Text generation method and electronic equipment

Abstract

The embodiment of the invention discloses a text generation method and electronic equipment. Variable parameters input by a user are obtained, the variable parameters comprise vehicle detail data, a middle paragraph template is generated according to the variable parameters and an initial paragraph template, a theme creation request sent by a client is received, and a first response is sent to the client so that the client can display a theme creation interface. And receiving theme creation confirmation information sent by the client, generating a target text according to the theme creation confirmation information and the middle paragraph template, and sending the target text to the client, so that the client displays the target text through a copywriting display interface. Therefore, through template splicing and verbal skill generalization, the quality and the manuscript generation efficiency of an oral broadcast manuscript are improved, and diversified oral broadcast manuscript requirements are met.

Description

Technical Field

[0001] The present invention relates to the field of text generation, and particularly to a text generation method and an electronic device. Background Art

[0002] At present, with the rapid development of the live broadcast industry, the efficient generation of live broadcast oral script has become increasingly important. Currently, it is a common method to use a large model to generate the oral script of a live broadcast room in a specific industry. In the prior art, mainly by means of an open-source framework, the text and data in a specific industry are provided to the large model as a knowledge base to achieve generation. Or, due to the limitations of the computing power and generation length of the large model, in order to obtain longer content, most large models need to splice several generated texts. However, these existing solutions have obvious defects. On the one hand, there are limitations on the number of words in the input and output texts of the large model. Using the form of knowledge base mounting will seriously damage the quality and diversity of the generated content. On the other hand, due to the limited length of the content output by the large model, the spliced long texts often have poor context logic and low text generation efficiency. In addition, the content generated by the large model is highly mechanical, with a single speech pattern, and poor readability and diversity. This makes the existing large models unable to meet the needs of the live broadcast industry for high-quality and diverse oral scripts when generating the oral script of a live broadcast room in a specific industry. Summary of the Invention

[0003] In view of this, embodiments of the present invention provide a text generation method and an electronic device, which can improve the quality of the oral script and the efficiency of script generation, and meet the diverse needs of oral scripts.

[0004] In a first aspect, an embodiment of the present invention provides a text generation method, which is applicable to a server, and the method includes:

[0005] Obtain variable parameters input by a user, where the variable parameters include vehicle detail data;

[0006] Generate an intermediate paragraph template according to the variable parameters and an initial paragraph template;

[0007] Receive a theme creation request sent by a client;

[0008] Send a first response to the client to enable the client to display a theme creation interface, where the first response includes candidate copy templates;

[0009] Receive the theme creation confirmation information sent by the client, where the theme creation confirmation information includes at least one of a platform name, a mode type, a theme content type, a generation model, a copy name, and a target copy template;

[0010] Generate a target text according to the theme creation confirmation information and the intermediate paragraph template;

[0011] Send the target text to the client so that the client displays the target text through the copy display interface.

[0012] In some embodiments, the method further includes:

[0013] Classify and store general manuscript sentences according to a preset classification, where the preset classification includes at least one of text content, usage, and theme;

[0014] Generate synonymous derivative sentences based on the general manuscript sentences through a large language model;

[0015] Generate an initial paragraph template based on the synonymous derivative sentences;

[0016] Generate an initial copy template based on the initial paragraph template.

[0017] In some embodiments, the generating an initial paragraph template based on the synonymous derivative sentences includes:

[0018] Store the synonymous derivative sentences according to the preset classification;

[0019] Identify the variable parameters of the synonymous derivative sentences, where the variable parameters include car name, parameter configuration, price, and price reduction amount;

[0020] Set variable slots on the synonymous derivative sentences according to the variable parameters to generate derivative sentence patterns;

[0021] Generate an initial paragraph template according to the derivative sentence patterns.

[0022] In some embodiments, the generating an initial paragraph template according to the derivative sentence patterns includes:

[0023] Receive a paragraph creation request sent by the client;

[0024] Send a second response to the client so that the client displays a paragraph creation interface, where the second response includes the derivative sentence patterns;

[0025] Receive the new paragraph confirmation information sent by the client, where the new paragraph confirmation information includes at least one of paragraph content type, paragraph template type, paragraph template name, and paragraph template structure, and the paragraph template structure includes manuscript type and the derivative sentence pattern selected by the user;

[0026] Generate an initial paragraph template according to the new paragraph confirmation information.

[0027] In some embodiments, the generating an initial copy template based on the initial paragraph template includes:

[0028] Receive a copy creation request sent by the client;

[0029] Send a third response to the client to enable the client to display a copywriting creation interface, where the third response includes the initial paragraph template;

[0030] Receive the new copywriting confirmation information sent by the client, where the new copywriting confirmation information includes at least one of the copywriting content type, copywriting mode, mode name, and template structure, and the template structure includes at least one of the paragraph content type, paragraph template type, and paragraph template name;

[0031] Generate an initial copywriting template according to the new copywriting confirmation information.

[0032] In some embodiments, the generating the target text according to the theme creation confirmation information and the intermediate paragraph template includes:

[0033] Determine a target large language model according to the generation model;

[0034] Generate a target paragraph template through the target large language model according to the intermediate paragraph template;

[0035] Generate the target text through the target copywriting template according to the target paragraph template.

[0036] In a second aspect, an embodiment of the present invention provides a text generation method applicable to a client, and the method includes:

[0037] Obtain the variable parameters input by the user and send them to the server to enable the server to generate an intermediate paragraph template according to the variable parameters and the initial paragraph template, where the variable parameters include vehicle detail data;

[0038] Send a theme creation request to the server;

[0039] Receive a first response sent by the server, where the first response includes a candidate copywriting template;

[0040] Display a theme creation interface according to the first response, where the theme creation interface includes at least one of a mode selection area, a model setting area, a text generation area, and a first confirmation control, the mode selection area includes a platform selection control, a mode type control, a text type control, and a copywriting template selection control, the model setting area includes a generation model selection control and a word count setting control, and the text generation area includes a generation method selection control and a copywriting naming control;

[0041] In response to the triggering of the first confirmation control, send theme creation confirmation information to the server, so that the server generates a target text based on the theme creation confirmation information and the intermediate paragraph template, where the theme creation confirmation information includes at least one of a platform name, a mode type, a theme content type, a generation model, a copywriting name, and a target copywriting template;

[0042] Display the target text through a copywriting display interface.

[0043] In some embodiments, the method further includes:

[0044] Send a paragraph creation request to the server;

[0045] Receive a second response sent by the server, where the second response includes derivative sentence patterns;

[0046] Display a paragraph creation interface according to the second response, where the paragraph creation interface includes at least one of a first new selection area, a paragraph content definition area, and a paragraph template setting area and a second confirmation control. The first new selection area includes a paragraph new blank control and a copy existing paragraph control. The paragraph content definition area includes a paragraph content selection control, a paragraph template type control, and a paragraph template name control. The template setting area includes a paragraph type selection control and a sentence pattern label selection control;

[0047] In response to the triggering of the second confirmation control, send new paragraph confirmation information to the server, so that the server generates an initial paragraph template according to the new paragraph confirmation information. The new paragraph confirmation information includes at least one of a paragraph content type, a paragraph template type, a paragraph template name, and a paragraph template structure. The paragraph template structure includes a manuscript type and a derivative sentence pattern selected by the user.

[0048] In some embodiments, the method further includes:

[0049] Send a copywriting creation request to the server;

[0050] Receive a third response sent by the server, where the third response includes an initial paragraph template;

[0051] Display a copywriting creation interface according to the third response, where the copywriting creation interface includes at least one of a second new selection area, a copywriting content definition area, and a copywriting template setting area and a third confirmation control. The second new selection area includes a copywriting new blank control and a copy existing copywriting control. The copywriting content definition area includes a copywriting content selection control, a copywriting template type control, and a copywriting template name control. The copywriting template setting area includes a paragraph template selection control, a paragraph template type control, and a paragraph template name selection control;

[0052] In response to the triggering of the third confirmation control, send new copy confirmation information to the server, so that the server generates an initial copy template according to the new copy confirmation information. The new copy confirmation information includes at least one of copy content type, copy mode, mode name, and template structure. The template structure includes at least one of paragraph content type, paragraph template type, and paragraph template name.

[0053] In a third aspect, an embodiment of the present invention provides an electronic device, including a memory and a processor. The memory is used to store one or more computer program instructions. Among them, the one or more computer program instructions are executed by the processor to implement the method described in the first aspect.

[0054] The technical solution of the embodiment of the present invention obtains variable parameters input by the user. The variable parameters include vehicle detail data, generates an intermediate paragraph template according to the variable parameters and the initial paragraph template, receives a theme creation request sent by the client, sends a first response to the client, so that the client displays a theme creation interface, receives the theme creation confirmation information sent by the client, generates a target text according to the theme creation confirmation information and the intermediate paragraph template, and sends the target text to the client, so that the client displays the target text through a copy display interface. Thus, through template splicing and speech generalization, the quality of the voiceover manuscript and the manuscript generation efficiency are improved, meeting the diverse voiceover manuscript requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Through the following description of the embodiments of the present invention with reference to the drawings, the above and other objects, features, and advantages of the present invention will become clearer. In the drawings:

[0056] Figure 1 is a schematic diagram of a text generation system according to an embodiment of the present invention;

[0057] Figure 2 is an information interaction diagram of a text generation system according to an embodiment of the present invention for obtaining an initial paragraph template;

[0058] Figure 3 is a schematic diagram of a paragraph creation interface according to an embodiment of the present invention;

[0059] Figure 4 is an information interaction diagram of a text generation system according to an embodiment of the present invention for obtaining an initial copy template;

[0060] Figure 5 is a schematic diagram of a copy creation interface according to an embodiment of the present invention;

[0061] Figure 6 is an information interaction diagram of a text generation system according to an embodiment of the present invention for obtaining a target text;

[0062] Figure 7 is a schematic diagram of the theme creation interface of an embodiment of the present invention;

[0063] Figure 8 is a flowchart of the text generation method of an embodiment of the present invention;

[0064] Figure 9 is a schematic diagram of the text generation device of an embodiment of the present invention;

[0065] Figure 10 is a schematic diagram of the electronic device of an embodiment of the present invention. Detailed implementation manners

[0066] The following describes the present application based on embodiments, but the present application is not limited to these embodiments. In the following detailed description of the present application, some specific details are described in detail. Those skilled in the art can fully understand the present application without the description of these details. In order to avoid obscuring the essence of the present application, well-known methods, processes, procedures, components, and circuits are not described in detail.

[0067] In addition, those of ordinary skill in the art should understand that the drawings provided herein are for illustrative purposes only, and the drawings are not necessarily drawn to scale.

[0068] Unless the context clearly requires otherwise, the words such as "including" and "comprising" in the entire application document should be interpreted as the meaning of including rather than exclusive or exhaustive; that is, the meaning of "including but not limited to".

[0069] In the description of the present application, 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 addition, in the description of the present application, unless otherwise specified, the meaning of "a plurality of" is two or more.

[0070] Figure 1 is a schematic diagram of the text generation system of an embodiment of the present invention. In Figure 1 the illustrated embodiment, the text generation system includes a client 11, a network 12, and a server 13. The server 13 obtains the variable parameters, theme creation requests, and theme creation confirmation information input by the user from the client 11 through the network 12, and the client 11 obtains a first response from the server 13 through the network 12, and the first response includes candidate copywriting templates.

[0071] Specifically, the client 11 can be various types of devices or applications. For example, it can be specialized software running on a personal computer, providing users with an intuitive graphical user interface. Users can conveniently perform various operations on the desktop, such as inputting variable parameters, selecting various modes and types, etc.; or it can be a web-based application accessed through a browser. Users do not need to install additional software. As long as they have a network connection and a browser, they can use this service on any device; or it can be an application installed on a smartphone or tablet. Users can generate live broadcast script manuscripts at any time and place.

[0072] Among them, in the paragraph creation stage, the client 11 sends a paragraph creation request to the server 13. After receiving the request, the server 13 returns a second response containing derivative sentence patterns. The client 11 displays a paragraph creation interface according to the second response. When the user triggers the second confirmation control, the client 11 sends new paragraph confirmation information to the server 13, including information such as paragraph content type, paragraph template type, paragraph template name, and paragraph template structure. The server 13 generates an initial paragraph template based on this information.

[0073] In the copywriting creation stage, the client 11 sends a copywriting creation request to the server 13. After receiving the request, the server 13 returns a third response containing the initial paragraph template. The client 11 displays a copywriting creation interface according to the third response. When the user triggers the third confirmation control, the client 11 sends new copywriting confirmation information to the server 13, including information such as copywriting content type, copywriting mode, mode name, and template structure. The server 13 generates an initial copywriting template based on this information.

[0074] In the text generation stage, the client 11 obtains the variable parameters input by the user, which contain vehicle details data, and then sends these parameters to the server 13. After receiving the parameters, the server 13 generates an intermediate paragraph template based on these parameters and the initial paragraph template. The client 11 sends a theme creation request to the server 13. After receiving the request, the server 13 returns a first response containing candidate copywriting templates. The client 11 displays a theme creation interface according to the first response. When the user triggers the first confirmation control, the client 11 sends theme creation confirmation information to the server 13, including information such as platform name, mode type, theme content type, generation model, copywriting name, and target copywriting template. The server 13 generates the target text based on this information and the intermediate paragraph template, and returns the target text to the client 11. After receiving the target text returned by the server 13, the client 11 displays it to the user through the copywriting display interface.

[0075] Among them, during the initial copywriting template generation phase, server 13 waits for client 11 to send a copywriting creation request. After receiving the request, it prepares to provide a relevant response. Server 13 sends a third response to client 11, which contains the initial paragraph template. After client 11 receives it, it displays a copywriting creation interface for the user to perform copywriting creation setting operations. Server 13 receives the new copywriting confirmation information sent by client 11, which includes important information such as the copywriting content type, copywriting mode, mode name, and template structure (including paragraph content type, paragraph template type, and paragraph template name, etc.). Server 13 generates an initial copywriting template based on the received new copywriting confirmation information for subsequent text generation work.

[0076] During the initial paragraph template generation phase, server 13 classifies and stores general manuscript sentences according to preset categories (such as text content, purpose, and theme, etc.) to build a basic manuscript sentence library. Server 13 uses a large language model to generate synonymous derivative sentences based on the stored general manuscript sentences to enrich the sentence resources, and stores the generated synonymous derivative sentences according to the preset categories for subsequent search and use. Server 13 identifies the variable parameters (such as car names, parameter configurations, etc.) in the synonymous derivative sentences and sets variable slots on these sentences to generate derivative sentence patterns. Server 13 waits for client 11 to send a paragraph creation request. When it receives this request, it proceeds to the next step. Server 13 sends a second response to client 11, which contains the derivative sentence patterns. After client 11 receives it, it will display a paragraph creation interface for the user to perform paragraph creation related settings. Server 13 receives the new paragraph confirmation information sent by client 11, which includes key information such as the paragraph content type, paragraph template type, paragraph template name, and paragraph template structure (including manuscript type and the derivative sentence pattern selected by the user). Server 13 generates an initial paragraph template based on the received new paragraph confirmation information to provide a basis for subsequent generation of the initial copywriting template.

[0077] In the text generation stage, server 13 obtains the variable parameters input by the user from client 11, which include vehicle details data. These data are important bases for generating the real-time live broadcast script subsequently. Server 13 generates an intermediate paragraph template according to the obtained variable parameters and the existing initial paragraph template. This step involves processing and integrating the input data and the existing template to prepare for generating the target text. Server 13 waits for client 11 to send a theme creation request. After receiving this request, it proceeds to the next operation. Server 13 sends a first response to client 11, which includes candidate copy templates. After receiving it, client 11 will display a theme creation interface according to this response, providing an entry for the user to select relevant parameters. Server 13 receives the theme creation confirmation information sent by client 11, which includes at least one key information such as platform name, mode type, theme content type, generation model, copy name, and target copy template. These information further clarify the specific requirements of the user for the generated text. Server 13 generates the target text according to the received theme creation confirmation information and the previously generated intermediate paragraph template. The specific process includes determining the target large language model according to the generation model, generating a target paragraph template based on the intermediate paragraph template through the target large language model, and then generating the final target text according to the target copy template based on the target paragraph template. Server 13 sends the generated target text to client 11, and client 11 will display it to the user through the copy display interface after receiving it.

[0078] Among them, the server 13 can be a single server or a server cluster composed of multiple servers.

[0079] The network 12 can be implemented through wired or wireless networks to ensure that client 11 can effectively establish a connection with server 13 and perform data interaction.

[0080] In the embodiment of the present invention, by obtaining the variable parameters input by the user, the variable parameters include vehicle details data, generating an intermediate paragraph template according to the variable parameters and the initial paragraph template, receiving the theme creation request sent by the client, sending a first response to the client to enable the client to display the theme creation interface, receiving the theme creation confirmation information sent by the client, generating the target text according to the theme creation confirmation information and the intermediate paragraph template, and sending the target text to the client to enable the client to display the target text through the copy display interface. Thus, through template splicing and speech generalization, the quality and generation efficiency of the broadcast script are improved, and the diverse needs of the broadcast script are met.

[0081] Figure 2 It is the information interaction diagram for the text generation system of the embodiment of the present invention to obtain the initial paragraph template. Figure 2The flowchart shows the information interaction process in which the player establishes a connection with the server and obtains the initial paragraph template during the text preparation stage, which specifically includes the following steps:

[0082] Step S201: Classify and store general manuscript sentences according to preset categories.

[0083] Specifically, the server first determines the preset classification types. For example, according to the text content, it can be divided into categories such as automotive performance description, appearance introduction, interior explanation, etc.; according to the usage, it can be divided into categories such as opening guidance, product promotion, interactive session, etc.; according to the theme, it can be divided into categories such as new car release, promotion activity, vehicle model comparison, etc. For each general manuscript sentence, the server analyzes its semantic, keyword and other features through natural language processing technology. For example, for the sentence "This car is equipped with a high-performance engine and has strong power", by analyzing the keywords "engine" and "power", it is classified into the "automotive performance description" category in the text content, and at the same time, according to its usage which may be for product promotion, it is also classified into the corresponding usage category. The classified general manuscript sentences are stored in the database. The database table structure can be designed to include text content classification fields, usage classification fields, theme classification fields, etc., which is convenient for subsequent rapid retrieval and invocation.

[0084] Step S202: Generate synonymous derivative sentences.

[0085] Specifically, the server reads the stored general manuscript sentences, calls the large language model, sends each general manuscript sentence as input to the large language model, and attaches specific prompt information, requiring the large language model to generate sentences with the same meaning but different expressions. For example, for the input sentence "The body lines of this car are smooth", the large language model may generate "This car has beautiful and smooth body lines", etc.

[0086] Step S203: Store the synonymous derivative sentences according to the preset categories.

[0087] Specifically, using natural language processing technology, analyze the semantics, keywords, etc. of the generated synonymous derivative sentences. These synonymous derivative sentences are stored in the database according to the previously determined categories. New records can be added to the original database table, or a special synonymous derivative sentence table can be established, which also includes sentence fields, various classification fields, etc., to ensure consistency with the classification system of general manuscript sentences and facilitate subsequent unified management and use.

[0088] Step S204: Identify the variable parameters of the synonymous derivative sentences.

[0089] Specifically, the server uses entity recognition technology to identify the variable parameters of the synonymous derivative statements. For example, through a trained named entity recognition model, the synonymous derivative statements are analyzed. For the statement "The price of this [car name] is [price] ten thousand yuan", the named entity recognition model can identify "[car name]" and "[price]" as variable parameters. For parameters in some specific fields, such as parameter configurations and price reduction amounts in the automotive industry, domain knowledge graphs may need to be combined for identification. By constructing an automotive domain knowledge graph, which contains information such as various automotive components, price systems, and promotional activities, the synonymous derivative statements are matched with the knowledge graph to accurately identify relevant variable parameters.

[0090] Step S205: Set variable slots on the synonymous derivative statements to generate derivative sentence patterns.

[0091] Specifically, the server determines the positions of the variable parameters in the synonymous derivative statements. For example, for the statement "This [car name] is equipped with an engine with [parameter configuration]", the positions of "[car name]" and "[parameter configuration]" in the statement are clarified. Specific marking symbols are used to replace the variable parameters to generate derivative sentence patterns. For example, the above statement is converted to "This [car name] is equipped with an engine with [parameter configuration]", where "[]" is used as the marking symbol for the variable slots. When generating paragraph templates later, specific vehicle information can be conveniently filled into these slots.

[0092] Step S206: Send a paragraph creation request.

[0093] Specifically, on the interface of the client application, the user clicks the function button or menu option for creating a paragraph template, triggering a paragraph creation request event. The client sends the paragraph creation request to the server through the network communication protocol.

[0094] Step S207: Return a second response.

[0095] Specifically, after receiving the paragraph creation request sent by the client, the server reads the previously generated and stored derivative sentence patterns from the database. These derivative sentence patterns are encapsulated into a second response and sent back to the client through the network communication protocol.

[0096] Step S208: Display a paragraph creation interface according to the second response.

[0097] Specifically, after receiving the second response returned by the server, the client parses the message content. According to the parsed derivative sentence patterns and other relevant information, a paragraph creation interface is constructed on the interface of the client application. The paragraph creation interface will display the available derivative sentence patterns for the user to select, for example, presented in a list form.

[0098] Figure 3It is a schematic diagram of the paragraph creation interface according to an embodiment of the present invention. As Figure 3 shown, the paragraph creation interface includes a first new selection area 31, a paragraph content definition area 32, and a paragraph template setting area 33, and a second confirmation control 34. Among them, the first new selection area 31 includes a paragraph new blank control 311 and a copy existing paragraph control 312. The paragraph content definition area 32 includes a paragraph type selection control 321, a paragraph template type control 322, and a paragraph template name control 323. The template setting area 33 includes a paragraph type selection control 331 and a sentence pattern label selection control 332.

[0099] Specifically, when the user triggers the paragraph new blank control 311, a brand-new paragraph template needs to be created from scratch on this page. The user can freely select various elements to construct the paragraph structure without being restricted by the existing templates. When the user triggers the copy existing paragraph control 312, it is to copy and modify based on the existing paragraph template, which can save time and effort and is very helpful for creating paragraph templates with similar themes or structures.

[0100] When the user triggers the paragraph type selection control 321 in the paragraph content definition area 32, the user can select the content type involved in this paragraph template from the preset options, such as "explanation of passing models" and "introduction of new car features", etc. By selecting the appropriate content type, the theme direction of this paragraph template can be clarified, which is convenient for subsequent classification management and use of the template. When the user triggers the paragraph template type control 322, the type of the paragraph template can be further refined. For example, under the content type of "explanation of passing models", there may be different template type options such as "explanation of vehicle appearance", "explanation of vehicle interior", "explanation of vehicle power", etc., to help the user more accurately define the use of the paragraph template. When the user triggers the paragraph template name control 323, the user can name the newly created paragraph template here, such as "paragraph template for the explanation of the appearance of the xxxx model XX vehicle", and the clear name is convenient for the user to quickly identify and find the required paragraph template among many templates.

[0101] When the user triggers the paragraph type selection control 331 in the template setting area 33, the user can select the document type, such as "lead generation" and "live room welfare introduction", etc. Different document types correspond to different expression purposes and styles. Selecting the appropriate document type helps to construct a paragraph structure that meets the requirements. When the user triggers the sentence pattern label selection control 332, the user can select a specific sentence pattern label from the sentence pattern list. Among them, when selecting the sentence pattern label, clicking on the drop-down box will pop up the sentence pattern area 333 to display the sentence pattern list. After selecting a certain sentence pattern, the user can also see the specific content example of this sentence pattern in the sentence pattern information area 334, which helps the user better judge whether it meets the requirements. For example, when the user selects sentence pattern 4 in the sentence pattern list in the sentence pattern area 333, the user can see the specific content example of "We have checked quite a few cars for you. Now let's calculate the final price of each car when it arrives" in the sentence pattern information area 334. In addition, the user can click the "+" button to add the selected sentence pattern to the paragraph template structure. The "×" button is used to delete the added sentence pattern, and the up and down arrow buttons can adjust the order of the added sentence patterns.

[0102] Step S209: Send a new paragraph confirmation message.

[0103] Specifically, when the user completes all the settings in each area of the above paragraph creation interface and determines the parameters and structure of the new paragraph template, the second confirmation control 34 can be triggered. After the user clicks the second confirmation control for creating the paragraph template, the client triggers a paragraph creation request event and sends the new paragraph confirmation message to the server through the network communication protocol.

[0104] Step S210: Generate an initial paragraph template according to the new paragraph confirmation message.

[0105] Specifically, after receiving the new paragraph confirmation message, the server parses the message content to obtain information such as the paragraph content type, paragraph template type, paragraph template name, and paragraph template structure. The server reads the relevant derivative sentence patterns from the database according to the parsed information and combines and arranges them according to the paragraph template structure. For example, if the paragraph template structure specifies the sequence of certain derivative sentence patterns, the server combines these sentence patterns in that order to generate the final initial paragraph template.

[0106] In an embodiment of the present invention, by obtaining variable parameters input by a user, the variable parameters including vehicle detail data, generating an intermediate paragraph template according to the variable parameters and an initial paragraph template, receiving a theme creation request sent by a client, sending a first response to the client so that the client displays a theme creation interface, receiving the theme creation confirmation information sent by the client, generating a target text according to the theme creation confirmation information and the intermediate paragraph template, and sending the target text to the client so that the client displays the target text through a copy display interface. Thus, through template splicing and speech generalization, the quality of the voiceover script and the script generation efficiency are improved, meeting the diverse voiceover script requirements.

[0107] Figure 4 It is an information interaction diagram for a text generation system in an embodiment of the present invention to obtain an initial copy template. Figure 4 It shows a flowchart of the information interaction in which a player establishes a connection with a server and obtains an initial copy template in the text preparation stage, specifically including the following steps:

[0108] Step S301: Send a copy creation request.

[0109] Specifically, in the application program of the client, when the user has a need to create a new copy template, a copy creation request is sent to the server.

[0110] Step S302: Return a third response.

[0111] Specifically, the server reads relevant initial paragraph template data from a database that pre-stores initial paragraph templates according to the copy creation request, and encapsulates it into a third response in a certain data format. The initial paragraph template data may include multiple initial paragraph templates of different content types and template types. The server sends the third response back to the client.

[0112] Step S303: Display a copy creation interface according to the third response.

[0113] Specifically, after the client receives the third response sent by the server, it parses the message and extracts the initial paragraph template data and other relevant information therein. The interface rendering module of the client constructs a copy creation interface on the interface of the application program according to these data and information.

[0114] Figure 5 It is a schematic diagram of the copy creation interface in an embodiment of the present invention. As Figure 5As shown, the copywriting creation interface includes a second new selection area 41, a copywriting content definition area 42, a copywriting template setting area 43, and a third confirmation control 44. Among them, the second new selection area 41 includes a blank copywriting creation control 411 and a control 412 for copying existing copywriting. The copywriting content definition area 42 includes a copywriting content selection control 421, a copywriting template type control 422, and a copywriting template name control 423. The copywriting template setting area 43 includes a paragraph template selection control 431, a paragraph template type control 432, and a paragraph template name selection control 433.

[0115] Specifically, when the user triggers the blank copywriting creation control 411, it means creating a brand-new copywriting template from scratch. The user can freely combine various elements to build a unique copywriting structure, which is suitable for scenarios with brand-new ideas or special requirements. When the user triggers the control 412 for copying existing copywriting, it is based on the existing copywriting template for copying and modification, especially suitable for creating copywriting templates with similar themes or structures.

[0116] When the user triggers the copywriting content selection control 421, by selecting appropriate copywriting content, the user clarifies the core content direction of the created copywriting template. For example, "explanation of a certain model", "introduction of new car release", and "popular science of car maintenance knowledge", etc., which facilitates subsequent classification management and retrieval of the template. When the user triggers the copywriting template type control 422, by selecting a suitable copywriting template, the user can further refine the type of the copywriting template. Under the content selection of "explanation of a certain model", there may be different template type options such as "product feature display" and "price advantage analysis", etc., which helps the user more accurately locate the purpose and style of the copywriting template. When the user triggers the copywriting template name control 423, the user can name the newly created copywriting template here, such as "Live promotion copywriting template for model xxxx of XX car". A clear and descriptive name facilitates the user to quickly find the required copywriting template among numerous templates and improves work efficiency.

[0117] When the user triggers the paragraph template selection control 431, the user can select a suitable paragraph template from the existing paragraph template library. Each paragraph template may correspond to different parts of the copywriting, such as the opening introduction, the elaboration of product highlights, the ending guidance, etc. When the user triggers the paragraph template type control 432, the user can select a paragraph template in the drop-down box showing the specific types of paragraph templates, such as "description of vehicle model appearance", "explanation of interior details", and "introduction of power performance", etc., to help the user more clearly understand the focus of each paragraph template so as to make a suitable choice. When the user triggers the paragraph template name selection control 433, the user can select the name of the already created paragraph template. In addition, the user can also click the "+" button to add the selected paragraph template to the copywriting template structure. The "×" button is used to delete the added paragraph template, and the up and down arrow buttons can adjust the order of the added paragraph templates.

[0118] Step S304: Send the new copywriting confirmation information.

[0119] Specifically, when the user completes all the settings in the second new selection area 41, the copywriting content definition area 42, and the copywriting template setting area 43, and determines the parameters and structure of the new copywriting template, the third confirmation control 44 is triggered. After the client captures the operation event of the user triggering the third confirmation control, it organizes and encapsulates all the information filled in and selected by the user on the copywriting creation interface to generate the new copywriting confirmation information. The new copywriting confirmation information includes the copywriting content type, the copywriting mode, the mode name, and the template structure composed of the selected paragraph content type, paragraph template type, and paragraph template name, etc. The client sends the new copywriting confirmation information to the server.

[0120] Step S305: Generate an initial copywriting template according to the new copywriting confirmation information.

[0121] Specifically, after the server receives the new copywriting confirmation information sent by the client, it parses the information and extracts the various parameters therein, such as the copywriting content type, the copywriting mode, the mode name, and the template structure, etc. The server selects the corresponding paragraph templates from the previously stored initial paragraph template data according to the requirements of the template structure for combination and arrangement. For example, if the template structure specifies to first use the paragraph template of the "description of vehicle model appearance" type and then use the paragraph template of the "guide for leaving contact information" type, the server will separately retrieve the corresponding paragraph templates from the database and combine them in order. After the combination is completed, the server generates the final initial copywriting template.

[0122] In an embodiment of the present invention, by obtaining variable parameters input by a user, where the variable parameters include vehicle detail data, generating an intermediate paragraph template according to the variable parameters and an initial paragraph template, receiving a theme creation request sent by a client, sending a first response to the client so that the client displays a theme creation interface, receiving the theme creation confirmation information sent by the client, generating a target text according to the theme creation confirmation information and the intermediate paragraph template, and sending the target text to the client so that the client displays the target text through a copy display interface. Thus, through template splicing and speech generalization, the quality of the voice-over manuscript and the manuscript generation efficiency are improved, meeting the diverse needs of voice-over manuscripts.

[0123] Figure 6 It is an information interaction diagram for a text generation system in an embodiment of the present invention to obtain a target text. Figure 6 It shows a flowchart of information interaction in which a player establishes a connection with a server and obtains a target text during the text generation stage, specifically including the following steps:

[0124] Step S401: Obtain variable parameters input by a user.

[0125] Specifically, when a user wants to edit a certain paragraph template to generate a new intermediate paragraph template, the user can trigger an edit control in the management interface of the client. After the user triggers the edit control, the client will pop up or jump to a new interface. In this new interface or area, input boxes, dropdown boxes, and other controls related to the paragraph template will be displayed to guide the user to input variable parameters. Thus, the client obtains the variable parameters input by the user.

[0126] Step S402: Send variable parameters.

[0127] Specifically, the client sends the received variable parameters to the server through a network communication protocol.

[0128] Step S403: Generate an intermediate paragraph template according to the variable parameters and an initial paragraph template.

[0129] Specifically, after the server receives the variable parameters sent by the client, it reads the corresponding initial paragraph templates from a pre-stored initial paragraph template database. These initial paragraph templates are obtained by processing and generating general manuscript sentences before, and contain some preset text structures and derivative sentence patterns. The server combines the variable parameters with the initial paragraph templates and fills the variable parameters into the corresponding variable slots in the initial paragraph templates. For example, if there is a derivative sentence pattern such as "[Car name] is a model with excellent performance" in the initial paragraph template, the server will fill the specific car name input by the user into the position of "[Car name]" to generate an intermediate paragraph template.

[0130] Step S404: Send a theme creation request.

[0131] Specifically, after the client sends the variable parameters and confirms that the server has received and processed them, the user can trigger an operation for theme creation on the client interface, such as triggering the "Start creating theme copywriting" control. After the client program captures this operation, it will encapsulate a theme creation request message, which may include the user's identity information, request identifier, etc., and then send this request to the server via the network.

[0132] Step S405: Return a first response.

[0133] Specifically, after the server receives the theme creation request from the client, it filters out a suitable candidate copywriting template from the pre-stored candidate copywriting template database. The candidate copywriting templates are stored classified according to different themes, purposes, and content types. The server encapsulates the filtered candidate copywriting templates and some response information into a first response and sends it back to the client according to the network protocol and data format agreed upon with the client.

[0134] Step S406: Display a theme creation interface according to the first response.

[0135] Specifically, after the client receives the first response returned by the server, it parses the message. Extract the candidate copywriting template data and other relevant information. The interface rendering module of the client dynamically generates a theme creation interface on the application interface according to these data and information. The theme creation interface will display the candidate copywriting templates for the user to select, and at the same time, it will also provide some other setting options, such as platform selection, mode type selection, theme content type selection, generation model selection, word count setting, and copywriting naming, etc.

[0136] Figure 7 is a schematic diagram of the theme creation interface of an embodiment of the present invention. As Figure 7 shown, the theme creation interface includes a mode selection area 51, a model setting area 52, a text generation area 53, and a first confirmation control 54. The mode selection area 51 includes a platform selection control 511, a mode type control 512, a text type control 513, and a copywriting template selection control 514. The model setting area 52 includes a generation model selection control 521 and a word count setting control 522. The text generation area 53 includes a generation method selection control 531 and a copywriting naming control 532.

[0137] Specifically, when the user triggers the platform selection control 511, by selecting the corresponding platform, the generated manuscript can be adapted to the style and rules of that platform. For example, the preferences of the audience and content specifications of different platforms vary. Selecting the appropriate platform helps improve the applicability of the manuscript. Herein, the platform can be a live streaming platform.

[0138] When the user triggers the mode type control 512, the user can select different modes according to actual needs, which can include mode type options such as "complete manuscript", "fragment content", and "copywriting for interactive sessions". For example, if only the copywriting for a specific interactive session during the live stream is needed, the "copywriting for interactive sessions" mode can be selected.

[0139] When the user triggers the text type control 513, the user can clarify the core theme direction of the manuscript by selecting different content types, including various text content types such as "explanation of passing models", "introduction of new product releases", and "copywriting for promotional activities", etc., which facilitates the subsequent targeted generation of content.

[0140] When the user triggers the copywriting template selection control 514, the user can, according to their own creative ideas and purposes, select an existing initial copywriting template as the basic framework. The initial copywriting template is designed according to different scenarios and requirements as mentioned above, such as the "lead for lead generation + question and answer" template, etc.

[0141] When the user triggers the generation model selection control 521, the user can select a suitable model to generate text according to their expectations for the quality and style of the manuscript. For example, some models are good at generating lively language, while others are more suitable for rigorous and professional expressions. The generation model selection control 521 can include a dropdown box listing different available generation models. Different generation models may vary in aspects such as language style, knowledge reserve, and generation ability.

[0142] Among them, the word count setting control 522 consists of a numeric input box, "+" and "-" buttons. The user can set the word count of the manuscript to be generated by clicking the "+" and "-" buttons or directly entering a number in the input box. For example, if set to "1000", the server will attempt to generate a manuscript of approximately 1000 words, which facilitates the user to control the length of the manuscript.

[0143] The generation method selection control 531 includes two options, "Yes" and "No", for selecting the generation method of the text. For example, selecting "Yes" indicates the use of the background automatic generation method, and the server will use relevant algorithms and data to generate text in the background, and the user does not need to wait for real-time feedback during the generation process; selecting "No" may indicate the use of other generation methods, such as step-by-step generation or real-time interactive generation, etc.

[0144] When the user triggers the copywriting naming control 532, the user can name the upcoming target text here, such as "XX model live promotion copywriting" or "New product release presentation draft", etc.

[0145] Step S407: Send the theme creation confirmation information.

[0146] Specifically, in the theme creation interface, the user completes various settings according to their own needs, such as selecting a suitable candidate copywriting template, determining the platform name, mode type, theme content type, generation model, setting the copywriting name, etc. When the user completes all the settings, the user triggers the first confirmation control 54 on the theme creation interface, triggering a confirmation operation event. After the client captures the event that the user triggers the first confirmation control, it sorts out and encapsulates all the setting information of the user on the theme creation interface. This information includes the platform name, mode type, theme content type, generation model, copywriting name, and target copywriting template, etc. Then, the encapsulated theme creation confirmation information is sent to the server according to the network protocol and data format agreed upon with the server.

[0147] Step S408: Generate the target text according to the theme creation confirmation information and the middle paragraph template.

[0148] Specifically, after the server receives the theme creation confirmation information sent by the client, it parses the information. According to the specified generation model therein, it determines the target large language model from the available list of large language models. Then, taking the middle paragraph template as the input, it passes it to the target large language model, allowing it to generate the target paragraph template according to its own algorithm and training data. Finally, the server combines the generated target paragraph template with the target copywriting template, further adjusts and improves the text content, and generates the final target text.

[0149] Step S409: Send the target text.

[0150] Specifically, the server encapsulates the generated target text into a suitable data format and adds necessary response header information, such as text encoding, content type, etc. Then, it sends the encapsulated target text back to the client through the network.

[0151] Step S410: Display the target text through the copywriting display interface.

[0152] Specifically, after the client receives the target text sent by the server, it decodes and parses the text. Then, the interface display module of the client displays the target text on the copywriting display interface. This interface can be a dedicated text display area with good typesetting and format settings, facilitating the user to view and edit the generated target text.

[0153] The embodiment of the present invention obtains variable parameters input by the user, the variable parameters include vehicle details data, generates an intermediate paragraph template according to the variable parameters and the initial paragraph template, receives a theme creation request sent by the client, sends a first response to the client so that the client displays a theme creation interface, receives theme creation confirmation information sent by the client, generates a target text according to the theme creation confirmation information and the intermediate paragraph template, and sends the target text to the client so that the client displays the target text through a text display interface. Thus, through template splicing and generalization of speech, the quality of oral manuscripts and the efficiency of manuscript generation are improved to meet the diverse needs of oral manuscripts.

[0154] Figure 8 It is a flow chart of the text generation method according to an embodiment of the present invention. Figure 8 The text generation method shown specifically includes the following steps:

[0155] Step S501: Obtain variable parameters input by the user, wherein the variable parameters include vehicle detail data.

[0156] Step S502: Generate an intermediate paragraph template according to the variable parameters and the initial paragraph template.

[0157] Step S503: Receive a theme creation request sent by the client.

[0158] Step S504: Send a first response to the client so that the client displays a theme creation interface, wherein the first response includes candidate copy templates.

[0159] Step S505: receiving the theme creation confirmation information sent by the client, wherein the theme creation confirmation information includes at least one of a platform name, a mode type, a theme content type, a generation model, a text name, and a target text template.

[0160] Step S506: Generate a target text based on the theme creation confirmation information and the middle paragraph template.

[0161] Step S507: sending the target text to the client, so that the client displays the target text through a text display interface.

[0162] The specific implementation of the above steps S501-S507 may refer to the above steps S401-S411, which will not be described in detail in this embodiment.

[0163] In an embodiment of the present invention, by obtaining variable parameters input by a user, where the variable parameters include vehicle details data, generating an intermediate paragraph template according to the variable parameters and an initial paragraph template, receiving a theme creation request sent by a client, sending a first response to the client so that the client displays a theme creation interface, receiving the theme creation confirmation information sent by the client, generating a target text according to the theme creation confirmation information and the intermediate paragraph template, and sending the target text to the client so that the client displays the target text through a copy display interface. Thus, by template splicing and speech generalization, the quality of the voice-over manuscript and the manuscript generation efficiency are improved, meeting the diverse requirements for voice-over manuscripts.

[0164] Figure 9 It is a schematic diagram of a text generation device according to an embodiment of the present invention. As Figure 9 shown, the text generation device according to an embodiment of the present invention includes a variable parameter acquisition module 81 for obtaining variable parameters input by a user, where the variable parameters include vehicle details data; an intermediate paragraph template generation module 82 for generating an intermediate paragraph template according to the variable parameters and an initial paragraph template; a theme creation request acquisition module 83 for receiving a theme creation request sent by a client; a first response sending module 84 for sending a first response to the client so that the client displays a theme creation interface, where the first response includes a candidate copy template; a theme creation confirmation information receiving module 85 for receiving the theme creation confirmation information sent by the client, where the theme creation confirmation information includes at least one of a platform name, a mode type, a theme content type, a generation model, a copy name, and a target copy template; a target text generation module 86 for generating a target text according to the theme creation confirmation information and the intermediate paragraph template; and a target text display module 87 for sending the target text to the client so that the client displays the target text through a copy display interface.

[0165] In an embodiment of the present invention, by obtaining variable parameters input by a user, where the variable parameters include vehicle details data, generating an intermediate paragraph template according to the variable parameters and an initial paragraph template, receiving a theme creation request sent by a client, sending a first response to the client so that the client displays a theme creation interface, receiving the theme creation confirmation information sent by the client, generating a target text according to the theme creation confirmation information and the intermediate paragraph template, and sending the target text to the client so that the client displays the target text through a copy display interface. Thus, by template splicing and speech generalization, the quality of the voice-over manuscript and the manuscript generation efficiency are improved, meeting the diverse requirements for voice-over manuscripts.

[0166] Figure 10 It is a schematic diagram of an electronic device according to an embodiment of the present invention. As Figure 10As shown Figure 10 The electronic device shown is a text generation device, which includes a general computer hardware structure, which at least includes a processor 91 and a memory 92. The processor 91 and the memory 92 are connected by a bus 93. The memory 92 is adapted to store instructions or programs executable by the processor 91. The processor 91 can be an independent microprocessor or a set of one or more microprocessors. Thus, by executing the instructions stored in the memory 92, the processor 91 executes the method flow of the embodiment of the present invention as described above to realize the processing of data and the control of other devices. The bus 93 connects the above-mentioned multiple components together and at the same time connects the above-mentioned components to a display controller 94, a display device, and an input / output (I / O) device 95. The input / output (I / O) device 95 can be a mouse, a keyboard, a modem, a network interface, a touch input device, a body sensing input device, a printer, and other devices well-known in the art. Typically, the input / output (I / O) device 95 is connected to the system through an input / output (I / O) controller 96.

[0167] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a device (equipment), or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can be implemented as a computer program product on one or more computer-readable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) containing computer-usable program code.

[0168] The present application is described with reference to the flowcharts of methods, devices (equipment), and computer program products according to the embodiments of the present application. It should be understood that each process in the flowchart can be implemented by computer program instructions.

[0169] These computer program instructions can be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured article including an instruction device, and the instruction device implements the process Figure 1 the functions specified in one process or multiple processes.

[0170] These computer program instructions can also be provided to the processor of a general computer, a special computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices produce a device for realizing the functions specified in Figure 1 one process or multiple processes.

[0171] Another embodiment of the present invention relates to a non-volatile storage medium for storing a computer-readable program, which is used for a computer to execute the above-mentioned partial or all method embodiments.

[0172] That is, those skilled in the art can understand that all or part of the steps in implementing the methods of the above embodiments can be completed by a program to specify relevant hardware. The program is stored in a storage medium, including several instructions for causing a device (which can be a single-chip microcomputer, a chip, etc.) or a processor to execute all or part of the steps of the methods described in the embodiments of the present application. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs.

[0173] For the solutions described in this specification and the embodiments, if they involve personal information processing, they will be processed on the premise of having a legal basis (such as obtaining the consent of the personal information subject, or being necessary for performing a contract, etc.), and will only be processed within the specified or agreed scope. A user's refusal to process personal information other than the necessary information required for basic functions will not affect the user's use of the basic functions.

[0174] The above are only the preferred embodiments of the present application and are not used to limit the present application. For those skilled in the art, various modifications and changes can be made to the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present application shall be included within the protection scope of the present application.

Claims

1. A text generation method, applicable to a server, characterized in that, The method includes: Obtaining variable parameters input by a user, where the variable parameters include vehicle details data; Generating an intermediate paragraph template according to the variable parameters and an initial paragraph template; Receiving a theme creation request sent by a client; Sending a first response to the client to enable the client to display a theme creation interface, where the first response includes candidate copy templates; Receiving the theme creation confirmation information sent by the client, where the theme creation confirmation information includes at least one of a platform name, a mode type, a theme content type, a generation model, a copy name, and a target copy template; Generating a target text according to the theme creation confirmation information and the intermediate paragraph template; Sending the target text to the client to enable the client to display the target text through a copy display interface.

2. The method according to claim 1, wherein The method further includes: Classifying and storing general manuscript sentences according to a preset classification, where the preset classification includes at least one of text content, usage, and theme; Generating synonymous derivative sentences according to the general manuscript sentences through a large language model; Generating an initial paragraph template according to the synonymous derivative sentences; Generating an initial copy template according to the initial paragraph template.

3. The method according to claim 2, characterized in that The generating an initial paragraph template according to the synonymous derivative sentences includes: Storing the synonymous derivative sentences according to the preset classification; Identifying variable parameters of the synonymous derivative sentences, where the variable parameters include a car name, parameter configuration, price, and price reduction amount; Setting variable slots on the synonymous derivative sentences according to the variable parameters to generate a derivative sentence pattern; Generating an initial paragraph template according to the derivative sentence pattern.

4. The method according to claim 3, wherein The generating an initial paragraph template according to the derivative sentence pattern includes: Receiving a paragraph creation request sent by a client; Sending a second response to the client to enable the client to display a paragraph creation interface, where the second response includes the derivative sentence pattern; Receiving a new paragraph confirmation information sent by the client, where the new paragraph confirmation information includes at least one of a paragraph content type, a paragraph template type, a paragraph template name, and a paragraph template structure, and the paragraph template structure includes a manuscript type and a derivative sentence pattern selected by a user; Generating an initial paragraph template according to the new paragraph confirmation information.

5. The method according to claim 2, wherein The generating an initial copy template according to the initial paragraph template includes: Receiving a copy creation request sent by a client; Sending a third response to the client to enable the client to display a copy creation interface, where the third response includes the initial paragraph template; Receiving a new copy confirmation information sent by the client, where the new copy confirmation information includes at least one of a copy content type, a copy mode, a mode name, and a template structure, and the template structure includes at least one of a paragraph content type, a paragraph template type, and a paragraph template name; Generating an initial copy template according to the new copy confirmation information.

6. The method according to claim 1, wherein The generating a target text according to the theme creation confirmation information and the intermediate paragraph template includes: Determining a target large language model according to the generation model; Generating a target paragraph template according to the intermediate paragraph template through the target large language model; Generating the target text according to the target copy template and the target paragraph template.

7. A text generation method, applicable to a client, characterized in that, The method includes: Obtaining variable parameters input by a user and sending them to a server, so that the server generates an intermediate paragraph template according to the variable parameters and an initial paragraph template, where the variable parameters include vehicle detail data; Sending a theme creation request to the server; Receiving a first response sent by the server, where the first response includes candidate copy templates; Displaying a theme creation interface according to the first response, where the theme creation interface includes at least one of a mode selection area, a model setting area, a text generation area, and a first confirmation control. The mode selection area includes a platform selection control, a mode type control, a text type control, and a copy template selection control. The model setting area includes a generation model selection control and a word count setting control. The text generation area includes a generation method selection control and a copy naming control; In response to the first confirmation control being triggered, sending theme creation confirmation information to the server, so that the server generates a target text according to the theme creation confirmation information and the intermediate paragraph template, where the theme creation confirmation information includes at least one of a platform name, a mode type, a theme content type, a generation model, a copy name, and a target copy template; Displaying the target text through a copy display interface.

8. The method according to claim 7, wherein The method further includes: Sending a paragraph creation request to the server; Receiving a second response sent by the server, where the second response includes derivative sentence patterns; Displaying a paragraph creation interface according to the second response, where the paragraph creation interface includes at least one of a first new selection area, a paragraph content definition area, and a paragraph template setting area, and a second confirmation control. The first new selection area includes a paragraph new blank control and a copy existing paragraph control. The paragraph content definition area includes a paragraph content selection control, a paragraph template type control, and a paragraph template name control. The template setting area includes a paragraph type selection control and a sentence pattern label selection control; In response to the second confirmation control being triggered, sending new paragraph confirmation information to the server, so that the server generates an initial paragraph template according to the new paragraph confirmation information, where the new paragraph confirmation information includes at least one of a paragraph content type, a paragraph template type, a paragraph template name, and a paragraph template structure, and the paragraph template structure includes a document type and a derivative sentence pattern selected by the user; 9. The method according to claim 7, wherein The method further includes: Sending a copy creation request to the server; Receiving a third response sent by the server, where the third response includes an initial paragraph template; Displaying a copy creation interface according to the third response, where the copy creation interface includes at least one of a second new selection area, a copy content definition area, and a copy template setting area, and a third confirmation control. The second new selection area includes a copy new blank control and a copy existing copy control. The copy content definition area includes a copy content selection control, a copy template type control, and a copy template name control. The copy template setting area includes a paragraph template selection control, a paragraph template type control, and a paragraph template name selection control; In response to the triggering of the third confirmation control, send new copywriting confirmation information to the server, so that the server generates an initial copywriting template according to the new copywriting confirmation information. The new copywriting confirmation information includes at least one of copywriting content type, copywriting mode, mode name, and template structure. The template structure includes at least one of paragraph content type, paragraph template type, and paragraph template name.

10. An electronic device, comprising a memory and a processor, characterized in that, The memory is used to store one or more computer program instructions, wherein the one or more computer program instructions are executed by the processor to implement the method described in any one of claims 1-6.

Images