Short drama generation method and device, equipment, storage medium and program product

Abstract

The invention provides a short drama script generation method and device, equipment, a storage medium and a program product, and relates to the technical field of artificial intelligence such as large language models, agents, video generation and script generation. The method comprises the following steps: acquiring an original drama text and a short drama demand parameter; generating a complete plot abstract based on key plot elements extracted from the original plot text; generating a script outline based on the complete plot abstract, and generating diversity planning consistent with a set episode number in the short episode demand parameters based on the script outline; providing matched creation evidences for each diversity plan by using the reference script template library to obtain diversity scripts which are obtained by performing content completion on the matched creation evidences and are consistent with the set episodes in number; the reference script template library is constructed and obtained based on a reference script template obtained by carrying out templating processing on a historical script with an actual score exceeding a preset score. The method can automatically and efficiently generate the drama of the movie.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of data processing, in particular to the technical field of artificial intelligence such as large language models, agents, video generation, script generation, and the like, and more particularly to a short drama script generation method and device, an electronic device, a computer readable storage medium, and a computer program product. BACKGROUND

[0002] With the explosive growth of short video content, vertical short dramas as a new form of content are becoming the focus of the film and television and Internet content industries. Short drama creation has the characteristics of short production cycle, fast content rhythm, and strong emotional conflict, which puts higher requirements on the structural rationality and rhythm control of the script creation.

[0003] However, the current short drama scripts are mostly completed by artificial screenwriting, relying on personal experience and team collaboration, and often have problems such as low efficiency, unstable structure, and difficult to guarantee quality. SUMMARY

[0004] The present disclosure provides a short drama script generation method and device, an electronic device, a computer readable storage medium, and a computer program product.

[0005] In a first aspect, the present disclosure provides a short drama script generation method, including: obtaining an original plot text and short drama requirement parameters; generating a complete plot summary based on key plot elements extracted from the original plot text; generating a script outline based on the complete plot summary, and generating a split planning consistent with the number of episodes set in the short drama requirement parameters based on the script outline; using a preset reference script template library to provide matched creation evidence for each split planning, obtaining each split script consistent with the number of episodes set after content completion of the matched creation evidence; wherein the reference script template library is constructed based on reference script templates obtained by template processing of historical scripts with actual scores exceeding a preset score, and the reference script templates are used to provide evidence-based creation evidence.

[0006] In a second aspect, the embodiments of the present disclosure provide a short drama script generation device, comprising: an input information acquisition unit configured to acquire an original plot text and short drama requirement parameters; a complete plot abstract generation unit configured to generate a complete plot abstract based on key plot elements extracted from the original plot text; a script outline and episode planning generation unit configured to generate a script outline based on the complete plot abstract, and generate episode planning consistent with the number of episodes set in the short drama requirement parameters based on the script outline; and an episode script generation unit configured to provide matched creation evidence for each episode planning by using a preset reference script template library, to obtain each episode script consistent with the number of episodes set after content completion of the matched creation evidence. The reference script template library is constructed based on reference script templates obtained by template processing of historical scripts with actual scores exceeding a preset score, and the reference script templates are used to provide evidence-based creation evidence.

[0007] In a third aspect, the embodiments of the present disclosure provide an electronic device, comprising: at least one processor; and a memory communicatively connected with the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to implement the short drama script generation method described in the first aspect when executed.

[0008] In a fourth aspect, the embodiments of the present disclosure provide a non-transitory computer-readable storage medium storing computer instructions, which are used to enable a computer to implement the short drama script generation method described in the first aspect when executed.

[0009] In a fifth aspect, the embodiments of the present disclosure provide a computer program product comprising a computer program, which is used to enable a processor to implement the steps of the short drama script generation method described in the first aspect when executed.

[0010] The short drama script generation scheme provided by the present disclosure first acquires an original plot text and short drama requirement parameters to ensure that the script creation closely matches the number of episodes, single-episode duration and other requirements set by the user, realizes personalized customization, secondly generates a complete plot abstract based on key plot elements to effectively retain the core of the story and refine the narrative structure, lays a foundation for subsequent creation, and ensures that each episode is compact and consistent with the overall episode framework by generating a script outline and episode planning, and finally provides professional-level creation evidence for episode planning by using a reference template library constructed based on high-score historical scripts, thereby enhancing the logicality and consistency of the script. The scheme significantly improves the efficiency and quality of short drama script generation through an automated process, thereby reducing the need for human involvement and production costs, and improving the professionalism and shootability of the script, and is suitable for large-scale short drama production.

[0011] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0012] Other features, objects, and advantages of this disclosure will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 This is an exemplary system architecture to which this disclosure can be applied; Figure 2 A flowchart of a short drama script generation method provided in this disclosure embodiment; Figure 3 A flowchart illustrating a method for generating episode plans based on a script outline, provided as an embodiment of this disclosure; Figure 4 A flowchart illustrating an evaluation-based modification method provided in this disclosure embodiment; Figure 5 A flowchart illustrating a complete short drama script generation scheme provided in this embodiment of the disclosure; Figure 6 A structural block diagram of a short drama script generation device provided in this disclosure embodiment; Figure 7 This is a schematic diagram of the structure of an electronic device suitable for performing a short drama script generation method, provided as an embodiment of the present disclosure. Detailed Implementation

[0013] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding; these should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description. It should be noted that, unless otherwise specified, the embodiments and features described in this disclosure can be combined with each other.

[0014] The collection, storage, use, processing, transmission, provision, and disclosure of user personal information involved in the technical solution disclosed herein comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0015] Figure 1 An exemplary system architecture 100 is shown, in which embodiments of the short drama script generation method, apparatus, electronic device, and computer-readable storage medium disclosed herein can be applied.

[0016] like Figure 1As shown, system architecture 100 may include terminal devices 101, 102, and 103, a network 104, and a server 105. Network 104 serves as the medium for providing communication links between terminal devices 101, 102, and 103 and server 105. Network 104 may include various connection types, such as wired or wireless communication links, or fiber optic cables, etc.

[0017] Users can use terminal devices 101, 102, and 103 to interact with server 105 via network 104 to receive or send messages, etc. Various applications for enabling information communication between the terminal devices 101, 102, and 103 and server 105 can be installed. These applications include short drama generation applications, script generation applications, and instant messaging applications.

[0018] Terminal devices 101, 102, and 103 and server 105 can be either hardware or software. When terminal devices 101, 102, and 103 are hardware, they can be various electronic devices with displays, including but not limited to smartphones, tablets, laptops, and desktop computers. When terminal devices 101, 102, and 103 are software, they can be installed in the aforementioned electronic devices, and can be implemented as multiple software programs or software modules, or as a single software program or software module; no specific limitation is made here. When server 105 is hardware, it can be implemented as a distributed server cluster composed of multiple servers, or as a single server. When server 105 is software, it can be implemented as multiple software programs or software modules, or as a single software program or software module; no specific limitation is made here.

[0019] Server 105 can provide various services through its built-in applications. Taking a script generation application that provides script creation services as an example, when running this application, server 105 can achieve the following effects: First, it receives the original plot text and short drama requirement parameters transmitted by the user using terminal devices 101, 102, and 103 via network 104; then, it generates a complete plot summary based on the key plot elements extracted from the original plot text; next, it generates a script outline based on the complete plot summary, and generates an episode plan consistent with the number of episodes set in the short drama requirement parameters based on the script outline; finally, it uses a preset reference script template library to provide matching creative evidence for each episode plan, resulting in episode scripts that are complete with the matching creative evidence and consistent with the set number of episodes. This reference script template library is constructed based on reference script templates obtained by templated processing of historical scripts with actual scores exceeding the preset score, and these reference script templates are used to provide evidence-based creative evidence.

[0020] It should be noted that the original plot text and short drama requirement parameters can be temporarily obtained from terminal devices 101, 102, and 103 via network 104, or they can be pre-stored locally on server 105 through various means. Therefore, when server 105 detects that this data is already stored locally (for example, when starting to process previously retained pending script generation tasks), it can choose to directly obtain this data from locally. In this case, the exemplary system architecture 100 may also exclude terminal devices 101, 102, and 103 and network 104.

[0021] Since script generation requires significant computing resources and power, the short drama script generation methods provided in the subsequent embodiments of this disclosure are generally executed by a server 105 with strong computing power and abundant computing resources. Correspondingly, the short drama script generation device is also generally located within the server 105. However, it should also be noted that when terminal devices 101, 102, and 103 also possess sufficient computing power and resources, they can also complete the aforementioned calculations performed by the server 105 through script generation applications installed on them, thereby outputting the same results as the server 105. Especially when multiple terminal devices with different computing capabilities exist simultaneously, but the script generation application determines that the terminal device has strong computing power and abundant remaining computing resources, it can allow the terminal device to perform the aforementioned calculations, thereby appropriately reducing the computing pressure on the server 105. Accordingly, the script generation application device can also be located within terminal devices 101, 102, and 103. In this case, the exemplary system architecture 100 may also exclude the server 105 and the network 104.

[0022] It should be understood that Figure 1 The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included.

[0023] Please refer to Figure 2 , Figure 2 A flowchart of a short drama script generation method provided in this disclosure embodiment, wherein process 200 includes the following steps: Step 201: Obtain the original plot text and short drama requirement parameters; This step is intended for the execution body of the short drama script generation method (e.g., Figure 1The server 105 shown obtains the original plot text and short drama requirement parameters. The original plot text, as the basic material for content generation, typically refers to the unedited source text (such as a novel or story outline). The executing entity uses natural language processing technology to analyze its semantic structure and extract key narrative elements (such as characters, scenes, and plot). Taking a novel as an example, the executing entity can obtain the original novel text and short drama requirement parameters, including the number of episodes, episode length, video style, and target audience.

[0024] In this embodiment, the executing entity can obtain the original plot text either by having the user upload a complete novel document or by having the user submit episode fragments in batches. The original plot text can also be preprocessed; for example, a pre-trained text embedding model can be used to convert the original plot text into a high-dimensional vector while preserving semantic relevance. Simultaneously, a chapter segmentation algorithm is used to automatically divide the text into logical paragraphs, providing a structured basis for the pre-generation of subsequent episodes. For instance, when inputting a 500,000-word novel, the executing entity will first segment it by chapter, then extract the core plot, character relationships, and other elements from each chapter, ultimately organizing them into indexable structured data.

[0025] The short drama requirement parameters describe the requirements for the short drama generated from the input original plot text. These parameters may include, for example, the number of episodes to determine narrative density (e.g., 80 episodes require high-frequency conflict), the episode length to constrain the pace of a single scene (1-2 minutes requires strong conflict design), the video style to influence visual language selection (e.g., "ancient style martial arts" requires ink-wash tones), and the target audience type to drive content preference adaptation (e.g., "Generation Z" requires fast cuts and internet memes). These short drama requirement parameters can be stored in memory as key-value pairs, forming a callable creative constraint matrix. For example, when a user selects the "suspense" style, it can automatically associate parameters such as dark tones and fast cuts. Furthermore, the aforementioned short drama requirement parameters can be flexibly configured through an interactive interface. For instance, users can select the drama type using checkboxes, set the number of episodes using sliders, and select a visual style template from a drop-down menu. In addition, when users do not specify the duration of a single episode, the industry average can be automatically used as the duration of a single episode; when the audience type is not clearly defined, age preferences can be inferred based on platform data; the executing entity can also perform coupling verification of short drama requirement parameters, automatically reject contradictory combinations, and avoid the risk of generation failure from the source.

[0026] Specifically, for serialized creation scenarios, cross-session parameter inheritance is also supported. This means that when a user pauses and resumes creation, the previous short drama requirement parameter template can be automatically loaded. The execution entity can optimize short drama requirement parameters through comparative testing, such as generating two versions of a short drama and then optimizing the short drama requirement parameter library based on click data. Furthermore, short drama requirement recommendation parameters can be derived based on text features. For example, the number of novel chapters can be counted to automatically recommend the number of matching episodes, or the distribution of text sentiment can be analyzed to recommend suitable video styles.

[0027] Furthermore, to enhance usability, the system can also include built-in industry template presets. For example, for common genres such as martial arts dramas and suspense dramas, there are preset template packages containing complete short drama requirement parameters such as the number of episodes, duration, and style. When a user selects the "Martial Arts Drama Template," the system automatically loads short drama requirement parameters such as an 80-episode setting and a cool color scheme, significantly reducing the configuration cost of short drama requirement parameters. These templates can be continuously updated by operations personnel to ensure they are in sync with market trends.

[0028] Step 202: Generate a complete plot summary based on the key plot elements extracted from the original plot text; Building upon step 201, this step aims to generate a complete plot summary by the aforementioned executing entity based on the key plot elements extracted from the original plot text. Specifically, the executing entity can perform deep semantic analysis on the input original plot text using a pre-trained model, locate named entities such as characters, scenes, and props through entity recognition, construct an interaction network between characters through relation extraction, and mark key plot turning points through event detection. This yields key plot elements including chapter summaries, named entities (i.e., character entities, scene entities, prop entities, etc.), the interaction network between characters, and key plot turning points. Furthermore, the extraction process can incorporate a graph attention mechanism to construct a traversable knowledge graph from these key plot elements using a graph neural network, where nodes represent entities and edges represent relation weights, providing quantifiable association criteria for subsequent creation.

[0029] For the key plot elements obtained, the executing entity can use dynamic template filling and adaptive recombination strategies to generate a structured complete plot summary. For example, firstly, the core event types of the key plot elements are extracted and matched with a preset narrative template library; then, an attention mechanism is used to calculate the relevance score between each key plot element and the template slot, and high-weight key plot elements are filled into the corresponding positions. To improve fluency, the executing entity can also call a fine-tuned language model to perform grammatical calibration on the filling results, ensuring that the output conforms to the concise narrative style required by the script. The structured complete plot summary can provide multi-dimensional constraints for subsequent episode planning. For example, when generating the "poisoning in the Imperial Garden" summary, parameters such as "high conflict intensity, time limit, and spatial closure" are output simultaneously. These data will directly affect the pacing control when dividing subsequent episodes.

[0030] Furthermore, frequently identified named entities (such as the protagonist's name) can be cached in memory to reduce redundant computations, while complex relational reasoning is processed in parallel using a distributed computing framework. For novels with millions of words, a sliding window approach is used for segmentation, and contextual attention is employed to maintain cross-segment coherence.

[0031] Step 203: Generate a script outline based on the complete plot summary, and generate an episode plan based on the script outline that matches the number of episodes set in the short drama requirement parameters; Building upon step 202, this step aims to have the aforementioned implementing entity generate a script outline based on the complete plot summary, and then generate an episode plan consistent with the number of episodes set in the short drama's requirements parameters based on the script outline. Specifically, the implementing entity can first identify the core narrative elements in the complete plot summary (such as the main storyline, key conflict points, and character development arcs), and then match the optimal structural framework (such as a four-act structure of "introduction, development, climax, and conclusion" or a three-act structure of "goal-obstacle-reversal") through a preset short drama narrative template library, thereby obtaining the script outline. During this process, the inter-act pacing ratio can also be dynamically adjusted to meet the single-episode length constraint in the short drama's requirements parameters. For example, if a single episode is set to 1.5 minutes, then the event density of each act must be adapted to a 90-second narrative capacity to avoid lengthy dialogues or slow setups.

[0032] After obtaining the script outline, the implementing entity can perform the following steps to obtain the episode plans: First, the script outline is initially divided according to the set number of episodes. For example, for an 80-episode script, the conflict density of each episode is calculated as "total number of conflicts / number of episodes," and episode boundaries are set at key plot points. Then, all episode units are scanned to identify the first appearance of characters / props / scenes, and their characteristic parameters (such as character appearance descriptions, weapon attributes, and architectural styles) are stored in a global object library to implement a global parameter inheritance mechanism. Finally, cross-episode consistency is ensured. That is, when the same object appears again in subsequent episodes, the parameters are automatically retrieved from the object library and copied. If a parameter needs to be modified in an episode (such as a character changing clothes), the new parameter version is only bound to that episode, and other episodes still inherit the original settings. For example, if the female lead is set to have "long, straight black hair" in episode 1, and changes to curly hair in episode 30, the change will be recorded as only affecting episode 30 and subsequent episodes, while the original hairstyle remains unchanged in the first 29 episodes.

[0033] For serialized content creation, when a user adds new episode text, the system can automatically compare it with the historical object library and perform feature alignment for duplicate characters in the new content. Simultaneously, a conflict density monitoring algorithm can optimize the episode structure in real time. Specifically, when an episode is detected to have insufficient conflict intensity, it automatically compresses transitional scenes or merges adjacent episodes to ensure that each episode meets the fast-paced requirements of a short drama.

[0034] Step 204: Use the preset reference script template library to provide matching creative evidence for each episode plan, and obtain the episode scripts after the content is completed with the matching creative evidence, which are consistent with the set number of episodes.

[0035] Building upon step 203, this step aims to have the aforementioned implementing entity utilize a pre-defined reference script template library to provide matching creative evidence for each episode plan, resulting in episode scripts that are complete with the matched creative evidence and consistent with the set number of episodes. The reference script template library is constructed based on templated reference script templates obtained by templating historical scripts with actual scores exceeding the pre-defined rating. These reference script templates serve to provide evidence-based creative evidence.

[0036] The reference script template library is built from in-depth research on high-quality historical scripts. Specifically, it can be constructed in the following way: short drama scripts with actual ratings exceeding a preset threshold are selected. Then, their structured features are extracted through a script deconstruction model. The script deconstruction model is based on the Transformer architecture and can identify high-value narrative elements in the script, such as scene transition patterns, conflict design templates, and character dialogue paradigms.

[0037] To obtain the scripts for each episode, the executing entity can first perform semantic matching, transforming the core plot of the current episode plan into semantic vectors, and then retrieving the template with the closest narrative structure from the template library using cosine similarity calculation. Next, demand-level matching is performed, filtering out target templates that meet market preferences based on factors such as audience type and video style in the short drama's demand parameters. Then, the target templates are broken down into fillable narrative units, and elements are replaced and logically calibrated according to the specific scenes in the episode plan, thus obtaining the scripts for each episode. Furthermore, an adversarial generation mechanism can be introduced, with a discriminator model detecting logical conflicts to ensure that new content is coherent and connected to the original plot.

[0038] Furthermore, a feedback channel can be established. When a script generated after content completion is marked as high-quality output by the evaluation object, the template optimization process can be automatically triggered. This involves extracting the script's innovative narrative structure, de-identifying it, and adding it to the library as a new template. It also supports users manually importing scripts from popular TV series and using a difference comparison algorithm to identify improvements compared to older templates, enabling continuous iteration of the template library. This "use-feedback-evolution" mechanism allows the template library to keep up with market trends and avoids rigid creative models.

[0039] When generating episode scripts based on a series plan, matching reference script templates are needed to provide creative evidence for content completion. To improve the final quality of the episode scripts, a target reference script template that matches both semantically and in terms of short drama requirements can be retrieved from the reference script template library. Then, based on the creative evidence provided by the target reference script template, the corresponding series plan is completed, resulting in episode scripts that match the set number of episodes. Semantically, the executing entity can convert key plot descriptions of the current series plan into high-dimensional semantic vectors and match them with scene description vectors in the template library using a cosine similarity algorithm. In terms of requirements, templates can be filtered based on short drama requirements to eliminate options with incompatible styles.

[0040] The reference script template library specifically includes scene narrative templates, character profile templates, and emotional atmosphere templates, which provide evidence of creation for scene narrative plots, character profile settings, and emotional atmosphere rendering plots, respectively. Based on this, the target reference script templates can provide the following three types of structured creative evidence: Scene narrative templates can provide creative evidence for the standardized construction of plot structure. Their technical principle is to deconstruct classic narrative patterns into fillable skeleton models, including parameterized fields such as scene spatiotemporal coordinates, event sequences, and conflict intensity indices; Character profile templates can provide character behavior logic chains, ensuring the depth and consistency of character development. Their core lies in establishing a multi-dimensional vector model of character characteristics, including physiological characteristics, social attributes, psychological motivations, and behavioral patterns; Emotional atmosphere templates can associate emotion types with audiovisual elements to provide emotional rendering formulas, achieving industrialized control of emotional expression.

[0041] Furthermore, a slot filling mechanism can be used in the content completion process. For example, abstract nodes in the series planning can be mapped to the event slots of the template to automatically fill in the pre-set detailed descriptions of the template. At the same time, a style calibrator is introduced to ensure that the filled content is consistent with the style of the original text. For example, if the original series uses colloquial dialogue, the execution entity will filter out overly formal lines in the template.

[0042] During content completion, the implementing entity can supplement the content of each act constituting the corresponding drama series based on the creative evidence provided by the target reference script template. That is, each drama series contains at least one act, and the scene and characters remain unchanged in each act. Different acts correspond to different conflict events that occur in different scenes involving the characters. An act refers to a series of events completed by the same group of characters in a fixed scene. It usually serves as the basic narrative unit of the script to ensure that content completion begins at the act level.

[0043] Specifically, the implementing entity can complete specific scenes through steps such as template matching, element replacement, and dynamic calibration. Specifically, it can retrieve matching templates from a reference template library based on conflict event types and scene tags; then, it extracts the general narrative skeleton from the templates and replaces abstract slots with the current instances: character ID mapping, prop binding, and dialogue conversion; finally, it uses a pre-trained language model to detect logical conflicts and automatically filter incompatible elements. Simultaneously, it can compress dialogue turns based on episode length constraints to ensure event density meets industry standards.

[0044] The short drama script generation method provided in this disclosure first ensures that the script creation closely matches the user-defined requirements such as the number of episodes and episode length by acquiring the original plot text and short drama requirement parameters, thus achieving personalized customization. Secondly, it generates a complete plot summary based on key plot elements, effectively preserving the core story and refining the narrative structure, laying the foundation for subsequent creation. Thirdly, it ensures that each episode is compact and conforms to the overall drama framework by generating a script outline and episode planning. Finally, it utilizes a reference template library built based on high-scoring historical scripts to provide professional-level creative evidence for episode planning, thereby enhancing the script's logic and consistency. This solution significantly improves the efficiency and quality of short drama script generation through automated processes, thereby reducing manual intervention, lowering production costs, and enhancing the script's professionalism and filmability, making it suitable for large-scale short drama production.

[0045] Building upon the aforementioned implementation, the executing entity can further extract structured reference script templates from the films and television works to be processed using a pre-defined script deconstruction model, thereby achieving more efficient acquisition of reference script templates. This script deconstruction model learns script extraction knowledge from training samples comprised of historical films and television works with actual ratings exceeding a preset score and sample reference script templates extracted from these historical works. Its core function is to analyze high-rated historical films and television works and extract structured narrative templates from them.

[0046] Based on the short drama script generation method disclosed in Process 200, this disclosure provides a specific implementation scheme for generating episode plans based on the script outline. Please refer to [link / reference]. Figure 3 The flowchart, wherein process 300 includes the following steps: Step 301: Divide the script outline according to the set number of episodes to obtain the original episode outlines.

[0047] This step involves the aforementioned execution entity dividing the script outline into original episode outlines based on narrative nodes, according to the number of episodes set by the user. During this process, a sliding window algorithm can be used to dynamically detect key plot turning points as episode boundaries, ensuring that each episode contains independent conflict events.

[0048] Step 302: Extract the setting parameters of each object that appears for the first time from each original episode outline in the order of episode growth, and summarize them to obtain the object setting parameter library.

[0049] Building upon step 301, this step aims to have the executing entity extract the setting parameters of each object appearing for the first time from each original episode outline in the order of episode progression, and compile them into an object setting parameter library. That is, when a new character, prop, or scene is detected to appear for the first time, its setting parameters (material, pattern, size, etc.) will be automatically extracted, and these parameters can be encapsulated into a structured object model and stored in the global object library.

[0050] The object setting parameter library can adopt a spatiotemporal index design. Each entry contains basic parameters, effective range, and version branch. The basic parameters record the original characteristics when they first appear. The effective range is bound to the "starting episode - ending episode" interval by default. The version branch refers to the generation of a new version when the characteristics of subsequent episodes change.

[0051] Step 303: According to the object setting parameter library, inherit the setting parameters of objects that appear in each original episode outline in the order of episode growth, so as to obtain each episode plan that makes the same object have consistent setting parameters.

[0052] Based on step 302, this step aims to have the executing entity inherit the setting parameters of objects that are not appearing for the first time in each original episode outline according to the order of episode growth, so as to obtain each episode plan that makes the same objects have consistent setting parameters.

[0053] Specifically, the episode planning can be determined by performing the following two phases: In the inheritance detection phase, the object identifiers in the current episode outline are scanned. If an object with the same name exists in the global library, parameter inheritance is triggered. In the version matching phase, the effective range of the object is retrieved based on the current episode number. If it is within the effective range of the basic parameters, the basic parameters are directly copied. If it enters a new version range, a new parameter set is loaded to handle boundary cases. When the same object has multiple version parameters in the same episode, a temporary copy is automatically created and a time-space marker is added to ensure that the main storyline is not disturbed.

[0054] Further, the implementing entity can extract new setting parameters for non-first-time appearances of objects from each original episode outline in chronological order of episode growth, and determine the target episode number affected by the new setting parameters. This target episode number is used to limit the portion of the original episode outline that inherits the setting parameters of the corresponding object using the updated setting parameters. Then, new setting parameters corresponding to the target episode number are added to the object's setting parameter library, and other episode numbers (not the target episode number) are added to the original setting parameters of the corresponding object. These other episode numbers are used to limit the portion of the original episode outline that inherits the setting parameters of the corresponding object using the original setting parameters. In other words, by scanning the original episode outlines in episode order, new setting parameters for non-first-time appearances of objects are identified. Then, a semantic comparison algorithm is used to detect the difference between the new parameters and the original parameters, and they are automatically marked as valid changes. Subsequently, the effective scope of the new parameters is determined, relevant plot context is extracted, and its continuous impact is analyzed to generate a set of target episode numbers. If the new parameters only affect a local plot, the target episode number is precisely limited.

[0055] The method for generating episode plans based on a script outline provided in this embodiment first divides the script outline according to a set number of episodes to obtain original episode outlines. Then, in the order of episode growth, the setting parameters of each first-appearing object are extracted from each original episode outline, and a set of object setting parameters is compiled to obtain an object setting parameter library. Finally, based on the object setting parameter library, the setting parameters of non-first-appearing objects in each original episode outline in the order of episode growth are inherited to obtain episode plans that ensure that the same objects have consistent setting parameters. This embodiment enforces parameter inheritance in the order of episodes, ensuring a high degree of consistency in the setting parameters of non-first-appearing objects in all subsequent episodes. By dynamically capturing the first-appearing objects to build a global object parameter library, the maintenance of episode script setting consistency is automatically completed, significantly improving the creation efficiency and quality of the multi-episode script basic framework.

[0056] Based on any of the above embodiments, the executing entity also discloses a specific implementation method for generating episode scripts: First, a preset content understanding agent extracts key plot elements from the original plot text to generate a complete plot summary; then, a preset script outline generation agent generates a script outline based on the complete plot summary, and a preset episode planning agent generates episode plans consistent with the number of episodes set in the short drama requirement parameters based on the script outline; finally, a preset episode script creation agent uses a reference script template library to provide matching creative evidence for each episode plan, resulting in episode scripts with content completion based on the matching creative evidence, consistent with the set number of episodes. This implementation method utilizes multiple agents to generate high-quality episode scripts with consistent logic and coherent content, reducing manual intervention and improving creation efficiency.

[0057] Based on the short drama script generation method disclosed in Process 200, this disclosure also... Figure 4 A specific implementation of an evaluation-based modification method is shown, and its process 400 includes the following steps: Step 401: Based on the short drama requirements parameters and complete plot summary, evaluate the script outline and / or each episode script, and generate modification suggestions if the evaluation fails; In this step, the aforementioned execution entity aims to evaluate the script outline and / or each episode script based on the short drama's requirements parameters and a complete plot summary, and generate modification suggestions if the evaluation fails. Similar to the embodiments mentioned above, the execution entity can utilize a pre-defined humanoid script evaluation agent to evaluate the script outline and / or each episode script based on the content of the short drama's requirements parameters and the complete plot summary output by the agent. This humanoid script evaluation agent quantitatively assesses the script outline and episode scripts based on a pre-defined multi-dimensional evaluation model, and the evaluation indicators may include conflict density, character consistency, pacing suitability, and market preference matching.

[0058] The process of generating modification suggestions may include: first, locating the root cause of the defect through attention weight analysis; then, matching the repair strategy template in the film and television knowledge base, i.e., calling the modification plan for the problem and generating modification suggestions. The priority of modification suggestions can be dynamically divided according to the scope of impact; for example, problems within a single episode are marked as low priority, while logical gaps across episodes are marked as urgent corrections.

[0059] Step 402: Adjust the corresponding script outline and / or episode scripts according to the modification suggestions until the revised script outline and / or episode scripts pass the evaluation.

[0060] Building upon step 401, this step aims to have the aforementioned implementing entity adjust the corresponding script outline and / or episode scripts according to the modification feedback until the revised script outline and / or episode scripts pass the evaluation. Continuing the previous example, the implementing entity can specifically utilize the script outline generation agent to adjust the script outline and / or episode script creation agent according to the modification feedback, and the episode script creation agent to adjust the episode scripts according to the modification feedback, until the revised script outline and / or episode scripts pass the evaluation of the humanoid script evaluation agent.

[0061] The aforementioned implementing entities may adjust the corresponding script outlines and / or episode scripts in the following ways: For structural defects at the script outline level, the script design agent can reallocate narrative weights and rearrange the event sequence according to conflict intensity; for localized issues in episode scripts, the creative agent can perform precise replacements based on modification suggestions. Furthermore, incremental evaluation can be initiated after each modification, re-evaluating only the affected episodes to avoid wasting resources on full-scale testing. For problematic segments that still fail after three rounds of optimization, a manual intervention channel is activated, and these are recorded as negative samples and fed back to the training model.

[0062] The evaluation-based modification method disclosed in this embodiment first evaluates the script outline and / or each episode script based on the short drama's requirement parameters and a complete plot summary. If the evaluation fails, modification suggestions are generated. Then, the corresponding script outline and / or episode scripts are adjusted according to the modification suggestions until the revised script outline and / or episode scripts pass the evaluation. This embodiment constructs a closed-loop iterative mechanism, forcing and guiding the script to continuously optimize until it meets preset core requirements. Automated evaluation and intelligent suggestion generation achieve standardized quality control in the scriptwriting process, significantly improving script development efficiency and reducing rework costs later.

[0063] To enhance understanding, this disclosure also includes... Figure 5A complete short drama script generation solution is provided, consisting of five core intelligent agent modules and a reference template library: a content understanding agent, a script design agent, a script creation agent, a human-like script evaluation agent, a script reference enhancement agent, and the reference template library. Each module is relatively independent, yet they collaborate through a unified agent scheduling mechanism to achieve closed-loop control and bidirectional optimization of the creative process. The following sections will introduce each intelligent agent and the reference template library: 1. Content Understanding Agent: Responsible for in-depth understanding and information extraction of the original novel text provided by the user. Through chapter content summarization LLM (Large Language Model), the system can extract the story outline, scene elements, character relationships, and key props for each chapter, providing structured material for subsequent script adaptation. This stage ensures that the semantic foundation from the original text to the script creation is accurate and rich.

[0064] 2. Script Design Intelligent Agent: Receives the output of the content understanding intelligent agent and automatically generates a script outline. Through episode division LLM and script outline LLM, the system intelligently divides the novel story into several short drama units, clearly defining the plot twists and main character storylines of each episode, generating a complete creative blueprint including background, main plot, and character relationships, ensuring the continuity and pacing control of the short dramas.

[0065] 3. Scriptwriting AI: Based on the design outline, the system generates scene scripts for each episode. Through episode analysis LLM and scene generation LLM, the system can generate structured script content that meets the standards of short dramas, including scene descriptions, character dialogues, action conflicts, and emotional development, ensuring that each scene has the characteristics of a short drama: "single scene, strong conflict, and consistent characters".

[0066] 4. Human-like Script Evaluation Agent: Simulating a professional script review mechanism, this agent combines user input requirements (such as style, emotion, and pacing) with a standardized script evaluation system to conduct multi-dimensional quality assessments of generated scripts. This agent includes an evaluation criterion selection module and a feedback generation module, capable of automatically matching script genre standards (such as romance, suspense, science fiction, etc.) and generating targeted strengths and weaknesses analyses and improvement suggestions. This module supports multiple rounds of feedback, forming a cyclical optimization mechanism of "creation—evaluation—improvement."

[0067] 5. Script Reference Enhancement Agent: Based on the concept of evidence retrieval, it automatically provides inspiration and enhanced content for script creation by matching high-quality materials in the reference template library. It includes a content quality assessment and applicability judgment module, as well as a tag retrieval and similarity retrieval mechanism. The executing agent can intelligently recommend relevant scene narratives, dialogue fragments and emotional atmosphere templates based on semantic similarity, realizing "evidence enhancement" in the creation process and significantly improving the script's detailed expression and artistic appeal.

[0068] 6. Reference Template Library: Stores and structures high-quality script excerpts and templates, including three main categories: scene narrative templates, character portrait templates, and emotional atmosphere templates. Initially, a standardized template structure was established through manual summarization and organization by script experts. Later, automated script deconstruction technology was used to achieve continuous expansion and tag-based management, building a creative knowledge base covering multiple types and styles.

[0069] Based on the aforementioned intelligence and template libraries, a specific multi-agent collaboration and optimization process includes the following steps: 1) Users input the original text of the novel and the short drama requirements (such as drama length, style, target audience, etc.); 2) The content understanding agent extracts core elements such as chapter outlines, characters, and scenes to form a structured content summary; 3) The script design agent generates a complete script outline and episode plan based on the understanding results; 4) With the support of the reference template library and script enhancement mechanism, the script creation intelligent agent generates detailed scene scripts for each episode; 5) The humanoid script evaluation agent performs quality assessments based on a standard library and outputs targeted feedback; 6) The system automatically optimizes the script or triggers a reference enhancement mechanism for re-creation based on the feedback results, forming a closed-loop dual optimization process.

[0070] 7) After multiple rounds of optimization, the script can be output as a short drama script or episode text that can be filmed, ensuring that the content can be produced and consumed by users.

[0071] This embodiment first obtains the original plot text and short drama requirement parameters to ensure that the script creation closely matches the user's requirements for the number of episodes, episode length, etc., achieving personalized customization. Secondly, it generates a complete plot summary based on key plot elements, effectively preserving the core story and refining the narrative structure, laying the foundation for subsequent creation. Thirdly, it generates a script outline and episode planning to ensure that each episode is compact and conforms to the overall drama framework. Finally, it utilizes a reference template library built based on high-scoring historical scripts to provide professional-level creative evidence for episode planning, thereby enhancing the script's logic and consistency. This solution significantly improves the efficiency and quality of short drama script generation through automated processes, thereby reducing manual intervention, lowering production costs, and enhancing the script's professionalism and filmability, making it suitable for large-scale short drama production.

[0072] Further reference Figure 6 As an implementation of the methods shown in the above figures, this disclosure provides an embodiment of a short drama script generation device, which is similar to... Figure 2 Corresponding to the method embodiments shown, this device can be specifically applied to various electronic devices.

[0073] likeFigure 6 As shown, the short drama script generation device 600 of this embodiment may include: an input information acquisition unit 601, a complete plot summary generation unit 602, a script outline and episode planning generation unit 603, and an episode script generation unit 604. The input information acquisition unit 601 is configured to acquire the original plot text and short drama requirement parameters; the complete plot summary generation unit 602 is configured to generate a complete plot summary based on key plot elements extracted from the original plot text; the script outline and episode planning generation unit 603 is configured to generate a script outline based on the complete plot summary, and generate an episode plan consistent with the number of episodes set in the short drama requirement parameters based on the script outline; the episode script generation unit 604 is configured to provide matching creative evidence for each episode plan using a preset reference script template library, obtaining episode scripts with content completion based on the matched creative evidence, consistent with the set number of episodes; wherein, the reference script template library is constructed based on reference script templates obtained by templated processing of historical scripts with actual scores exceeding the preset score, and the reference script templates are used to provide evidence-based creative evidence.

[0074] In this embodiment, the specific processing and technical effects of the input information acquisition unit 601, the complete plot summary generation unit 602, the script outline and episode planning generation unit 603, and the episode script generation unit 604 in the short drama script generation device 600 can be referred to respectively. Figure 2 The relevant descriptions of steps 201-204 in the corresponding embodiments will not be repeated here.

[0075] In some other optional implementations of this embodiment, the input information acquisition unit 601 is further configured to acquire the original novel text and short drama requirement parameters, including the number of episodes, episode length, video style, and audience type.

[0076] In some other optional implementations of this embodiment, the complete plot summary generation unit 602 is further configured to: extract key plot elements, including chapter outlines, characters, objects, and scenes, from the original plot text; and generate a structured complete plot summary based on the key plot elements.

[0077] In some other optional implementations of this embodiment, the script outline and episode planning generation unit 603 includes an episode planning generation subunit configured to generate episode plans consistent with the number of episodes set in the short drama requirement parameters based on the script outline. The episode planning generation subunit includes: an episode division module configured to divide the script outline according to the set number of episodes to obtain each original episode outline; a setting parameter extraction module configured to extract the setting parameters of each first-appearing object from each original episode outline in the order of episode growth and summarize them to obtain an object setting parameter library; and a setting parameter inheritance module configured to inherit the setting parameters of non-first-appearing objects in each original episode outline in the order of episode growth according to the object setting parameter library, so as to obtain each episode plan that makes the same object have consistent setting parameters.

[0078] In some other optional implementations of this embodiment, the episode planning generation subunit may further include: a new setting parameter extraction module, configured to extract new setting parameters of non-first-time-appearing objects sequentially from each original episode outline according to the episode growth order, and determine the target episode number affected by the new setting parameters; wherein, the target episode number is used to limit the portion of the original episode outline that inherits the setting parameters of the corresponding object using the setting parameters updated by the new setting parameters; The module for adding episode numbers based on impact is configured to add new setting parameters corresponding to the target episode number to the corresponding object in the object setting parameter library, and to add other episode numbers that are not the target episode number to the original setting parameters of the corresponding object; wherein, other episode numbers are used to limit the inheritance of setting parameters of the corresponding object using the original setting parameters.

[0079] In some other optional implementations of this embodiment, the episode script generation unit 604 is further configured to: for each episode plan in the episode planning, retrieve a target reference script template from the reference script template library that matches both the semantic level and the requirement level of the short sentence requirement parameters, and complete the content of the corresponding episode plan based on the creative evidence provided by the target reference script template to obtain an episode script consistent with the set number of episodes.

[0080] In some other optional implementations of this embodiment, the reference script template library includes scene narrative templates, character profile templates, and emotional atmosphere templates, which are used to provide evidence of creation for scene narrative plots, character profile settings, and emotional atmosphere rendering plots, respectively.

[0081] In some other optional implementations of this embodiment, the short drama script generation device 600 may further include: an automated reference script template extraction unit, configured to extract structured reference script templates from the film and television works to be processed through a preset script deconstruction model; wherein the script deconstruction model learns script extraction knowledge from training samples consisting of historical film and television works with actual scores exceeding preset scores and sample reference script templates parsed from historical film and television works.

[0082] In some other optional implementations of this embodiment, the episode script generation unit 604 includes a completion processing subunit configured to complete the content of the corresponding series plan based on the creative evidence provided by the target reference script template. The completion processing subunit is further configured to: complete the content of each scene constituting the corresponding series based on the creative evidence provided by the target reference script template; wherein, the series contains at least one scene, the scenes and characters remain unchanged in each scene, and different scenes correspond to different conflict events that occur to the characters in different scenes.

[0083] In some other optional implementations of this embodiment, the complete plot summary generation unit 602 is further configured to: generate a complete plot summary using key plot elements extracted from the original plot text by a preset content understanding agent; In some other optional implementations of this embodiment, the script outline and episode planning generation unit 603 is further configured to: generate a script outline based on a complete plot summary using a preset script outline generation agent, and generate an episode plan consistent with the number of episodes set in the short drama requirement parameters using a preset episode planning agent based on the script outline. In some other optional implementations of this embodiment, the episode script generation unit 604 is further configured to: use a preset episode script creation agent to provide matching creation evidence for each episode plan using a reference script template library, and obtain each episode script after content completion with the matching creation evidence, which is consistent with the set number of episodes.

[0084] In some other optional implementations of this embodiment, the short drama script generation device 600 may further include: an evaluation unit configured to evaluate the script outline and / or each episode script based on the short drama requirement parameters and a complete plot summary, and generate modification suggestions if the evaluation fails; and an adjustment unit configured to adjust the corresponding script outline and / or episode script based on the modification suggestions until the modified script outline and / or episode script passes the evaluation.

[0085] In some other optional implementations of this embodiment, the short drama script generation device 600 may further include: a humanoid script evaluation intelligence agent evaluation processing unit, configured to use a preset humanoid script evaluation intelligence agent to understand the complete plot summary output by the intelligence agent based on the content of the short drama requirement parameters, evaluate the script outline output by the script outline generation intelligence agent and / or each episode script output by the episode script creation intelligence agent, and return the generated modification opinions to the script outline generation intelligence agent and / or the episode script creation intelligence agent when the evaluation fails, so that the script outline generation intelligence agent adjusts the script outline according to the modification opinions and / or the episode script creation intelligence agent adjusts the episode scripts according to the modification opinions, until the modified script outline and / or episode scripts pass the evaluation of the humanoid script evaluation intelligence agent.

[0086] This embodiment exists as a device embodiment corresponding to the method embodiment described above. The short drama script generation device provided in this embodiment first obtains the original plot text and short drama requirement parameters to ensure that the script creation closely matches the user-set requirements such as the number of episodes and the length of each episode, achieving personalized customization. Secondly, it generates a complete plot summary based on key plot elements, effectively preserving the core of the story and refining the narrative structure, laying the foundation for subsequent creation. Furthermore, it ensures that each episode is compact and conforms to the overall drama framework by generating a script outline and episode planning. Finally, it utilizes a reference template library built based on high-scoring historical scripts to provide professional-level creative evidence for episode planning, thereby enhancing the script's logic and consistency. This solution significantly improves the efficiency and quality of short drama script generation through automated processes, thereby reducing manual intervention, lowering production costs, and enhancing the script's professionalism and filmability, making it suitable for large-scale short drama production.

[0087] According to embodiments of this disclosure, this disclosure also provides an electronic device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, which, when executed by the at least one processor, enable the at least one processor to implement the short drama script generation method described in any of the above embodiments.

[0088] According to embodiments of this disclosure, this disclosure also provides a readable storage medium storing computer instructions that enable a computer to implement the short drama script generation method described in any of the above embodiments when executed.

[0089] According to embodiments of this disclosure, this disclosure also provides a computer program product that, when executed by a processor, can implement the short drama script generation method described in any of the above embodiments.

[0090] Figure 7A schematic block diagram of an example electronic device 700 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0091] like Figure 7 As shown, device 700 includes a computing unit 701, which can perform various appropriate actions and processes based on a computer program stored in read-only memory (ROM) 702 or a computer program loaded into random access memory (RAM) 703 from storage unit 708. The RAM 703 may also store various programs and data required for the operation of device 700. The computing unit 701, ROM 702, and RAM 703 are interconnected via bus 704. Input / output (I / O) interface 705 is also connected to bus 704.

[0092] Multiple components in device 700 are connected to I / O interface 705, including: input unit 706, such as keyboard, mouse, etc.; output unit 707, such as various types of monitors, speakers, etc.; storage unit 708, such as disk, optical disk, etc.; and communication unit 709, such as network card, modem, wireless transceiver, etc. Communication unit 709 allows device 700 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0093] The computing unit 701 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 701 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 701 performs the various methods and processes described above, such as the short drama script generation method. For example, in some embodiments, the short drama script generation method may be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 708. In some embodiments, part or all of the computer program may be loaded and / or installed on device 700 via ROM 702 and / or communication unit 709. When the computer program is loaded into RAM 703 and executed by the computing unit 701, one or more steps of the short drama script generation method described above may be performed. Alternatively, in other embodiments, the computing unit 701 may be configured to perform the short drama script generation method by any other suitable means (e.g., by means of firmware).

[0094] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0095] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0096] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0097] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0098] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as a data server), or computing systems that include middleware components (e.g., an application server), or computing systems that include frontend components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., a communication network). Examples of communication networks include local area networks (LANs), wide area networks (WANs), and the Internet.

[0099] Computer systems can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and Virtual Private Server (VPS) services, such as high management difficulty and weak business scalability.

[0100] According to the technical solution of this disclosure, firstly, by acquiring the original plot text and short drama requirement parameters, the script creation is ensured to closely match the user's requirements for the number of episodes, episode length, etc., achieving personalized customization. Secondly, a complete plot summary is generated based on key plot elements, effectively preserving the core of the story and refining the narrative structure, laying the foundation for subsequent creation. Furthermore, by generating a script outline and episode planning, each episode's content is ensured to be compact and conform to the overall drama framework. Finally, a reference template library built based on high-scoring historical scripts provides professional-level creative evidence for episode planning, thereby enhancing the script's logic and consistency. This solution significantly improves the efficiency and quality of short drama script generation through automated processes, thereby reducing manual intervention, lowering production costs, and enhancing the script's professionalism and filmability, making it suitable for large-scale short drama production.

[0101] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this disclosure can be achieved, and this is not limited herein.

[0102] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A method for generating short drama scripts, comprising: Obtain the original plot text and short drama requirement parameters; A complete plot summary is generated based on the key plot elements extracted from the original plot text; A script outline is generated based on the complete plot summary, and an episode plan is generated based on the script outline, which is consistent with the number of episodes set in the short drama requirement parameters; A preset reference script template library is used to provide matching creative evidence for each episode plan, resulting in episode scripts that are completed with the matching creative evidence and are consistent with the set number of episodes; wherein, the reference script template library is constructed based on reference script templates obtained by template processing of historical scripts with actual scores exceeding the preset score, and the reference script templates are used to provide evidence of creation.

2. The method according to claim 1, wherein, The process of obtaining the original plot text and short drama requirement parameters includes: Obtain the original novel text and short drama requirements parameters, including the number of episodes, episode length, video style, and target audience type.

3. The method according to claim 1, wherein, The process of generating a complete plot summary based on key plot elements extracted from the original plot text includes: Extract key plot elements, including chapter summaries, characters, objects, and scenes, from the original plot text. A structured, complete plot summary is generated based on the key plot elements.

4. The method according to claim 1, wherein, The generation of episode plans based on the script outline, which matches the number of episodes set in the short drama requirement parameters, includes: The script outline is divided according to the set number of episodes to obtain the original episode outlines; The setting parameters of each object appearing for the first time are extracted from each original episode outline in the order of episode growth, and the object setting parameter library is obtained by summarizing them. Based on the object setting parameter library, the setting parameters of objects that are not the first appearance in each original episode outline according to the growth order of the series are inherited in sequence to obtain each episode plan that makes the same object have consistent setting parameters.

5. The method according to claim 4, further comprising: According to the order of episode growth, new setting parameters of objects that are not appearing for the first time are extracted from each of the original episode outlines, and the target episode number affected by the new setting parameters is determined; wherein, the target episode number is used to limit the part of the original episode outline that uses the setting parameters updated by the new setting parameters to inherit the setting parameters of the corresponding object. In the object setting parameter library, new setting parameters corresponding to the target series number are added to the corresponding object, and other series numbers other than the target series number are added to the original setting parameters of the corresponding object; wherein, the other series numbers are used to limit the original episode outlines that inherit the setting parameters of the corresponding object using the original setting parameters.

6. The method according to claim 1, wherein, The step of using a preset reference script template library to provide matching creative evidence for the episode planning, and obtaining episode scripts that are completed with the matched creative evidence and match the set number of episodes, includes: For each episode in the episode planning, a target reference script template that matches both the semantic level and the requirement level of the short sentence requirement parameters is retrieved from the reference script template library. The corresponding episode planning is then supplemented with content based on the creative evidence provided by the target reference script template to obtain an episode script that matches the set number of episodes.

7. The method according to claim 6, wherein, The reference script template library includes scene narrative templates, character profile templates, and emotional atmosphere templates, which are used to provide evidence of creation for scene narrative plots, character profile settings, and emotional atmosphere rendering plots, respectively.

8. The method according to claim 7, further comprising: The script deconstruction model extracts structured reference script templates from the films and television works to be processed. The script deconstruction model learns script extraction knowledge from training samples consisting of historical films and television works with actual scores exceeding the preset score and sample reference script templates parsed from the historical films and television works.

9. The method according to claim 6, wherein, The process of supplementing the content of the corresponding TV series plan based on the creative evidence provided by the target reference script template includes: Based on the creative evidence provided by the target reference script template, the content of each act constituting the corresponding drama series is supplemented; wherein, the drama series contains at least one act, the scene and characters remain unchanged in each act, and different acts correspond to different conflict events that occur to the characters in different scenes.

10. The method according to any one of claims 1-9, wherein, The process of generating a complete plot summary based on key plot elements extracted from the original plot text includes: A complete plot summary is generated by using a pre-defined content understanding agent to extract key plot elements from the original plot text; The process of generating a script outline based on the complete plot summary, and generating an episode plan based on the script outline that matches the number of episodes set in the short drama requirement parameters, includes: A pre-defined script outline generation agent generates a script outline based on the complete plot summary, and a pre-defined episode planning agent generates an episode plan based on the script outline that matches the number of episodes set in the short drama requirement parameters. The step of using a preset reference script template library to provide matching creative evidence for each episode plan, and obtaining episode scripts that are completed with the matched creative evidence and match the set number of episodes, includes: Using a pre-defined episode script creation agent, the reference script template library is used to provide matching creation evidence for each episode plan, resulting in episode scripts that are completed with the matching creation evidence and are consistent with the set number of episodes.

11. The method according to any one of claims 1-9, further comprising: Based on the short drama requirement parameters and the complete plot summary, the script outline and / or each episode script are evaluated, and modification suggestions are generated if the evaluation fails. The script outline and / or episode scripts will be adjusted according to the aforementioned modification suggestions until the revised script outline and / or episode scripts pass the evaluation.

12. The method of claim 10, further comprising: Using a pre-defined humanoid script evaluation agent, the system understands the complete plot summary output by the agent based on the content of the short drama requirement parameters. It then evaluates the script outline generated by the script outline generation agent and / or the episode scripts generated by the episode script creation agent. If the evaluation fails, the system returns modification suggestions to the script outline generation agent and / or the episode script creation agent, allowing the script outline generation agent to adjust the script outline and / or the episode script creation agent to adjust the episode scripts according to the modification suggestions, until the modified script outline and / or episode scripts pass the evaluation of the humanoid script evaluation agent.

13. A short drama script generation device, comprising: The input information acquisition unit is configured to acquire the original plot text and short drama requirement parameters; The complete plot summary generation unit is configured to generate a complete plot summary based on key plot elements extracted from the original plot text; The script outline and episode planning generation unit is configured to generate a script outline based on the complete plot summary, and generate an episode plan based on the script outline that matches the number of episodes set in the short drama requirement parameters; The episode script generation unit is configured to use a preset reference script template library to provide matching creative evidence for each episode plan, and to obtain episode scripts that are completed with the matching creative evidence and are consistent with the set number of episodes; wherein, the reference script template library is constructed based on reference script templates obtained by template processing of historical scripts with actual scores exceeding the preset score, and the reference script templates are used to provide evidence of creation.

14. An electronic device, comprising: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the short drama script generation method according to any one of claims 1-12.

15. A non-transitory computer-readable storage medium storing computer instructions for causing the computer to perform the short drama script generation method according to any one of claims 1-12.

16. A computer program product comprising a computer program that, when executed by a processor, implements the steps of the short drama script generation method according to any one of claims 1-12.

Images