect data corresponding to the delivery booth, and the estimated delivery effect improvement data is determined based on the first historical delivery effect data of the first delivery label set and the second historical delivery effect data of the second delivery label set in the homogeneous delivery label set corresponding to the delivery booth. Therefore, by implementing the delivery of the target delivery content corresponding to the delivery booth and monitoring the actual delivery effect data of the delivery booth within the preset observation time, the delivery effect of the delivery booth is evaluated. When there is a large gap between the actual delivery effect improvement data corresponding to the delivery booth and the estimated delivery effect improvement data, the delivery label set labels corresponding to the delivery booth and the reference delivery material in the delivery label mapping relationship are adjusted to ensure that the delivery content generated based on the adjusted delivery label set in the future can bring about an improvement in delivery effect. When the actual delivery effect improvement data corresponding to the delivery booth is close to the estimated delivery effect improvement data, it means that the delivery content generated based on the delivery label set corresponding to the delivery booth and the reference delivery material can already bring about an improvement in delivery effect. Therefore, the delivery label set corresponding to the delivery booth and the reference delivery material in the delivery label mapping relationship does not need to be updated.

[0129] In the delivery content generation method provided in the embodiment of the present specification, the historical delivery booth and the historical delivery data corresponding to the historical delivery booth can be obtained to obtain the historical delivery label set for the historical delivery data and the historical delivery booth, and then the delivery purpose corresponding to the historical delivery booth can be determined, and the key indicators corresponding to the delivery purpose can be determined. The historical delivery label set is subjected to label adjustment processing based on the key indicators to obtain a reference delivery label set, and then a delivery label mapping relationship is established based on the reference delivery label set, the reference delivery data corresponding to the reference delivery label set, and the reference delivery booth. Thus, the delivery label mapping relationship is pre-configured, so that in actual application, the delivery label set corresponding to the reference delivery data and the delivery booth can be directly queried in the delivery label mapping relationship, and the delivery label set can be determined quickly and accurately. Afterwards, the target text generation mode corresponding to the delivery label set is determined, and the delivery text generation large model is called based on the target text generation mode, the reference delivery data, and the delivery label set to perform delivery text generation processing to obtain the delivery text. Therefore, in different scenarios, the delivery text generation method suitable for the scenario can be flexibly selected. For example, in the scenario where the target text generation mode is a text generation mode based on production materials, that is, in the scenario where there is a reference delivery material, by designing different series prompt words for different types of materials, and then obtaining the target speech text through the series prompt words, that is, instructing the delivery text large model to perform continuous text processing with strong correlation logic according to the continuous prompt words one after another to obtain the target speech text, the delivery text large model can accurately generate the target speech text that matches the demand. Then, the target entity word is determined based on the delivery tag set, and the target entity word is processed by text-image processing to obtain the delivery picture, or the delivery tag set is processed by image retrieval adaptation processing based on the preset picture database to obtain the delivery picture, and then the text layout information is determined based on the delivery tag set, and the delivery text and delivery picture are combined based on the text layout information to obtain the target delivery content. Therefore, by matching the delivery picture, the delivery text and delivery picture generate rich target delivery content. In summary, the above method steps realize the automatic and intelligent generation of delivery text, improve the efficiency and convenience of generating delivery text, and subsequently generate target delivery content based on the delivery text and delivery image, realizing the automatic and intelligent generation of target delivery content, thereby improving the efficiency and convenience of generating target delivery content and avoiding the problem of inefficiency caused by relying on manual experience to generate delivery content.

[0130] The following will be combined Figure 6 , the delivery content generation device provided in the embodiment of this specification is introduced in detail. It should be noted that, Figure 6 The delivery content generation device shown is used to execute this instruction Figure 1 to Figure 6For the convenience of explanation, only the part related to the embodiment of this specification is shown. For the specific technical details not disclosed, please refer to this specification. Figure 1 to Figure 6 The embodiment shown.

[0131] See also Figure 6 , which shows a schematic diagram of the structure of the delivery content generation device of the embodiment of this specification. The delivery content generation device 1 can be implemented as all or part of the device through software, hardware or a combination of both. According to some embodiments, the delivery content generation device 1 includes a label determination module 11, a text generation module 12 and a content generation module 13, which are specifically used to:

[0132] The label determination module 11 is used to determine the placement booth to be placed, obtain reference placement information for the placement booth, and determine a placement label set based on the reference placement information and the placement booth;

[0133] A text generation module 12, configured to call a large model for generating delivery text based on the delivery tag set and the reference delivery data to perform delivery text generation processing to obtain delivery text;

[0134] The content generation module 13 is used to determine the delivery image based on the delivery tag set, and to generate target delivery content based on the delivery tag set, the delivery text and the delivery image.

[0135] Optionally, the label determination module 11 is specifically used for:

[0136] Get the delivery tag mapping relationship;

[0137] The placement tag set corresponding to the reference placement data and the placement booth is searched in the placement tag mapping relationship.

[0138] Optionally, the delivery content generating device 1 further includes:

[0139] The first processing module is used to obtain historical placement booths and historical placement data corresponding to the historical placement booths, and obtain a historical placement tag set for the historical placement data and the historical placement booths;

[0140] The second processing module is used to determine the placement purpose corresponding to the historical placement booth and determine the key indicators corresponding to the placement purpose;

[0141] A third processing module is used to perform label adjustment processing on the historical delivery label set based on the key indicator to obtain a reference delivery label set;

[0142] The fourth processing module is used to establish a delivery label mapping relationship based on the reference delivery label set, the reference delivery data corresponding to the reference delivery label set, and the reference delivery booth.

[0143] Optionally, the fourth processing module is specifically used for:

[0144] Determine the historical delivery effect data corresponding to each of the historical delivery tag sets based on the key indicators;

[0145] Determine a homogeneous delivery label set from all the historical delivery label sets, and determine a first delivery label set and a second delivery label set from the homogeneous delivery label set, wherein first historical delivery effect data of the first delivery label set is greater than second historical delivery effect data of the second delivery label set;

[0146] Among all the historical delivery label sets, label set adjustment processing is performed on the second delivery label set based on the first delivery label set to obtain a reference delivery label set.

[0147] Optionally, the text generation module 12 includes:

[0148] A first generating unit, used to determine a target text generation mode corresponding to the delivery tag set;

[0149] The second generation unit is used to call the delivery text generation model based on the target text generation mode, the reference delivery data and the delivery tag set to perform delivery text generation processing to obtain the delivery text.

[0150] Optionally, the second generation unit includes:

[0151] A speech text generation subunit is used for, if the target text generation mode is the first text generation mode, performing speech constraint content generation processing based on the reference delivery data and the delivery tag set to obtain speech generation prompt words, and calling the delivery text generation macro model to perform speech generation processing based on the speech generation prompt words to obtain the target speech text;

[0152] A layout knowledge generation subunit, used to perform layout element screening based on the delivery tag set to obtain layout reference knowledge;

[0153] The delivery text generation subunit is used to generate layout adaptation prompt words based on the target speech text and the layout reference knowledge, and call the delivery text generation model to perform layout adaptation processing based on the layout adaptation prompt words to obtain the delivery text.

[0154] Optional, the speech text generation subunit includes:

[0155] A prompt word generating unit is used to determine reference production data based on the reference delivery data, determine speech generation reference knowledge based on the reference production data and the delivery tag set, and generate speech generation prompt words based on the speech generation reference knowledge.

[0156] Optionally, the prompt word generating unit is specifically used for:

[0157] If the reference production material is a landing page of the delivery subject, then an index extraction process is performed based on the reference production material to obtain designated index data, a delivery label summary prompt word is determined based on the designated index data and the delivery label set, the delivery text generation large model is called to perform delivery label summary processing based on the delivery label summary prompt word to obtain a delivery label summary text, and the delivery label summary text is determined as a reference knowledge for speech generation;

[0158] If the reference production data is a text associated with a delivery subject, then an indicator extraction process is performed based on the reference production data to obtain designated indicator data, a delivery label summary prompt word is determined based on the designated indicator data and the delivery label set, the delivery text generation macromodel is called to perform delivery label summary processing based on the delivery label summary prompt word to obtain a delivery label summary text, the delivery label summary text is determined as indicator extraction reference knowledge, an indicator extraction prompt word is determined based on the indicator extraction reference knowledge, and the delivery text generation macromodel is called to generate speech generation reference knowledge based on the indicator extraction prompt word;

[0159] If the reference production material is a delivery subject database, then the associated indicator screening prompt words are determined based on the reference production material and the delivery tag set, and the delivery text generation large model is called to perform associated indicator screening processing based on the associated indicator screening prompt words to obtain indicator screening results, and the indicator screening results are determined as reference knowledge for speech generation.

[0160] Optionally, the second generating unit is specifically configured to:

[0161] If the target text generation mode is the second text generation mode, performing layout element screening processing based on the delivery tag set to obtain layout reference knowledge;

[0162] Determine text content thinking prompt words based on the layout reference knowledge and the reference delivery data, and call the delivery text generation model to generate reference text content based on the text content thinking prompt words;

[0163] Based on the reference text content, rewriting and refining prompt words are determined, and the delivery text generation model is called to perform text rewriting and refining processing on the reference text content based on the rewriting and refining prompt words to obtain the delivery text.

[0164] Optionally, the content generation module 13 is specifically used for:

[0165] Determine a target entity word based on the delivery tag set, perform text-based graph processing on the target entity word, and obtain a delivery picture; or,

[0166] Based on a preset image database, image retrieval and adaptation processing is performed on the delivery tag set to obtain delivery images.

[0167] Optionally, the content generation module 13 is specifically used for:

[0168] The text layout information is determined based on the delivery tag set, and the delivery text and the delivery image are combined based on the text layout information to obtain target delivery content.

[0169] Please refer to Figure 7 , which shows a schematic diagram of the structure of an electronic device provided by an exemplary embodiment of this specification. The electronic device in this specification may include one or more of the following components: a processor 110, a memory 120, an input device 130, an output device 140, and a bus 150. The processor 110, the memory 120, the input device 130, and the output device 140 may be connected via the bus 150.

[0170] The processor 110 may include one or more processing cores. The processor 110 uses various interfaces and lines to connect various parts of the entire terminal, and executes various functions and processes data of the terminal 100 by running or executing instructions, programs, code sets or instruction sets stored in the memory 120, and calling data stored in the memory 120. Optionally, the processor 110 can be implemented in at least one hardware form of digital signal processing (DSP), field-programmable gate array (FPGA), and programmable logic array (PLA). The processor 110 can integrate one or a combination of a central processing unit (CPU), a graphics processing unit (GPU), and a modem. Among them, the CPU mainly processes the operating system, user interface, and application programs; the GPU is responsible for rendering and drawing display content; and the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 110, but may be implemented separately through a communication chip.

[0171] The memory 120 may include a random access memory (RAM) or a read-only memory (ROM). Optionally, the memory 120 includes a non-transitory computer-readable storage medium. The memory 120 may be used to store instructions, programs, codes, code sets or instruction sets. The memory 120 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for implementing at least one function (e.g., a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the following various method embodiments, etc. The operating system may be an Android system, including a system deeply developed based on the Android system, an IOS system developed by Apple, including a system deeply developed based on the IOS system or other systems.

[0172] In order for the operating system to distinguish the specific application scenarios of third-party applications, it is necessary to open up data communication between third-party applications and the operating system so that the operating system can obtain the current scenario information of third-party applications at any time, and then perform targeted system resource adaptation based on the current scenario.

[0173] The input device 130 is used to receive input commands or data, and includes but is not limited to a keyboard, a mouse, a camera, a microphone, or a touch device. The output device 140 is used to output commands or data, and includes but is not limited to a display device and a speaker. In one example, the input device 130 and the output device 140 can be combined, and the input device 130 and the output device 140 are touch screen displays.

[0174] The touch display screen can be designed as a full screen, a curved screen or a special-shaped screen. The touch display screen can also be designed as a combination of a full screen and a curved screen, or a combination of a special-shaped screen and a curved screen, which is not limited in the embodiments of this specification.

[0175] In addition, those skilled in the art will appreciate that the structure of the electronic device shown in the above drawings does not constitute a limitation on the electronic device, and the electronic device may include more or fewer components than shown, or combine certain components, or arrange the components differently. For example, the electronic device also includes a radio frequency circuit, an input unit, a sensor, an audio circuit, a wireless fidelity (WiFi) module, a power supply, a Bluetooth module and other components, which will not be described in detail here.

[0176] exist Figure 7In the electronic device shown, the processor 110 can be used to call the program of the delivery content generation method stored in the memory 120, and specifically perform the following operations:

[0177] Determine a placement booth to be placed, obtain reference placement information for the placement booth, and determine a placement tag set based on the reference placement information and the placement booth;

[0178] Based on the delivery tag set and the reference delivery data, a delivery text generation model is called to perform delivery text generation processing to obtain a delivery text;

[0179] A delivery image is determined based on the delivery tag set, and target delivery content is generated based on the delivery tag set, the delivery text and the delivery image.

[0180] In one embodiment, when the processor 110 executes the step of determining the delivery tag set based on the reference delivery data and the delivery booth, the processor 110 specifically performs the following operations:

[0181] Get the delivery tag mapping relationship;

[0182] The placement tag set corresponding to the reference placement data and the placement booth is searched in the placement tag mapping relationship.

[0183] In one embodiment, before executing the acquisition of the delivery tag mapping relationship, the processor 110 further performs the following operations:

[0184] Obtain a historical placement booth and historical placement data corresponding to the historical placement booth, and obtain a historical placement tag set for the historical placement data and the historical placement booth;

[0185] Determine the placement purpose corresponding to the historical placement booth, and determine the key indicators corresponding to the placement purpose;

[0186] Performing label adjustment processing on the historical delivery label set based on the key indicator to obtain a reference delivery label set;

[0187] A delivery tag mapping relationship is established based on the reference delivery tag set, the reference delivery data corresponding to the reference delivery tag set, and the reference delivery booth.

[0188] In one embodiment, when the processor 110 performs the step of performing label adjustment processing on the historical delivery label set based on the key indicator to obtain a reference delivery label set, the processor 110 specifically performs the following operations:

[0189] Determine the historical delivery effect data corresponding to each of the historical delivery tag sets based on the key indicators;

[0190] Determine a homogeneous delivery label set from all the historical delivery label sets, and determine a first delivery label set and a second delivery label set from the homogeneous delivery label set, wherein first historical delivery effect data of the first delivery label set is greater than second historical delivery effect data of the second delivery label set;

[0191] Among all the historical delivery label sets, label set adjustment processing is performed on the second delivery label set based on the first delivery label set to obtain a reference delivery label set.

[0192] In one embodiment, when the processor 110 executes the step of calling the delivery text generation model based on the delivery tag set and the reference delivery data to perform delivery text generation processing to obtain the delivery text, the processor 110 specifically performs the following operations:

[0193] Determine a target text generation mode corresponding to the delivery tag set;

[0194] Based on the target text generation mode, the reference delivery data and the delivery tag set, a delivery text generation model is called to perform delivery text generation processing to obtain a delivery text.

[0195] In one embodiment, when the processor 110 executes the step of calling the delivery text generation model based on the target text generation mode, the reference delivery data and the delivery tag set to perform delivery text generation processing to obtain the delivery text, the following operations are specifically performed:

[0196] If the target text generation mode is the first text generation mode, a speech constraint content generation process is performed based on the reference delivery data and the delivery tag set to obtain speech generation prompt words, and a delivery text generation model is called to perform speech generation process based on the speech generation prompt words to obtain a target speech text;

[0197] Performing layout element screening based on the delivery tag set to obtain layout reference knowledge;

[0198] A layout adaptation prompt word is generated based on the target speech text and the layout reference knowledge, and the delivery text generation model is called to perform layout adaptation processing based on the layout adaptation prompt word to obtain the delivery text.

[0199] In one embodiment, when the processor 110 executes the step of performing speech constraint content generation processing based on the reference delivery data and the delivery tag set to obtain speech generation prompt words, the following operations are specifically performed:

[0200] Reference production data is determined based on the reference delivery data, speech generation reference knowledge is determined based on the reference production data and the delivery tag set, and speech generation prompt words are generated based on the speech generation reference knowledge.

[0201] In one embodiment, when the processor 110 executes the step of determining the speech generation reference knowledge based on the reference production data and the delivery tag set, the following operations are specifically performed:

[0202] If the reference production material is a landing page of the delivery subject, then an index extraction process is performed based on the reference production material to obtain designated index data, a delivery label summary prompt word is determined based on the designated index data and the delivery label set, the delivery text generation large model is called to perform delivery label summary processing based on the delivery label summary prompt word to obtain a delivery label summary text, and the delivery label summary text is determined as a reference knowledge for speech generation;

[0203] If the reference production data is a text associated with a delivery subject, then an indicator extraction process is performed based on the reference production data to obtain designated indicator data, a delivery label summary prompt word is determined based on the designated indicator data and the delivery label set, the delivery text generation macromodel is called to perform delivery label summary processing based on the delivery label summary prompt word to obtain a delivery label summary text, the delivery label summary text is determined as indicator extraction reference knowledge, an indicator extraction prompt word is determined based on the indicator extraction reference knowledge, and the delivery text generation macromodel is called to generate speech generation reference knowledge based on the indicator extraction prompt word;

[0204] If the reference production material is a delivery subject database, then the associated indicator screening prompt words are determined based on the reference production material and the delivery tag set, and the delivery text generation large model is called to perform associated indicator screening processing based on the associated indicator screening prompt words to obtain indicator screening results, and the indicator screening results are determined as reference knowledge for speech generation.

[0205] In one embodiment, when the processor 110 executes the step of calling the delivery text generation model based on the target text generation mode, the reference delivery data and the delivery tag set to perform delivery text generation processing to obtain the delivery text, the following operations are specifically performed:

[0206] If the target text generation mode is the second text generation mode, performing layout element screening processing based on the delivery tag set to obtain layout reference knowledge;

[0207] Determine text content thinking prompt words based on the layout reference knowledge and the reference delivery data, and call the delivery text generation model to generate reference text content based on the text content thinking prompt words;

[0208] Based on the reference text content, rewriting and refining prompt words are determined, and the delivery text generation model is called to perform text rewriting and refining processing on the reference text content based on the rewriting and refining prompt words to obtain the delivery text.

[0209] In one embodiment, when the processor 110 executes the step of determining the delivery image based on the delivery tag set, the processor 110 specifically performs the following operations:

[0210] Determine a target entity word based on the delivery tag set, perform text-based graph processing on the target entity word, and obtain a delivery picture; or,

[0211] Based on a preset image database, image retrieval and adaptation processing is performed on the delivery tag set to obtain delivery images.

[0212] In one embodiment, when the processor 110 executes the step of generating the target delivery content based on the delivery tag set, the delivery text, and the delivery image, the following operations are specifically performed:

[0213] The text layout information is determined based on the delivery tag set, and the delivery text and the delivery image are combined based on the text layout information to obtain target delivery content.

[0214] The embodiments of this specification also provide a computer program product, which stores at least one instruction, and the at least one instruction is loaded and executed by the processor to implement the delivery content generation device described in the above embodiments.

[0215] Those skilled in the art should be aware that in one or more of the above examples, the functions described in the embodiments of this specification can be implemented using hardware, software, firmware, or any combination thereof. When implemented using software, these functions can be stored in a computer-readable medium or transmitted as one or more instructions or codes on a computer-readable medium. Computer-readable media include computer storage media and communication media, wherein communication media include any media that facilitates the transmission of computer programs from one place to another. Storage media can be any available media that can be accessed by general or special-purpose computers.

[0216] The above description is only an optional embodiment of this specification and is not intended to limit this specification. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of this specification should be included in the protection scope of this specification.

[0217] The above is a description of a specific embodiment of the specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A method for generating delivery content, the method comprising: Determine a placement booth to be placed, obtain reference placement information for the placement booth, and determine a placement tag set based on the reference placement information and the placement booth; Based on the delivery tag set and the reference delivery data, a delivery text generation model is called to perform delivery text generation processing to obtain a delivery text; A delivery image is determined based on the delivery tag set, and target delivery content is generated based on the delivery tag set, the delivery text and the delivery image.

2. The method according to claim 1, wherein determining the delivery tag set based on the reference delivery data and the delivery booth comprises: Get the delivery tag mapping relationship; The placement tag set corresponding to the reference placement data and the placement booth is searched in the placement tag mapping relationship.

3. The method according to claim 2, before obtaining the delivery tag mapping relationship, further comprising: Obtain a historical placement booth and historical placement data corresponding to the historical placement booth, and obtain a historical placement tag set for the historical placement data and the historical placement booth; Determine the placement purpose corresponding to the historical placement booth, and determine the key indicators corresponding to the placement purpose; Performing label adjustment processing on the historical delivery label set based on the key indicator to obtain a reference delivery label set; A delivery tag mapping relationship is established based on the reference delivery tag set, the reference delivery data corresponding to the reference delivery tag set, and the reference delivery booth.

4. The method according to claim 3, wherein the step of performing label adjustment processing on the historical delivery label set based on the key indicator to obtain a reference delivery label set comprises: Determine the historical delivery effect data corresponding to each of the historical delivery tag sets based on the key indicators; Determine a homogeneous delivery label set from all the historical delivery label sets, and determine a first delivery label set and a second delivery label set from the homogeneous delivery label set, wherein first historical delivery effect data of the first delivery label set is greater than second historical delivery effect data of the second delivery label set; Among all the historical delivery label sets, label set adjustment processing is performed on the second delivery label set based on the first delivery label set to obtain a reference delivery label set.

5. The method according to claim 1, wherein the step of calling a large model for generating delivery text based on the delivery tag set and the reference delivery data to generate delivery text to obtain delivery text comprises: Determine a target text generation mode corresponding to the delivery tag set; Based on the target text generation mode, the reference delivery data and the delivery tag set, a delivery text generation model is called to perform delivery text generation processing to obtain a delivery text.

6. The method according to claim 5, wherein the step of calling a large delivery text generation model based on the target text generation mode, the reference delivery data and the delivery tag set to perform delivery text generation processing to obtain the delivery text comprises: If the target text generation mode is the first text generation mode, a speech constraint content generation process is performed based on the reference delivery data and the delivery tag set to obtain speech generation prompt words, and a delivery text generation model is called to perform speech generation process based on the speech generation prompt words to obtain a target speech text; Performing layout element screening based on the delivery tag set to obtain layout reference knowledge; A layout adaptation prompt word is generated based on the target speech text and the layout reference knowledge, and the delivery text generation model is called to perform layout adaptation processing based on the layout adaptation prompt word to obtain the delivery text.

7. According to the method of claim 6, the step of generating speech constraint content based on the reference delivery data and the delivery tag set to obtain speech generation prompt words comprises: Reference production data is determined based on the reference delivery data, speech generation reference knowledge is determined based on the reference production data and the delivery tag set, and speech generation prompt words are generated based on the speech generation reference knowledge.

8. The method according to claim 7, wherein determining the speech generation reference knowledge based on the reference production data and the delivery tag set comprises: If the reference production material is a landing page of the delivery subject, then an index extraction process is performed based on the reference production material to obtain designated index data, a delivery label summary prompt word is determined based on the designated index data and the delivery label set, the delivery text generation large model is called to perform delivery label summary processing based on the delivery label summary prompt word to obtain a delivery label summary text, and the delivery label summary text is determined as a reference knowledge for speech generation; If the reference production data is a text associated with a delivery subject, then an indicator extraction process is performed based on the reference production data to obtain designated indicator data, a delivery label summary prompt word is determined based on the designated indicator data and the delivery label set, the delivery text generation macromodel is called to perform delivery label summary processing based on the delivery label summary prompt word to obtain a delivery label summary text, the delivery label summary text is determined as indicator extraction reference knowledge, an indicator extraction prompt word is determined based on the indicator extraction reference knowledge, and the delivery text generation macromodel is called to generate speech generation reference knowledge based on the indicator extraction prompt word; If the reference production material is a delivery subject database, then the associated indicator screening prompt words are determined based on the reference production material and the delivery tag set, and the delivery text generation large model is called to perform associated indicator screening processing based on the associated indicator screening prompt words to obtain indicator screening results, and the indicator screening results are determined as reference knowledge for speech generation.

9. The method according to claim 5, wherein the step of calling a large delivery text generation model based on the target text generation mode, the reference delivery data and the delivery tag set to perform delivery text generation processing to obtain the delivery text comprises: If the target text generation mode is the second text generation mode, performing layout element screening processing based on the delivery tag set to obtain layout reference knowledge; Determine text content thinking prompt words based on the layout reference knowledge and the reference delivery data, and call the delivery text generation model to generate reference text content based on the text content thinking prompt words; Based on the reference text content, rewriting and refining prompt words are determined, and the delivery text generation model is called to perform text rewriting and refining processing on the reference text content based on the rewriting and refining prompt words to obtain the delivery text.

10. The method according to claim 1, wherein determining the delivery image based on the delivery tag set comprises: Determine a target entity word based on the delivery tag set, perform text-based graph processing on the target entity word, and obtain a delivery picture; or, Based on a preset image database, image retrieval and adaptation processing is performed on the delivery tag set to obtain delivery images.

11. The method according to claim 1, wherein generating target delivery content based on the delivery tag set, the delivery text and the delivery image comprises: The text layout information is determined based on the delivery tag set, and the delivery text and the delivery image are combined based on the text layout information to obtain target delivery content.

12. A delivery content generation device, the device comprising: A label determination module is used to determine a placement booth to be placed, obtain reference placement information for the placement booth, and determine a placement label set based on the reference placement information and the placement booth; A text generation module, used to call a large model for generating delivery text based on the delivery tag set and the reference delivery data to perform delivery text generation processing to obtain delivery text; The content generation module is used to determine the delivery image based on the delivery tag set, and generate target delivery content based on the delivery tag set, the delivery text and the delivery image.

13. A computer storage medium storing a plurality of instructions, wherein the instructions are suitable for being loaded by a processor and executing the method according to any one of claims 1 to 11.

14. A computer program product, wherein the computer program product stores at least one instruction, wherein the at least one instruction is loaded by a processor and executes the method according to any one of claims 1 to 11.

15. An electronic device, comprising: A processor and a memory; wherein the memory stores a computer program, and the computer program is suitable for being loaded by the processor and executing the method according to any one of claims 1 to 11.

Images