Drama script generation method and device, computer equipment and storage medium

Abstract

The invention provides a drama script generation method and device, computer equipment and a storage medium, and belongs to the technical field of computers.The method specifically comprises the steps that target text content obtained by a user terminal is received, and the target text content at least comprises the overall concept, theme and drama tension of a script; performing text analysis on the target text content through a learning model, randomly generating a first story setting and a first character setting of at least one script, and sending the first story setting and the first character setting to the user terminal; generating a script outline based on second story setting and second figure setting fed back by the user terminal; refining the script outline to obtain an initial script, and sending the initial script to a user terminal; and receiving verification information fed back by the user terminal to the initial script, and generating a complete target script. Through the processing scheme of the application, the drama and relevance of the drama generated by artificial intelligence are improved, and it is ensured that the high-quality drama drama with consistency, coherence and drama is generated by artificial intelligence.

Description

Technical Field

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

[0002] Artificial intelligence (AI) technology is developing at an astonishing speed, profoundly affecting various fields, and text creation is no exception. As a technology that can understand and generate natural language, the application of AI in writing has gradually shown its huge potential and broad prospects. From news reporting to novel writing, AI is changing users' traditional understanding of writing and greatly improving the efficiency of users' text creation.

[0003] In the process of drama script creation, although artificial intelligence (or large language models) can be used to learn the script based on massive data, and then use the "emergence" ability of artificial intelligence to generate a variety of drama scripts. However, such drama scripts either lack details, fail to grasp the rules of drama creation, fail to generate high-quality literary works, and fail to bring about essential quality improvements; or pay too much attention to details, but drama uses stage performance as the main means, and presents the plot through the performance of actors and stage settings. Details cannot be presented in a theater environment at all. Therefore, in actual work, the consistency, coherence, and dramatic performance of drama scripts generated by artificial intelligence are insufficient, and cannot meet the needs of drama scripts in theaters. Summary of the invention

[0004] Therefore, in order to overcome the shortcomings of the above-mentioned prior art, the present invention provides a drama script generation method, device, computer equipment and storage medium. The drama script generation method allows the learning model to generate the script setting by restricting the overall concept, theme and dramatic tension, and then generates a high-quality drama script with consistency, coherence and drama based on the script setting.

[0005] In order to achieve the above-mentioned purpose, the present invention provides a method for generating a drama script, comprising: receiving target text content obtained by a user terminal, the target text content at least including the overall concept, theme and dramatic tension of the script, and the dramatic tension is used to increase the conflict and / or thoughtfulness of the script; performing text analysis on the target text content through a learning model, and randomly generating a first story setting and a first character setting of at least one script, and sending them to the user terminal; generating a script outline based on a second story setting and a second character setting fed back by the user terminal; refining the script outline to obtain an initial script, and sending it to the user terminal; receiving verification information fed back by the user terminal on the initial script, and generating a complete target script.

[0006] In one of the embodiments, before receiving the target text content obtained by the user terminal, it also includes: receiving template text content input by the user, the template text content at least including the template overall concept, template theme and template dramatic tension as the template; using a learning model to perform text learning on the template text content, and generating script rules corresponding to the template overall concept, template theme and template dramatic tension.

[0007] In one of the embodiments, the target text content obtained by the receiving user terminal includes: receiving a text help request from a user terminal; based on the text help request, generating at least one set of help text replies according to the script rules, and sending the help text replies to the user terminal; receiving feedback information from the user terminal, and generating the target text content obtained by the user terminal based on the feedback information and the help text replies.

[0008] In one of the embodiments, the learning model is at least one of a Transformer neural network, a convolutional neural network, a recurrent neural network, a long short-term memory network, and a generative adversarial network.

[0009] In one embodiment, refining the script outline to obtain an initial script includes: obtaining corresponding drama structure rules according to a preset drama type; splitting the script outline according to the drama structure rules to obtain at least one act outline; splitting the act outline into several scenes and refining them to obtain an initial script.

[0010] In one of the embodiments, the act outline is split into several scenes and refined to obtain an initial script, including: splitting the act outline into several scenes, and generating specific script requirements based on the scenes; taking character features, scene situations and key plots as inputs for text writing, and adding the specific script requirements to form a scene script and sending it to a user terminal; after receiving feedback from the user terminal, all the scene scripts are merged to obtain an initial script.

[0011] In one of the embodiments, it also includes: generating a sharing link carrying the target script and setting corresponding sharing permissions; sending the sharing link to a specific user terminal, and receiving the modified content fed back by the specific user terminal according to the sharing permissions.

[0012] A device for generating a drama script, the device comprising: a receiving module, for receiving target text content obtained by a user terminal, the target text content at least including the overall concept, theme and dramatic tension of the script, the dramatic tension being used to increase the conflict and / or thoughtfulness of the script; a setting generation module, for performing text learning on the target text content through a learning model, and randomly generating a first story setting and a first character setting of at least one script, and sending the settings to the user terminal; an outline generation module, for generating a script outline based on a second story setting and a second character setting fed back by the user terminal; a script refinement module, for refining the script outline to obtain an initial script; and a verification module, for receiving verification information fed back by the user terminal on the initial script, and generating a complete target script.

[0013] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and wherein the processor implements the steps of the above method when executing the computer program.

[0014] A computer-readable storage medium stores a computer program, which implements the steps of the above method when executed by a processor.

[0015] Compared with the prior art, the advantages of the present invention are: by limiting the overall concept, theme and dramatic tension, the learning model generates the script setting, and the entire script generation process is divided into multiple related links, which can not only automatically generate the script outline and script content by artificial intelligence, but also introduce user terminals for manual intervention at any time, avoiding the problem of insufficient controllability of overall generation. Not only that, by limiting the overall concept, theme and dramatic tension, so that it runs through the entire script generation process, the drama and relevance of the script generated by artificial intelligence are improved, ensuring that artificial intelligence generates high-quality drama scripts with consistency, coherence and drama. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0017] Figure 1 An application scenario diagram of a drama script generation method in an embodiment;

[0018] Figure 2 A schematic diagram of a process flow of a method for generating a drama script in one embodiment;

[0019] Figure 3is a flow chart of a method for generating a drama script in another embodiment;

[0020] Figure 4 is a text content input interface displayed by a user terminal in an embodiment;

[0021] Figure 5 A flowchart of a script refinement step in an embodiment;

[0022] Figure 6 is a structural block diagram of a drama script generating device in one embodiment;

[0023] Figure 7 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0024] The embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0025] The following describes the implementation methods of the present application through specific examples, and those skilled in the art can easily understand other advantages and effects of the present application from the contents disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The present application can also be implemented or applied through other different specific implementation methods, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present application. It should be noted that, in the absence of conflict, the following embodiments and the features in the embodiments can be combined with each other. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without making creative work belong to the scope of protection of the present application.

[0026] It should be noted that various aspects of the embodiments within the scope of protection of the present invention are described below. It should be apparent that the aspects described herein can be embodied in a wide variety of forms, and any specific structure and / or function described herein is merely illustrative. Based on the present application, it should be understood by those skilled in the art that an aspect described herein can be implemented independently of any other aspect, and two or more of these aspects can be combined in various ways. For example, any number and aspect described herein can be used to implement the device and / or practice the method. In addition, other structures and / or functionalities other than one or more of the aspects described herein can be used to implement this device and / or practice this method.

[0027] It should also be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present application. The drawings only show components related to the present application rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed arbitrarily, and the component layout may also be more complicated.

[0028] Additionally, in the following description, specific details are provided to facilitate a thorough understanding of the examples. However, it will be understood by those skilled in the art that the aspects described may be practiced without these specific details.

[0029] The drama script generation method provided in this application can be applied to Figure 1 In the application environment shown. The user terminal 102 or 106 communicates with the server 104 through a network. The server 104 receives the target text content obtained by the user terminal 102 or 106. The target text content can be input by the user or generated by the server 104 and displayed to the user for confirmation via the user terminal 102 or 106; the server 104 performs text analysis on the target text content through a learning model, and randomly generates a first story setting and a first character setting of at least one script, and sends them to the user terminal 102 or 106; the user terminal 102 or 106 displays the first story setting and the first character setting, and the user can input the first story setting and the first character setting. The user terminal 102 or 106 can feed back the feedback content to the server 104; the server 104 generates a script outline based on the second story setting and the second character setting fed back by the user terminal 102 or 106; the server 104 refines the script outline to obtain an initial script and sends it to the user terminal 102 or 106; the user terminal 102 or 106 displays the initial script, the user verifies the initial script, and uploads the verification information for the initial script; the server 104 receives the verification information fed back by the user terminal on the initial script and generates a complete target script. Among them, the user terminal 102 or 106 can be, but not limited to, various personal computers, laptops, smart phones, tablet computers and portable smart devices, and the server 104 can be implemented by an independent server or a server cluster consisting of multiple servers.

[0030] In one embodiment, Figure 2 As shown, a method for generating a drama script is provided, and the method is applied to Figure 1 The server in the example is used to illustrate the following steps:

[0031] Step 201, receiving target text content obtained by a user terminal, the target text content at least includes the overall concept, theme and dramatic tension of the script, and the dramatic tension is used to increase the conflict and / or thought of the script.

[0032] The target text content includes at least the overall concept, theme and dramatic tension of the script.

[0033] The overall concept is the outline of the story. The outline can include four elements: main characters, time and place, important plots, and ending. Users can be guided to enter the corresponding content through the user terminal by asking questions such as "What kind of story do you want to write? Who are the two parties in the story conflict? What is the cause of the conflict? What is the result?" The overall concept can be a complete text or a simple short sentence or paragraph.

[0034] The theme is the practical significance of the story, corresponding to the ideology. The theme can guide the coherence of the drama script, and can also be used to create multi-layered and content-rich works in a targeted manner.

[0035] Dramatic tension is used to increase the conflict and / or ideological content of the script. Ideological content refers to the ideological significance expressed in the image system depicted in literary and artistic works, that is, the social significance expressed by the life phenomena reflected in the works and the subjective evaluation of the writer. Dramatic tension is used to reflect the contradiction of the theme of the drama, so dramatic tension needs to show the contradiction of the story. For example, "As the director of a hospital, XX should seek benefits for employees, but this time she stood against everyone. Because she must fulfill the duties of an angel in white and ensure the healthy development of the medical industry." The contradiction in this dramatic tension is the contradiction between "employee welfare" and "medical industry."

[0036] The target text content may be directly input and obtained by the user through the user terminal 102; the target text content may also be a text request for help sent by the user through the user terminal 102 to the server 104, and the server 104 randomly generates text content according to pre-trained script rules and sends it to the user terminal 102, and the user terminal 102 displays the obtained text content and is confirmed by the user. The user terminal 102 sends the target text content directly input by the user, or the text content obtained by the user terminal 102 from the server 104 and confirmed by the user, to the server 104.

[0037] Step 202: Perform text analysis on the target text content through a learning model, and randomly generate a first story setting and a first character setting of at least one script, and send them to a user terminal.

[0038] In one embodiment, the learning model is at least one of a Transformer neural network, a convolutional neural network (CNN), a recurrent neural network (RNN), a long short-term memory network (LSTM), a generative adversarial network (GAN), etc. The learning model is not limited to these networks, as long as it can perform artificial intelligence activities. The learning model is a ready-made artificial intelligence model, or it can be learned and trained by the user according to needs. The Transformer neural network is a sequence neural network model based on an attention mechanism. The Transformer only uses a self-attention mechanism to process input sequences and output sequences, so it can be calculated in parallel, greatly improving the computing efficiency.

[0039] When the learning model is a ready-made AI model, the server may not need to be pre-trained. For example, the learning model may be one of the AI ​​models such as GPT-3.5, GPT4, Baidu Qianfan, Tongyi Qianwen, etc. When the learning model is customized by the user, the server may first pre-train the Transformer neural network, convolutional neural network (CNN), recurrent neural network (RNN), long short-term memory network (LSTM) or generative adversarial network (GAN) to improve the consistency, coherence and drama of the final text.

[0040] When using Transformer neural networks, convolutional neural networks (CNN), recurrent neural networks (RNN), long short-term memory networks (LSTM), and generative adversarial networks (GAN), the server can perform semantic analysis and learning on a large number of drama texts, analyze and determine the story setting, character setting, overall conception, theme and dramatic tension of the drama script, and thus determine the long-term dependencies or script rules in the text content.

[0041] The first character setting may include the character name, and the corresponding character type, character situation, etc. The character type is the role corresponding to the character name, and may include the protagonist, the protagonist's ally, the main antagonist, the secondary antagonist, and the like. The type of character type may be selected by the user on the user terminal according to the complexity of the story. The first story setting may include character actions, character portraits, and the like. Character actions include associated events corresponding to the character name, results corresponding to the associated events, and the like. The character portrait includes the language style and work style of the character corresponding to the character name. The first story setting and the first character setting are both directly generated by the server based on the target text content, and no review or modification by the user terminal is introduced at this time.

[0042] The server performs text analysis on the target text content through the learning model. The server can randomly generate the first story setting and the first character setting of at least one script and send them to the user terminal. The server can generate the first story setting and the first character setting of multiple scripts at one time and then send them together to the user terminal. The server can also only generate the first story setting and the first character setting of one script at a time and send them to the user terminal, and then regenerate the first story setting and the first character setting of another script according to the feedback of the user terminal; the server can also determine the second story setting and the second character setting according to the feedback of the user terminal. The second story setting and the second character setting are obtained by the user reviewing and even modifying the first story setting and the first character setting through the user terminal.

[0043] Step 203, generating a script outline based on the second story setting and the second character setting fed back by the user terminal.

[0044] The user terminal 102 will receive the first story setting and the first character setting of the script in step 202 and display them to the user, and the user can determine whether the first story setting and the first character setting are appropriate. When it is determined to be appropriate, the user can directly perform a confirmation operation through the user terminal 102, and at this time, the user terminal 102 directly feeds back the confirmed first story setting and the first character setting as the second story setting and the second character setting to the server 104; when it is determined to be appropriate, the user can also input a partial change instruction through the user terminal 102, and at this time, the user terminal 102 will change the first story setting and the first character setting according to the partial change instruction, and obtain the second story setting and the second character setting, and the user terminal 102 feeds back the second story setting and the second character setting to the server 104. When it is determined to be inappropriate, the user can directly perform a denial operation through the user terminal 102, and at this time, the user terminal 102 feeds back a re-generated request, and the server 104 regenerates the first story setting and the first character setting of another script according to the request fed back by the user terminal, and sends it to the user terminal 102 again. In one embodiment, the server stores the first story settings and the first character settings generated multiple times, and the server can display the first story settings and the first character settings of the corresponding round according to the selection requirements of the user terminal.

[0045] The script outline is composed of a sequence of key events. The key event sequence is the core clue that drives the development of the story, and is also the key basis for the subsequent splitting into several acts or scenes. An act generally refers to the opening and closing of the curtain on the stage, indicating a larger plot development paragraph, indicating a major change in time, place, and scenery, and is a relatively complete paragraph. A scene refers to a smaller paragraph in an act, usually a plot caused by spatial transformation or time interval in a scene. Acts and scenes are both markers for the time and space switching of the drama plot, and are used to divide the structure and plot development of the drama. An act can contain multiple scenes, and each scene is a small paragraph in an act. A key event sequence can contain only one key event, or it can contain multiple key events that develop in time. When there are multiple key events in the key event sequence, the server can set delimiters and keywords in the script outline to distinguish different key events. At this time, each key event corresponds to an act.

[0046] The server generates a script outline based on the second story setting and the second character setting fed back by the user terminal. In one embodiment, the server can query similar template scripts from the template script database based on the second story setting and the second character setting fed back by the user terminal, and splice the template script, the second story setting and the second character setting in combination with the overall conception, theme and dramatic tension of the script in the target text content, and directly obtain the script outline. The template script can include a pre-set key event sequence. The template script can be a complete drama script, or it can only have a framework script of the key event sequence. The framework script can only include the situation (the time and place of this scene), the initial challenge (which character encountered what challenge? What difficulties or concerns?), the development (how other characters provide help, opposition or pressure), the key event (what key event occurred?), the result (what new dilemma did the character encounter, or the final result?). The server substitutes the second story setting and the second character setting into the template script according to the overall conception, theme and dramatic tension of the script in the target text content to generate a script outline.

[0047] In one embodiment, the server may also autonomously generate a key event sequence according to the second story setting and the second character setting. The server may generate a script outline according to the generated key event sequence. When the server generates the key event sequence, the server may send the key event sequence to the user terminal, and the user may adjust the key event sequence through the user terminal. The server generates a script outline according to the adjusted key event sequence.

[0048] Whether the server generates a script outline based on the second story setting and the second character setting in the template script, or generates a script outline based on the adjusted key event sequence, the server can expand the challenges of the characters in the outline and the development of the plot through the learning model. Specifically, the server should be tortuous and dramatic in plot design, rather than very direct expression. For example, Juliet was forced to marry by her parents, and the action taken by her character was not to cry and refuse, but to drink poison and pretend to die in order to escape marriage. The server can also show the character's personality and characteristics through actions and plots, such as Juliet's bravery and decisiveness. In plot design, when there are two options of language expression and body movements at the same time, the server can replace body movements with language expression (in the script outline, the language expression can be described as "character A and character B quarrel", and then in the refinement of the subsequent step 204, the corresponding dialogue is added). The server can enrich the plot and introduce the union or conflict between characters with different positions. The more plots, the better the consistency, coherence and drama of the target script obtained in the end. The server can determine whether the dramatic action is significant and vivid, such as "Juliet took the poison" instead of "Juliet was thinking about Romeo", etc. The server can succinctly explain that a character took a certain dramatic action, or an event occurred, or a conflict occurred between characters, etc.

[0049] Step 204, refine the script outline to obtain the initial script, and send it to the user terminal.

[0050] In one embodiment, the server directly refines the script outline to obtain the initial script and sends it to the user terminal. In one embodiment, the server can also split the script outline into scenes based on the key events in the key event sequence, and then further split the key events according to the drama structure rules to obtain scenes, and then generate text content containing corresponding roles, situations and key plots based on the split scenes, and finally combine the text content of all scenes according to the order and combination in the key event sequence to obtain the initial script. The server sends the initial script to the user terminal. In the refinement process, the server can be based on the plot of the drama stage to be easy to perform, and the refined content is mainly dialogue and allows body movements; the refined content cannot use micro-movements such as "tears" and "choking in the throat", and avoids only emotional descriptions. For example, the server can generate "A walks back and forth worriedly" in the script, but will not generate "A feels very worried".

[0051] Step 205, receiving verification information fed back by the user terminal on the initial script, and generating a complete target script.

[0052] The user terminal receives the initial script and verifies it by the user. The user can verify whether the initial script is suitable. There are three situations for verification: the first is that it is qualified without modification; the second is that it is qualified after modification; and the third is that it is unqualified.

[0053] When it is determined to be the first case, the user can directly perform a confirmation operation through the user terminal 102, and the user terminal 102 feeds back the confirmation information to the server 104. The server 104 sets the initial script as the target script. The server can set a "verified" label according to the confirmation operation and store it in correspondence with the initial script for subsequent text confirmation or retrieval.

[0054] When it is determined to be the second case, the user can also input a modification instruction through the user terminal 102, and the user terminal 102 feeds back the modification instruction to the server 104. The server 104 modifies the initial script according to the modification instruction and generates a target script. At this time, the server 104 can also set a "verified" label according to the modification instruction, and store the label in correspondence with the target script.

[0055] When it is determined to be the third type, the user can directly perform a denial operation through the user terminal 102. At this time, the user terminal 102 feeds back a regeneration request, and the server 104 regenerates the initial script of another script according to the request fed back by the user terminal, and sends it to the user terminal 102 again for verification.

[0056] The above method, by limiting the overall concept, theme and dramatic tension, allows the learning model to generate the script setting, and splits the entire script generation process into multiple related links. It can not only automatically generate the script outline and script content by artificial intelligence, but also introduce user terminals for manual intervention at any time, avoiding the problem of insufficient controllability of overall generation. Not only that, by limiting the overall concept, theme and dramatic tension, so that it runs through the entire script generation process, the drama and relevance of the script generated by artificial intelligence are improved, ensuring that artificial intelligence generates high-quality drama scripts with consistency, coherence and drama.

[0057] like Figure 3 As shown, in one embodiment, before receiving the target text content obtained by the user terminal, the following steps are also included:

[0058] Step 301, receiving the template text content input by the user, the template text content at least including the overall concept of the template, the theme of the template and the dramatic tension of the template.

[0059] The server may receive the template text content input by the user, and the template text content at least includes the overall concept of the template, the theme of the template, and the dramatic tension of the template. The template text content may be a complete drama script, or may only include the overall concept of the template, the theme of the template, and the dramatic tension of the template. There is no specific limit on the number of words or content of the overall concept of the template, the theme of the template, and the dramatic tension of the template, but the overall concept of the template is the outline of the template story, the theme of the template is the realistic significance of the template story, and the dramatic tension of the template reflects the contradiction of the theme of the drama.

[0060] Step 302, using a learning model to perform text learning on the template text content, and generating script rules corresponding to the overall concept of the template, the theme of the template, and the dramatic tension of the template.

[0061] The server uses a learning model to perform text learning on the template text content, and generates script rules corresponding to the overall concept of the template, the theme of the template, and the dramatic tension of the template. The server can use a learning model to perform text learning on multiple different template text contents, and generate script rules corresponding to the overall concept of the template, the theme of the template, and the dramatic tension of the template; the server can also use a learning model to perform text learning on a different template text content and generate a template story setting and a template character setting of at least one script, and send it to the user terminal, and the user terminal displays the received template story setting and template character setting to the user, and the user can judge whether the template story setting and template character setting are appropriate, and then the server generates script rules corresponding to the overall concept of the template, the theme of the template, and the dramatic tension of the template according to the template story setting and template character setting determined to be appropriate.

[0062] In one embodiment, the server can iteratively learn the template story setting and template character setting that are appropriate each time, as well as the second story setting and second character setting that are fed back by the user terminal through the learning model, thereby improving the ability of the learning model and further enhancing the drama of the script generated by artificial intelligence.

[0063] In one embodiment, receiving the target text content obtained by the user terminal includes the following steps: receiving a text help request sent by the user terminal; generating at least one set of help text replies based on the text help request and according to the script rules, and sending the help text replies to the user terminal; receiving feedback information sent by the user terminal, and generating the target text content obtained by the user terminal based on the feedback information and the help text replies.

[0064] The user terminal 106 sends a text request for help, and the server 104 receives the text request for help sent by the user terminal 106 . Figure 4It can be a text content input interface displayed on the user terminal. The position indicated by arrow 1 in the figure is a button for generating a text help request. When the user clicks the arrow position on the user terminal 106, the user terminal 106 generates a text help request and sends the text help request to the server 104. Figure 4 In order to facilitate guiding user input, the overall concept is divided into background and concept.

[0065] Based on the text help request, the server 104 generates at least one set of help text replies according to the script rules, and sends the help text replies to the user terminal 106. The server can generate multiple sets of help text replies and send them to the user terminal 106 at the same time, or it can generate only one set of help text replies and send them to the user terminal 106.

[0066] The user terminal 106 receives and displays the help text reply. When the user is satisfied with the help text reply, the user can send feedback information including adoption to the server 104 through the user terminal 106. The server generates the target text content obtained by the user terminal based on the feedback information and the help text reply.

[0067] When the user needs to modify the help text reply, the user can send feedback information including additional requirements to the server 104 through the user terminal 106. The server 104 modifies the help text reply based on the additional requirements in the feedback information and sends it to the user terminal 106 for reconfirmation. Then the user can send feedback information including adoption to the server 104 through the user terminal 106. The server generates the target text content obtained by the user terminal based on the feedback information and the help text reply.

[0068] When the user is not satisfied with the text help response received, the user may send a text help request again to the server 104 via the user terminal 106 . The server 104 will regenerate at least one new set of text help responses according to the script rules and send the new text help responses to the user terminal 106 .

[0069] The above method can assist users in writing texts by using the content generation model of the entire system when users encounter difficulties in writing for various reasons (such as lack of inspiration, only trivial text ideas, only ideas but no writing ability, and the content of the creation cannot be connected with the existing content, etc.). It can generate multiple target text contents, and the user determines whether the generated auxiliary content meets the creation goal until the target text content that meets the creation goal is obtained. Since the target text content is generated very quickly and the quality of the generation is guaranteed, it can greatly reduce the time required for human users to think and create, greatly improving the efficiency of script creation.

[0070] like Figure 5As shown, in one embodiment, refining the script outline to obtain an initial script includes the following steps:

[0071] Step 401, obtaining corresponding drama structure rules according to a preset drama type.

[0072] Different types of drama have different rules of dramatic structure. For example, the common dramatic structure rules in Western drama are the three unities, also known as the "three unities". The three unities require that the drama creation maintain consistency between time, place and plot, that is, the story narrated in a play takes place within one day (one day and night), the location is in one scene, and the plot is subject to one theme. The common dramatic structure rule of Chinese drama is "one person, one thing". The "one person" in "one person, one thing" refers to the core character that can implicate all the characters in the whole play; "one thing" refers to the core event that can affect all the plots. Therefore, "one person, one thing" can be understood as the backbone of the whole play, and it is the "pillar" that can carry all the characters and plots.

[0073] The server can obtain corresponding drama structure rules according to the preset drama type.

[0074] Step 402, split the script outline according to the drama structure rules to obtain at least one act outline.

[0075] The server splits the script outline according to the drama structure rules to obtain at least one act outline. The server can analyze the nodes where each key event in the key event sequence is located according to the drama structure rules, such as the beginning, progress, climax, and ending of the play, and then the server splits the script outline into act outlines according to each node. When there is only one key event in the key event sequence of the script outline, the script outline may only be split into one act outline.

[0076] Step 403, split the act outline into several scenes and refine them to obtain an initial script.

[0077] The server splits the act outline into several scenes and refines them to obtain the initial script. The server can split the act outline into several scenes according to the judgment of the scene. For example, the server can split it in the order of "explanation scene", "preparation scene" (there can be multiple), "discovery and sudden turn scene", "climax scene" (climax or conflict must be arranged), "rendering scene" (further promote climax or conflict), "transition scene" (connection with the next scene). In some act outlines, there may only be "explanation scene", "climax scene", "transition scene" and the like. In one embodiment, when dividing the scene, the server will also determine whether there are multiple characters in the same plot. When a new character is added and the opponent's play changes, the server determines that the drama after the new character is added is independently divided into a new scene.

[0078] The server supplements the plot that should be there but is not based on the outline text corresponding to each scene to make the whole drama story coherent.

[0079] In one embodiment, the server may also obtain the context of the act outline and further refine the act outline based on the context.

[0080] The above method can generate scripts by scene, which reduces the challenge of producing long works for large models. And through the overall design, the relevance of the scripts generated by scene is guaranteed.

[0081] In one of the embodiments, the act outline is split into several scenes and refined to obtain an initial script, including: splitting the act outline into several scenes, and generating specific requirements for the script based on the scenes; taking character characteristics, scene situations and key plots as inputs for text writing, and adding specific requirements for the script to form a scene script and send it to a user terminal; after receiving feedback from the user terminal, all scene scripts are merged to obtain an initial script.

[0082] The server divides the outline into several scenes and generates specific script requirements based on the scenes. For example, the specific script requirements corresponding to "explain the scene" are the environment, time, characters, etc. that need to be introduced in the scene.

[0083] The server uses character features, scene situations and key plots as inputs for text writing, and adds specific requirements for the script to form a scene script. The server can send the scene and the corresponding specific requirements of the script, as well as the scene script to the user terminal. At this time, the server can send different scene scripts to different user terminals, or to the same user terminal. In one embodiment, the server can also obtain the context of each scene script in the outline of the act and the theme of the target text content, and limit the context and the theme of the target text content to the specific requirements of the script to form a scene script; the server can also send the context and the theme of the target text content together with the scene script to the user terminal. This can further allow the theme and dramatic tension to run through the entire script generation process, enhance the drama and relevance of the script generated by artificial intelligence, and ensure that artificial intelligence generates high-quality drama scripts with consistency, coherence and drama.

[0084] The user verifies the generated scene script and provides feedback on the scene script through the user terminal. The server adjusts the scene script based on the feedback and merges all the scene scripts to obtain the initial script.

[0085] The above method allows users to adjust the script of each scene in a timely manner, reducing the difficulty of adjustment and increasing the speed of initial script verification.

[0086] In one embodiment, the following steps are also included: generating a sharing link carrying the target script and setting corresponding sharing permissions; sending the sharing link to a specific user terminal, and receiving the modified content fed back by the specific user terminal according to the sharing permissions.

[0087] The server may generate a sharing link carrying the target script and set corresponding sharing permissions, which may include at least one of browsing, modifying, sharing, and copying permissions.

[0088] The server sends the sharing link to a specific user terminal and receives the modification content fed back by the specific user terminal according to the sharing permissions. When other users open the sharing link through a specific user terminal, they can process the target script according to the corresponding sharing permissions. The server receives the modification content fed back by them and determines whether to modify the target script according to the sharing permissions. When the sharing permissions include modifiable content, the server modifies the target script and can also mark the changes at the same time. The server can store or modify the corresponding sharing permissions.

[0089] In one embodiment, Figure 6 As shown, a drama script generation device is provided, which includes a receiving module 501, a setting generation module 502, an outline generation module 503, a script refinement module 504 and a verification module 505.

[0090] The receiving module 501 is used to receive the target text content obtained by the user terminal. The target text content at least includes the overall concept, theme and dramatic tension of the script. The dramatic tension is used to increase the conflict and / or thought of the script.

[0091] The setting generation module 502 is used to perform text analysis on the target text content through a learning model, and randomly generate a first story setting and a first character setting of at least one script, and send them to the user terminal.

[0092] The outline generation module 503 is used to generate a script outline based on the second story setting and the second character setting fed back by the user terminal.

[0093] The script refinement module 504 is used to refine the script outline to obtain an initial script.

[0094] The verification module 505 is used to receive the verification information fed back by the user terminal on the initial script and generate a complete target script.

[0095] In one embodiment, the drama script generating device further comprises:

[0096] The template receiving module is used to receive the template text content input by the user, and the template text content at least includes the overall concept of the template, the template theme and the template dramatic tension.

[0097] The script rule generation module is used to use a learning model to perform text learning on the template text content and generate script rules corresponding to the template's overall concept, template theme and template dramatic tension.

[0098] In one embodiment, the receiving module includes:

[0099] The help unit is used to receive a text help request sent by a user terminal.

[0100] The text content generating unit is used to generate at least one set of help text replies based on the text help request and according to the script rules, and send the help text replies to the user terminal.

[0101] The feedback unit is used to receive feedback information sent by the user terminal, and generate target text content obtained by the user terminal based on the feedback information and the help text reply.

[0102] In one embodiment, the script refinement module includes:

[0103] The rule acquisition unit is used to acquire corresponding drama structure rules according to the preset drama type.

[0104] The act splitting unit is used to split the script outline according to the drama structure rules to obtain at least one act outline.

[0105] The refinement unit is used to split the act outline into several scenes and refine them to obtain the initial script.

[0106] In one embodiment, the script refinement module includes:

[0107] The scene splitting unit is used to split the act outline into several scenes and generate specific script requirements based on the scenes.

[0108] The scene script generation unit is used to take character features, scene situations and key plots as input for text writing, and add specific script requirements to form a scene script and send it to the user terminal.

[0109] The merging unit is used to merge all the scene scripts after receiving the feedback from the user terminal to obtain the initial script.

[0110] In one embodiment, the drama script generating device further comprises:

[0111] The sharing link generation module is used to generate a sharing link carrying the target script and set the corresponding sharing permissions.

[0112] The sending module is used to send the sharing link to a specific user terminal and receive the modified content fed back by the specific user terminal according to the sharing authority.

[0113] The specific definition of the drama script generation device can be found in the definition of the drama script generation method above, which will not be repeated here. Each module in the above drama script generation device can be implemented in whole or in part by software, hardware and their combination. The above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0114] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 7 As shown. The computer device includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data such as template scripts. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for generating a drama script is implemented.

[0115] Those skilled in the art will understand that Figure 7 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0116] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the following steps when executing the computer program: receiving target text content obtained by a user terminal, the target text content at least including the overall concept, theme and dramatic tension of the script, and the dramatic tension is used to increase the conflict and / or thoughtfulness of the script; performing text analysis on the target text content through a learning model, and randomly generating a first story setting and a first character setting of at least one script, and sending them to the user terminal; generating a script outline based on a second story setting and a second character setting fed back by the user terminal; refining the script outline to obtain an initial script, and sending it to the user terminal; receiving verification information fed back by the user terminal on the initial script, and generating a complete target script.

[0117] In one embodiment, the processor executes a computer program to implement, before receiving the target text content obtained by the user terminal, the method also includes: receiving template text content input by the user, the template text content at least including the overall concept of the template, the theme of the template and the dramatic tension of the template as the template; using a learning model to perform text learning on the template text content, and generating script rules corresponding to the overall concept of the template, the theme of the template and the dramatic tension of the template.

[0118] In one embodiment, the target text content obtained by the receiving user terminal is implemented when the processor executes the computer program, including: receiving a text help request sent by the user terminal; based on the text help request, generating at least one set of help text replies according to script rules, and sending the help text replies to the user terminal; receiving feedback information sent by the user terminal, and generating the target text content obtained by the user terminal based on the feedback information and the help text replies.

[0119] In one embodiment, when the processor executes a computer program, the script outline is refined to obtain an initial script, including: obtaining corresponding drama structure rules according to a preset drama type; splitting the script outline according to the drama structure rules to obtain at least one act outline; splitting the act outline into several scenes and refining them to obtain the initial script.

[0120] In one embodiment, when the processor executes the computer program, it splits the act outline into several scenes and refines them to obtain an initial script, including: splitting the act outline into several scenes, and generating specific requirements for the script based on the scenes; using character features, scene situations and key plots as inputs for text writing, and adding specific requirements for the script to form a scene script and send it to the user terminal; after receiving feedback from the user terminal, all scene scripts are merged to obtain the initial script.

[0121] In one embodiment, when the processor executes the computer program, the following steps are also implemented: generating a sharing link carrying the target script and setting corresponding sharing permissions; sending the sharing link to a specific user terminal, and receiving the modified content fed back by the specific user terminal according to the sharing permissions.

[0122] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented: receiving target text content obtained by a user terminal, the target text content at least including the overall concept, theme and dramatic tension of the script, and the dramatic tension is used to increase the conflict and / or thoughtfulness of the script; performing text analysis on the target text content through a learning model, and randomly generating a first story setting and a first character setting of at least one script, and sending them to the user terminal; generating a script outline based on a second story setting and a second character setting fed back by the user terminal; refining the script outline to obtain an initial script, and sending it to the user terminal; receiving verification information fed back by the user terminal on the initial script, and generating a complete target script.

[0123] In one embodiment, the computer program implemented when executed by the processor, before receiving the target text content obtained by the user terminal, also includes: receiving template text content input by the user, the template text content at least including the template overall concept, template theme and template dramatic tension as the template; using a learning model to perform text learning on the template text content, and generating script rules corresponding to the template overall concept, template theme and template dramatic tension.

[0124] In one embodiment, the computer program implemented when executed by the processor receives the target text content obtained by the user terminal, including: receiving a text help request sent by the user terminal; based on the text help request, generating at least one set of help text replies according to script rules, and sending the help text replies to the user terminal; receiving feedback information sent by the user terminal, and generating the target text content obtained by the user terminal based on the feedback information and the help text replies.

[0125] In one embodiment, when the computer program is executed by a processor, the script outline is refined to obtain an initial script, including: obtaining corresponding drama structure rules according to a preset drama type; splitting the script outline according to the drama structure rules to obtain at least one act outline; splitting the act outline into several scenes and refining them to obtain an initial script.

[0126] In one embodiment, when a computer program is executed by a processor, the act outline is split into several scenes and refined to obtain an initial script, including: splitting the act outline into several scenes, and generating specific requirements for the script based on the scenes; using character features, scene situations and key plots as inputs for text writing, and adding specific requirements for the script to form a scene script and send it to a user terminal; after receiving feedback from the user terminal, all scene scripts are merged to obtain an initial script.

[0127] In one embodiment, when the computer program is executed by the processor, it generates a sharing link carrying the target script and sets corresponding sharing permissions; sends the sharing link to a specific user terminal, and receives the modified content fed back by the specific user terminal according to the sharing permissions.

[0128] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by any technician familiar with the technical field within the technical scope disclosed in the present application should be included in the protection scope of the present application.

Claims

1. A method for generating a drama script, characterized in that: include: Receiving target text content obtained by a user terminal, the target text content at least including the overall concept, theme and dramatic tension of the script, the dramatic tension being used to increase the conflict and / or thought of the script; Performing text analysis on the target text content through a learning model, and randomly generating a first story setting and a first character setting of at least one script, and sending the first story setting and the first character setting to the user terminal; Generate a script outline based on the second story setting and the second character setting fed back by the user terminal; Refining the script outline to obtain an initial script, and sending it to the user terminal; Receive verification information fed back by the user terminal on the initial script, and generate a complete target script.

2. The method for generating a drama script according to claim 1, characterized in that: Before receiving the target text content obtained by the user terminal, the method further includes: Receiving template text content input by a user, wherein the template text content at least includes the overall concept of the template, the theme of the template, and the dramatic tension of the template; A learning model is used to perform text learning on the template text content to generate script rules corresponding to the overall concept of the template, the theme of the template, and the dramatic tension of the template.

3. The method for generating a drama script according to claim 2, characterized in that: The receiving target text content obtained by the user terminal includes: Receiving a text request for help from a user terminal; Based on the text request for help, generating at least one set of text responses for help according to the script rules, and sending the text responses for help to the user terminal; Feedback information sent by the user terminal is received, and target text content obtained by the user terminal is generated based on the feedback information and the help text reply.

4. The method for generating a drama script according to claim 1, characterized in that: The learning model is at least one of a Transformer neural network, a convolutional neural network, a recurrent neural network, a long short-term memory network, and a generative adversarial network.

5. The drama script generation method according to claim 1, characterized in that: The script outline is refined to obtain an initial script, including: Obtain corresponding drama structure rules according to preset drama types; Splitting the script outline according to the drama structure rule to obtain at least one act outline; The act outline is split into several scenes and refined to obtain an initial script.

6. The method for generating a drama script according to claim 5, characterized in that: The said splitting the said act outline into several scenes and refining them to obtain an initial script includes: Splitting the act outline into a number of scenes, and generating specific script requirements based on the scenes; Taking the character characteristics, scene situation and key plot as the input of text writing, and adding the specific requirements of the script, a scene script is formed and sent to the user terminal; After receiving feedback from the user terminal, all the scene scripts are merged to obtain an initial script.

7. The method for generating a drama script according to claim 1, characterized in that: Also includes: Generate a sharing link carrying the target script and set corresponding sharing permissions; The sharing link is sent to a specific user terminal, and the modified content fed back by the specific user terminal is received according to the sharing authority.

8. A drama script generating device, characterized in that: The device comprises: A receiving module, configured to receive target text content obtained by a user terminal, wherein the target text content includes at least the overall concept, theme and dramatic tension of the script, wherein the dramatic tension is used to increase the conflict and / or ideological content of the script; A setting generation module, configured to perform text learning on the target text content through a learning model, and randomly generate a first story setting and a first character setting of at least one script, and send the first story setting and the first character setting to the user terminal; An outline generation module, used to generate a script outline based on the second story setting and the second character setting fed back by the user terminal; A script refinement module, used to refine the script outline to obtain an initial script; The verification module is used to receive the verification information fed back by the user terminal on the initial script and generate a complete target script.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images