Content processing method and device, electronic equipment, storage medium and program product

Abstract

The invention discloses a content processing method and device, electronic equipment, a storage medium and a program product, and is applied to the technical field of artificial intelligence. The method comprises the steps that a target template in a preset content template library is obtained, the content template library comprises a plurality of preset templates, each preset template comprises at least one structured content module, each structured content module comprises structured content and a corresponding weight, and the structured content is used for guiding output content information; the weight corresponding to the structured content is used for guiding the proportion of the output content in the total output content; generating metadata of the creation content based on the initial content of the creation content and the target template; and inputting the metadata of the creation content into a preset large model, and outputting target content matched with the target template through the guidance of the weight corresponding to the structured content module in the target template and the structured content. In this way, the structure consistency, the output efficiency and the output quality of the output target content can be effectively improved.

Description

Technical Field

[0001] This application belongs to the field of artificial intelligence technology, specifically relating to a content processing method, a content processing device, an electronic device, a computer-readable storage medium, and a computer program product. Background Technology

[0002] With the rapid development of science and technology, the text generation capabilities of artificial intelligence (AI) models have been continuously enhanced, and current AI models have been widely applied in many content creation scenarios. However, due to the probabilistic nature of AI model generation mechanisms, the structure and style of content output from the same input often vary considerably. This randomness leads to insufficient adaptability of the produced content, making it difficult to accurately meet the specific requirements of the publicity and distribution scenario regarding brand tone, communication focus, etc. Moreover, the output content often requires significant manual revision and calibration, increasing time and manpower costs, and easily affecting project efficiency due to repeated adjustments.

[0003] Therefore, how to standardize the output of AI models is a technical problem that urgently needs to be solved. Summary of the Invention

[0004] This application aims to address at least one of the technical problems existing in the prior art. To this end, this application proposes a content processing method, content processing apparatus, electronic device, computer-readable storage medium, and computer program product, which can effectively improve the structural consistency, output efficiency, and output quality of the output target content.

[0005] Firstly, this application provides a content processing method, including: Obtain a target template from a preset content template library. The content template library includes multiple preset templates. Each preset template includes at least one structured content module. Each structured content module includes structured content and corresponding weights. The structured content is used to guide the output content information, and the weights corresponding to the structured content are used to guide the proportion of the output content to the total output content. Metadata of the created content is generated based on the initial content of the created content and the target template; The metadata of the created content is input into a preset large model. Guided by the corresponding weights of the structured content modules in the target template and the structured content, the target content that matches the target template is output.

[0006] Secondly, this application provides a content processing apparatus, comprising: The acquisition module is used to acquire a target template from a preset content template library. The content template library includes multiple preset templates. Each preset template includes at least one structured content module. The structured content module includes structured content and corresponding weights. The structured content is used to guide the output content information, and the weights corresponding to the structured content are used to guide the proportion of the output content to the total output content. A generation module is used to generate metadata for the created content based on the initial content of the created content and the target template; The input / output module is used to input the metadata of the created content into a preset large model, and output target content that matches the target template through the corresponding weights of the structured content module in the target template and the guidance of the structured content.

[0007] Thirdly, this application provides an electronic device including a memory and a processor. The memory stores a computer program, and the processor executes the above-described content processing method by calling the computer program stored in the memory.

[0008] Fourthly, this application provides a computer-readable storage medium storing a computer program adapted for loading by a processor to execute the above-described content processing method.

[0009] Fifthly, this application provides a computer program product including computer instructions, which, when executed by a processor, implement the above-described content processing method.

[0010] The content processing method, content processing apparatus, electronic device, computer-readable storage medium, and computer program product provided in this application's embodiments select a preset template adapted to the usage requirements from a preset content template library as the target template. Then, the initial content of the created content and the target template are input as metadata into a preset large model. The preset large model guides the output of each component of the target content based on the structured content in each structured content module of the target template, and specifies the proportion of each component in the total output target content according to the weight corresponding to each structured content. In this way, the output target content is a combination of component content generated by various structured content in different proportions, which ensures that the output target content matches the target template, avoids deviations caused by free generation, and effectively improves the standardization and structural consistency of the output target content.

[0011] Meanwhile, the structured content in each module of the target template helps the pre-set large model clarify the creative direction, ensure content adaptability, avoid output deviating from requirements, improve output efficiency and quality, and reduce manual revision costs. Furthermore, new users can directly call upon the target template to quickly generate target content that meets their needs, lowering the learning curve.

[0012] Additional aspects and advantages of embodiments of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of embodiments of this application. Attached Figure Description

[0013] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which: Figure 1 This is an application scenario diagram of the content processing method provided in the embodiments of this application; Figure 2 This is a first flowchart illustrating the content processing method provided in the embodiments of this application; Figure 3 This is a schematic diagram showing the preset template provided in the embodiments of this application; Figure 4 This is a first schematic diagram illustrating the generation of target content using the content processing method provided in this application embodiment; Figure 5 This is a second schematic diagram illustrating the generation of target content using the content processing method provided in this application embodiment; Figure 6 This is a second flowchart illustrating the content processing method provided in the embodiments of this application; Figure 7 This is a schematic diagram of the basic information of the configuration preset template of the content processing method provided in the embodiments of this application; Figure 8 This is a first schematic diagram of a structured content module of a configuration preset template for the content processing method provided in this application embodiment; Figure 9 This is a second schematic diagram of the structured content module of the configuration preset template of the content processing method provided in the embodiments of this application; Figure 10 This is a third schematic diagram of a structured content module of a configuration preset template for the content processing method provided in this application embodiment; Figure 11 This is a third flowchart illustrating the content processing method provided in the embodiments of this application; Figure 12 This is a fourth schematic diagram of the structured content module of the configuration preset template of the content processing method provided in the embodiments of this application; Figure 13This is a schematic diagram of the fourth process of the content processing method provided in the embodiments of this application; Figure 14 This is the fifth schematic diagram of the structured content module of the configuration preset template of the content processing method provided in the embodiments of this application; Figure 15 This is a schematic diagram of a preset template for previewing the content processing method provided in the embodiments of this application; Figure 16 This is a schematic diagram of the content processing apparatus provided in the embodiments of this application; Figure 17 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application; Figure 18 This is a schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0014] The embodiments of this application are described in detail below. Examples of the embodiments of this application are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain this application, and should not be construed as limiting this application.

[0015] In the description of this application, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, features defined as "first" or "second" may explicitly or implicitly include one or more features. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0016] For ease of understanding, some of the technical terms used in this application are explained below: Artificial Intelligence (AI) is the theory, methods, technology, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to achieve optimal results. In other words, AI is a comprehensive technology within computer science that attempts to understand the essence of intelligence and produce new intelligent machines that can react in a way similar to human intelligence. AI studies the design principles and implementation methods of various intelligent machines, enabling them to possess perception, reasoning, and decision-making capabilities.

[0017] Artificial intelligence (AI) is a comprehensive discipline encompassing a wide range of fields, including both hardware and software technologies. Fundamental AI technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, pre-trained model technology, and operating / interactive systems. Pre-trained models, also known as large-scale models or foundational models, can be fine-tuned and widely applied to downstream tasks across various AI fields. AI software technologies primarily include computer vision, speech processing, natural language processing, and machine learning / deep learning.

[0018] Machine Learning (ML) is a multidisciplinary field involving probability theory, statistics, approximation theory, convex analysis, and algorithm complexity theory. It specifically studies how computers can simulate or implement human learning behavior to acquire new knowledge or skills and reorganize existing knowledge structures to continuously improve their performance. Machine learning is the core of artificial intelligence and the fundamental way to endow computers with intelligence; its applications span all areas of artificial intelligence. Machine learning and deep learning typically include techniques such as artificial neural networks, belief networks, reinforcement learning, transfer learning, and inductive learning. Pre-trained models are the latest development in deep learning, integrating all of these techniques.

[0019] Deep learning is a new research direction in the field of machine learning. Its goal is to enable machines to have analytical and learning capabilities like humans, thereby solving complex pattern recognition problems. Deep learning aims to learn the inherent patterns and hierarchical representations of sample data. The information gained in this learning process is of great help in interpreting data such as text, images, and sound.

[0020] The core of deep learning lies in neural networks, which consist of a large number of artificial neurons. Each neuron processes input data through connections to weights and activation functions. The hierarchical structure of a neural network includes input layers, hidden layers, and output layers, with multiple hidden layers possible. In deep learning, data is passed through the forward propagation process of the neural network to obtain the output, and then backpropagation is used to update the weights and biases of the neural network, making it better suited to the training data.

[0021] A large model refers to an artificial intelligence model trained on massive amounts of data, possessing a huge number of parameters (typically in the billions to hundreds of billions). It is capable of handling complex tasks and exhibits strong generalization ability and cross-domain adaptability. It is a typical example of "scale-driven intelligence enhancement" in the field of artificial intelligence. Its core characteristic is the combination of "large parameters, big data, and powerful computing power" to achieve deep understanding and generation of multimodal information such as language, images, and speech.

[0022] Large models include the following categories: (1) Large Language Model (LLM): Focused on natural language processing, it can understand and generate text and complete tasks such as chat, translation, summarization, and code generation.

[0023] (2) Large visual model: It processes image or video data and has the ability to recognize, generate and edit images.

[0024] (3) Multimodal large model: Simultaneously process multiple data such as text, image, and voice to achieve cross-modal understanding and generation (such as "picture description" and "generating video from text").

[0025] (4) Vertical domain large model: Based on the general large model, fine-tuned, focusing on specific fields (such as medical, financial and legal), with professional knowledge and skills.

[0026] For example, to achieve user intent recognition in the fields of text generation and video playback in this application, the large model can be fine-tuned to obtain the knowledge and skills of professionals in the fields of text generation and video playback, thereby focusing on the intent recognition of users watching text and / or video.

[0027] In view of the problems existing in the background art, the embodiments of this application provide a content processing method, a content processing apparatus, an electronic device, a computer-readable storage medium, and a computer program product.

[0028] To facilitate understanding, the application scenarios of this application will be introduced below: This application provides a content processing method, a content processing apparatus, an electronic device, a computer-readable storage medium, and a computer program product. Specifically, the content processing method of this application can be executed by an electronic device, which can be a terminal or a server. The terminal can include, but is not limited to, smartphones, tablets, laptops, smart TVs, wearable smart devices, smart vehicle terminals, etc. The terminal can also include a client, which can be a game client, a browser client, an instant messaging client, or a mini-program, etc. The server can be an independent physical server, a server cluster composed of multiple physical servers, or a distributed system. It can also be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0029] For example, when this content processing method runs on a terminal device, the terminal device may include a display screen and a processor. The display screen is used to present a content-generated screen and receive instructions from the user interacting with the content-generated screen. The processor is used to store a content-generating application, run the content-generating application, generate a content-generated screen, respond to instructions, and control the display of the content-generated screen on the display screen. When the user operates the content-generated screen through the display screen, the content-generated screen can control the local content of the terminal device in response to the received operation instructions. The terminal device can provide a graphical user interface to the user in various ways, such as rendering the interface on the terminal device's display screen or presenting the graphical user interface through holographic projection.

[0030] For example, if the content processing method runs on a server, it can be implemented and executed based on a cloud content generation system. A cloud content generation system refers to a content generation method based on cloud computing. A cloud content generation system includes servers and client devices. The main body running the content generation application and the main body presenting the generated content are separate. The storage and execution of the content processing method are completed on the server. The presentation of the generated content is completed on the client. The client is mainly used for receiving and sending content generation data and presenting the generated content. For example, the client can be a display device with data transmission capabilities located close to the user, such as a mobile terminal, television, computer, PDA, personal digital assistant, head-mounted display device, etc. However, the terminal device for processing the content generation data is the server in the cloud. During content generation, the user operates the client to send instructions to the server. The server controls the content generation process according to the instructions, encodes and compresses the generated content data, returns it to the client via the network, and finally, the client decodes and outputs the generated content.

[0031] It should be noted that, in this embodiment, the execution entity of the content processing method can be a terminal device or a server. The terminal device can be a local terminal device or the client device mentioned above in cloud content generation. This embodiment does not limit the type of execution entity.

[0032] It is understood that in the specific implementation of this application, user object data, context data and other related data are involved. When the embodiments of this application are applied to specific products or technologies, user permission or consent is required, and the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0033] For example, in conjunction with the above description, Figure 1This application illustrates a content generation system 100 for implementing a content processing method, as provided in an embodiment of this application. The content generation system 100 may include at least one terminal 10, at least one server 20, at least one database 30, and a network. The user-held terminal 10 can connect to different servers 20 via the network. The terminal 10 can be any device with computing hardware capable of supporting and executing software application tools corresponding to content generation.

[0034] In the aforementioned content generation system 100, terminal 10 is used to install and run the content generation application. In some cases, the content generation application may not need to be pre-installed on terminal 10; users can directly access the content generation application through a browser or other client. Users can log in to the content generation application using a registered account to control content generation. When a user logs in, terminal 10 sends a login request to server 20. Server 20 verifies the user's account and determines the content generation mechanism corresponding to the account based on the login request. If verification is successful, a login success notification is returned to terminal 10. During the content generation process, terminal 10 and server 20 interact, with terminal 10 sending various information to server 20. Server 20 determines the display data for terminal 10 based on the stored content generation mechanism and the received information, and sends the display data back to terminal 10 so that terminal 10 can display the data sent by server 20 to the user.

[0035] In possible application scenarios, different terminals 10 may be served by different servers 20. Therefore, in order to distinguish the servers 20 corresponding to different content-generating terminals 10, the embodiments of this application will use the terms "first" and "second" to describe them. In fact, the servers 20 corresponding to different content-generating terminals 10 can be the same server 20. Therefore, without distinguishing between "first" and "second", it can be understood that the terminals 10 corresponding to accounts located in the same content generation scenario are served by the same server 20.

[0036] Furthermore, in a content generation system 100 that includes multiple terminals, multiple servers, and multiple networks, different terminals can connect to each other through different networks and servers. The network can be a wireless network or a wired network; for example, wireless networks include Wi-Fi, LAN, cellular networks, 2G networks, 3G networks, 4G networks, and 5G networks. Additionally, different terminals can also connect to other terminals or servers using their own Bluetooth networks or hotspot networks. Furthermore, the system 100 can include multiple databases coupled to different servers, and can continuously store content generation-related information in the databases as different users generate content online.

[0037] It should be noted that in this embodiment, multiple terminal devices run the same content generation application. Therefore, data interaction between multiple terminal devices can be achieved through the content generation application's server. Thus, sending data from terminal device 1 to terminal device 2 can be understood as: terminal device 1 sends data to the content generation application's server, and the server sends the data to terminal device 2. Receiving data from terminal device 2 can be understood as: terminal device 1 receives data sent by the content generation application's server, which is the data sent by terminal device 2 to the server. Alternatively, there may be no content generation server, and terminal device 1 directly sends content generation-related data to terminal device 2.

[0038] It should be noted that, Figure 1 The schematic diagram of the content generation system shown is merely an example. The content generation system 100 described in this application embodiment is for the purpose of more clearly illustrating the technical solutions of this application embodiment and does not constitute a limitation on the technical solutions provided in this application embodiment. As those skilled in the art will know, with the evolution of content generation systems and the emergence of new business scenarios, the technical solutions provided in this application embodiment are also applicable to similar technical problems.

[0039] It should be noted that the triggering operations mentioned in the subsequent detailed description of the content processing methods provided in the embodiments of this application can all be regarded as triggering operations performed by the user through a finger or by controlling a medium such as a mouse, keyboard, or stylus. The specific medium used can be determined according to the type of electronic device. For example, when the electronic device is a touch screen device such as a mobile phone or tablet, the user can operate on the touch screen using any suitable object or accessory such as a finger or stylus. When the terminal device is a non-touch screen terminal device such as a desktop computer or laptop, the user can operate using an external device such as a mouse or keyboard.

[0040] The technical solution of this application will be described in detail below through specific embodiments. It should be noted that the following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments.

[0041] Based on the above introduction of technical terms and related scenarios, this application provides a content processing method, which will be described in detail below: Please refer to Figure 2 The content processing method provided in this application embodiment is implemented by steps 011, 012 and 013, which are described in detail below.

[0042] Step 011: Obtain the target template from the preset content template library; The preset content template library is a database that stores multiple preset templates. A preset template is a pre-defined overall framework used to constrain content generation. Each preset template includes at least one structured content module.

[0043] The structured content module is the component that carries specific content guidance. It includes structured content and its corresponding weights. The structured content is the information used to guide the output; the weights of the structured content are information indicating the proportion of that content in the total output content. The target template is a preset template selected from the preset content template library that is adapted to specific creative needs.

[0044] Specifically, a pre-set content template library uniformly stores and manages the preset templates needed for various application scenarios, facilitating quick querying and selection of target templates. From multiple preset templates in the library, a target template that suits the creative needs is selected. The target module, through pre-defined structured content modules, clearly defines the direction and proportion of its corresponding output content, standardizing the output content and improving the consistency and efficiency of content generation.

[0045] For example, please see Figure 3 A preset template is displayed through a graphical user interface. The preset template name allows users to quickly identify it. This preset template includes four structured content modules: Structured Content Module A, Structured Content Module B, Structured Content Module C, and Structured Content Module D. Structured Content Module A includes the corresponding weight A and structured content A; Structured Content Modules B, C, and D each contain their respective weights and structured content. This helps users quickly understand the template's structure and decide whether to use it as their target template.

[0046] For example, the default template is a four-part character profile template, which includes four structured content modules. The four structured content modules are: Character Introduction / Biography, Character Deeds / Entry into the World, Landmark Events / Growth, and Character Achievements / National Sentiment.

[0047] The structured content guidance output in the character opening / biography module includes character perception, background environment, and personality traits, which are used to establish a connection between character perception and emotion, with a corresponding weight of 20%. Among them, the structured content guidance output in the Character Deeds / Worldly Experiences module includes clear and intuitive worldly experiences and stage-by-stage growth, which is used to showcase the character's core abilities and enhance the character's charm, with a corresponding weight of 20%; The structured content guidance output in the landmark events / growth content module includes the character's highlight moments, at least one landmark event, and changes in the national situation, in order to enhance the character's appeal, with a corresponding weight of 50%. Among them, the structured content guidance output in the character achievement / national sentiment content includes character image freezing, hero shaping and national sentiment, which are used to explain the character's significance to the country and value to history, with a corresponding weight of 10%; Step 012: Generate metadata for the created content based on the initial content and target template; Step 013: Input the metadata of the created content into the preset large model, and output the target content that matches the target template by using the corresponding weights of the structured content modules in the target template and the guidance of the structured content.

[0048] The creative content refers to the final output. The initial content consists of the original materials or core ideas for the creation. For example, the initial content can be a short clip of material of a few hundred words, a summary of inspiration, or a description of the game setting. It does not necessarily require a complete story or script; providing only basic semantic context is also acceptable.

[0049] Metadata is key information describing the created content, integrating initial content and target templates. It includes the core requirements and structural constraints of the created content. The pre-set large model is a pre-trained large-scale AI model. This pre-set large model has the ability to output content based on the input metadata. For example, the pre-set large model is a large language model (LLM).

[0050] The target content is the final content generated by the pre-defined large model based on metadata, which meets the requirements of the target template.

[0051] Specifically, the initial content of the created content and the target template are used as metadata and input into a pre-set large model. The pre-set large model guides the output of the target content based on the structured content in each structured content module of the target template, reorganizes the text, and allocates more semantic space and narrative density to the high-weighted components according to the weight of each structured content and the proportion of the corresponding components in the total output target content.

[0052] For example, please see Figure 4 During the generation of target content, the graphical user interface displays the name of the created content, the name of the target template, a box for inputting the initial content, a generation control, and a box for the target content. The box for inputting the initial content is used to display the initial content of the created content in real time. After the initial content input is completed, the generation control is triggered, and the preset large model generates the target content based on the initial content and the target template, which is then displayed in real time in the box corresponding to the target content.

[0053] For example, the target template is a four-section character profile template, with the default large model being LLM. The initial content for creation is a description of Zhuge Liang: Zhuge Liang was a figure from the Three Kingdoms period. In his youth, he lived in seclusion in Longzhong and was known as the "Sleeping Dragon." Liu Bei visited him three times before he finally came out of seclusion to assist him, proposing the strategic plan of "dividing the world into three." Later, Zhuge Liang became the Chancellor of Shu Han, known for his loyalty, integrity, and hands-on approach. He launched several Northern Expeditions, though unsuccessful, leaving behind the historical evaluation of "dedicating himself to the cause until his dying breath." Folk tales such as "borrowing arrows with straw boats" and "the empty city stratagem" also highlight his wisdom.

[0054] The LLM model, based on a four-part character profile template and initial content, outputs four parts. The first part is the character's introduction / biography: In the Longzhong Mountains, the young Zhuge Liang studied and farmed, earning the title "Sleeping Dragon." He had the world in his heart, yet lived in seclusion. Liu Bei visited him three times before finally persuading him to join his cause. The second part is the character's deeds / entry into the world: Zhuge Liang outlined Liu Bei's blueprint for "dividing the world into three," assisting him in establishing himself in Jingzhou and conquering Yizhou, laying the foundation for Shu Han. His loyalty and wisdom made him an indispensable pillar of Shu Han. The third part is landmark events / growth: The ingenuity of borrowing arrows with straw boats, the composure of the Empty City Stratagem, the perseverance of six expeditions to Qishan… countless highlights shaped his image as a "wise sage." Despite repeated setbacks, he always devoted himself to the cause. The fourth part is the character's achievements / national sentiment: He died at Wuzhangyuan, his life written with loyalty and wisdom, praised by later generations as "dying only after fulfilling his duty." Zhuge Liang was not only a prime minister, but also a spiritual beacon in the memory of the nation.

[0055] In this way, the output target content is a combination of component content generated by various structured content in different proportions, which ensures that the output target content matches the target template, avoids deviations caused by free generation, effectively improves the standardization of the output target content, and ensures that the target content has a consistent structural rhythm and information density.

[0056] Optionally, the metadata includes localization information. This localization information is part of the created content and is used to assist the pre-defined large model in adapting the output target content to localization requirements. The localization information is obtained in the same way as the initial content, through user input. For example, localization information includes, but is not limited to: game background settings (such as worldview and historical context), character relationship networks (relationships between characters), cultural tags (such as traditional customs and values), folk tales, allusions, and symbolic elements. Please refer to [link / reference]. Figure 5 During the generation of target content, the graphical user interface also displays controls for localization enhancement. By triggering the controls for localization enhancement, users can input localization information in the corresponding boxes to help accurately generate target content that is adapted to the target region and local culture.

[0057] Meanwhile, the structured content in each module of the target template helps the pre-set large model clarify the creative direction, ensure content adaptability, avoid output deviating from requirements, improve output efficiency and quality, and reduce the cost of manual revisions. Furthermore, new users can directly use the target template to quickly get started and generate target content that meets their needs, lowering the learning curve.

[0058] In one optional embodiment, the metadata also includes regional information of the created content, a preset tag library and regional information matching, with different preset tag libraries set for different regions.

[0059] Among these, regional information refers to the geographical scope of the content being created. The preset tag library is a collection of tags categorized by region.

[0060] Specifically, the pre-set large model matches the corresponding pre-set tag library based on regional information, and makes the output content fit the tags in the pre-set tag library, thereby adapting to the cultural context, language habits, game world view and other elements of the corresponding region, improving the relevance of the target content, making the target content more easily accepted by the audience in the specific region, and adapting to the needs of localized promotion.

[0061] For example, the preset tag library for China may include: historical allusions, Confucian values, national sentiment, and aesthetic imagery; the preset tag library for Japan may include: samurai spirit, fatalism, and subtle emotional expression; and the preset tag library for Europe and America may include: heroism, free will, and personal growth narratives.

[0062] Optionally, the metadata may also include at least one of the following: identification information, type information, version information, and region information of the created content.

[0063] Identification information consists of symbols or codes used to uniquely identify creative content. This identification information ensures that creative content is accurately located during management and circulation, facilitating subsequent retrieval and traceability, and improving the efficiency of creative content management.

[0064] Type information defines the attribute category of the content. For example, the type information of the created content could be script, advertising copy, press release, etc. Type information can guide the pre-defined large model to match the corresponding text style and text specifications.

[0065] Version information records the iteration status of the created content. Version information clarifies the update stage of the created content, avoids version confusion, facilitates subsequent reproduction and auditing, and improves the traceability of the created content.

[0066] Specifically, the identification information, type information, version information, and regional information of the created content precisely constrain the generation direction of the preset large model from multiple dimensions, enhance localization and scene adaptability, and combine regional information to make the content more in line with the characteristics of the target audience, effectively avoiding the output target content from deviating from expectations.

[0067] In one alternative embodiment, please refer to Figure 6 The content processing method also includes step 014, which will be explained in detail below.

[0068] Step 014: In response to the first configuration operation on the structured content module in the preset template, configure the weight, first prompt word and second prompt word of the target structured content module, and store the configured preset template in the content template library.

[0069] The structured content module of the preset template includes a first prompt word and a second prompt word. The first prompt word is used to constrain the semantic intent and generation logic of the content output by the corresponding structured content module. The second prompt word includes multiple preset prompt words, which are used to supplement and constrain details such as content style and expression.

[0070] The first configuration operation involves adjusting and setting the weights, first prompt words, and second prompt words of the structured content modules. For example, the first configuration operation allows users to modify values ​​or input text using their fingers or by controlling a mouse, keyboard, or stylus.

[0071] Specifically, based on usage requirements, a new preset template can be generated by performing a first configuration operation on at least one structured content module in the selected preset template. This new preset template can meet the usage requirements of new application scenarios. Saving the configured preset template enables template reuse and reduces repetitive labor costs.

[0072] For example, please see Figure 7 During the modification or creation of preset templates, the graphical user interface displays basic information controls, structured content template controls, and preview and publish controls for the preset template. Triggering the basic information controls (gray fill indicates triggering) displays the basic information of the preset template. This basic information includes: the preset template's identifier, type, version, region, template description (briefly explaining the preset template), and applicable products (indicating the application products / scope to which the preset template is applicable).

[0073] For example, please see Figure 8 During the modification or creation of preset templates, the graphical user interface displays basic information controls, structured content template controls, and preview and publish controls for the preset template. Triggering a structured content template control (gray fill indicates triggering) displays the relevant content of the preset template's structured content template. The basic information controls corresponding to the thick outlines indicate that the preset template has completed the configuration of its basic information. The relevant content of the structured content template includes: total weight (to help determine if the sum of all weights is 100%), weight distribution of each structured content module, content flow (displaying the order in which the content guided by each structured content module is generated), structure visualization bar (displaying the proportion of the content guided by each structured content module in the target content), and target content structure (displaying the order of each structured content module, exemplarily represented in the diagram as A, B, C, and D in sequence).

[0074] For example, please see Figure 9 During the modification or creation of a preset template, the graphical user interface displays basic information controls for the preset template, structured content template controls, and preview and publish controls. Triggering the controls corresponding to structured content template A displays content related to structured content template A. The content related to structured content template A includes: first prompt word, second prompt word, third prompt word, fourth prompt word, and fifth prompt word, as well as a weight adjustment bar (used to adjust the weight A corresponding to structured content module A).

[0075] Alternatively, please continue reading Figure 9By triggering the control corresponding to structured content template A, the graphical user interface also displays the control corresponding to the newly added structured content template. By triggering the control corresponding to the newly added structured content template, a new structured content template can be added, along with the various prompts and their corresponding weights for the new template.

[0076] Optionally, please refer to Figure 6 The content processing method also includes steps 015 and 016, which are explained in detail below.

[0077] Step 015: Evaluate the match between the target content and the first prompt word; Step 016: If the matching degree is lower than the preset matching degree, re-output the target content based on the metadata.

[0078] The preset matching degree is a threshold set based on experience or a custom setting. For example, 90%.

[0079] Specifically, after the pre-defined large model outputs target content based on the metadata of the created content, the target content is evaluated. This evaluation can be achieved through methods such as semantic alignment, semantic similarity calculation, and keyword matching to calculate the matching degree between the target content and the first prompt word.

[0080] For example, the matching degree can be calculated using the embedding similarity calculation method in semantic similarity calculation. The calculation process includes: converting the target content and the first prompt word into high-dimensional vectors (or embedding vectors), calculating the cosine similarity or Euclidean distance between the two high-dimensional vectors (or embedding vectors), and then normalizing the transformation to obtain the matching degree between the target content and the first prompt word.

[0081] If the match between the target content and the first prompt word is higher than or equal to the preset match rate, it indicates that the output target content has met the expected standard and there is no need to re-output the target content. If the match between the target content and the first prompt word is lower than the preset match rate, it indicates that the output target content has not met the expected standard, and the preset large model needs to re-output the target content based on metadata until the match between the target content and the first prompt word is higher than or equal to the preset match rate, at which point output will stop.

[0082] For example, please see Figure 4 or Figure 5 During the generation of target content, the graphical user interface also displays controls for evaluating the matching degree. By triggering the controls for evaluating the matching degree, the matching degree between the generated target content and the first prompt word of the target template can be evaluated and displayed on the graphical user interface.

[0083] In this way, by evaluating the matching degree of the target content, we can ensure that the target content accurately responds to the core constraints, improve the reliability of the target content, and reduce the cost of manual revision.

[0084] In one alternative embodiment, please refer to Figure 6 The content processing method also includes step 017, which will be explained in detail below.

[0085] Step 017: In response to the second configuration operation on the structured content module in the preset template, configure the third prompt word and the fourth prompt word, and store the configured preset template in the content template library.

[0086] The structured content in the preset template's structured content module also includes a third prompt and a fourth prompt. The third prompt includes an example of content output that matches the corresponding structured content module; the fourth prompt is used to guide the style of the narration and / or subtitles in the output content of the corresponding structured content module.

[0087] For example, the third prompt includes: descriptive text, image links containing text, and video clip links containing narration or subtitles. The fourth prompt includes: solemn, lyrical, documentary, poetic expression, historical narrative, and promotional announcements.

[0088] The second configuration operation involves adjusting and setting the third and fourth prompts for the structured content module. For example, the second configuration operation allows the user to modify values ​​or input text using their finger, a mouse, keyboard, or stylus.

[0089] Specifically, the third cue word in the structured content provides examples that intuitively demonstrate the output standards, reducing the understanding cost of the pre-defined large model. This allows the pre-defined large model to more easily output target content that meets the expected needs based on the examples corresponding to the third cue word. The fourth cue word in the structured content is used to precisely constrain the narration / subtitle style of the target content, so that the target content can accurately adapt to the presentation requirements of multi-media content such as videos and advertisements.

[0090] Based on usage requirements, by performing a second configuration operation on at least one structured content module in the selected preset template, detailed configurations can be added. This allows the generated preset template to cover more specific scenarios, improving the flexibility of the preset template. Saving the configured preset template enables template reuse, reducing repetitive labor costs.

[0091] For example, please see Figure 9When modifying or adjusting the structured content modules of a preset template, after triggering structured content module A, the graphical user interface displays controls corresponding to the third and fourth prompt words for structured content module A. By triggering the controls corresponding to the third and fourth prompt words respectively, the third and fourth prompt words can be modified to achieve new configurations.

[0092] In one alternative embodiment, please refer to Figure 6 The content processing method also includes step 018, which will be explained in detail below.

[0093] Step 018: In response to the third configuration operation on the structured content module in the preset template, configure the fifth prompt word and store the configured preset template in the content template library.

[0094] The structured content also includes a fifth prompt word, which is used to guide the corresponding preset template output to match the content of the localized tags.

[0095] The third configuration operation is an interactive operation that adjusts and sets the fifth prompt word of the structured content in the structured content module of the preset template. For example, the third configuration operation is performed by the user through operations such as modifying values ​​or inputting text using a finger, mouse, keyboard, or stylus.

[0096] Localization tags are predefined keywords used to identify the cultural customs, language habits, and other characteristics of a specific region. These tags help the pre-defined large-scale model quickly determine key regional information, improving the accuracy of content localization.

[0097] In one alternative embodiment, please refer to Figure 10 Localization tags include at least one of the following: historical tags, style tags, color tags, and emotion tags.

[0098] Historical tags are used to identify the historical background, traditions, or iconic historical elements of a specific region. Examples include tags related to turbulent times, heroic epics, national sentiment, and ancient capital culture. Historical tags are used to integrate the target content with the region's historical heritage.

[0099] Style tags define the language expression, aesthetic preferences, or content presentation style of a specific region. Examples include epic narrative, realism, Eastern aesthetics, the delicate style of Jiangnan, and the bold and unrestrained expression of Northwest China. Style tags ensure that the style of the output content aligns with the habits of the regional audience.

[0100] Among them, role-based tags are labels that point to the identity, occupation, or social role of typical people in a specific region. For example, wise man, monarch, wanderer, farmer, impoverished nobleman, etc. Role-based tags are used to make the characters or perspectives in the output content more relevant to the target regional group.

[0101] Among these, emotion-related tags are those that resonate with the collective emotions and values ​​of an audience in a specific region. Examples include sacrifice, bond, a sense of destiny, everyday life, and a sense of belonging to one's hometown. Emotion-related tags are used to ensure that the content delivered evokes emotional resonance with the target audience in that region.

[0102] Specifically, different types of localization tags help serve as a reference for the pre-set model, ensuring that the output content is as closely aligned with the culture, customs, and other characteristics of the target audience as possible. Based on the needs of different regional audiences, by performing a third-party configuration operation on at least one structured content module in the selected pre-set template, detailed configurations can be added. This allows the generated new pre-set templates to adapt to the promotion and dissemination needs of different regions, enhancing the flexibility of the pre-set templates. Saving the configured pre-set templates enables template reuse, reducing repetitive labor costs. The third-party configuration operation also reduces the cost of content modification due to regional differences, improving generation efficiency.

[0103] In one alternative embodiment, please refer to Figure 11 Step 018 includes: Step 0181: In response to the third configuration operation on the structured content module in the preset template, configure the localized tags corresponding to at least one of the worldview, character relationships and narrative context, and store the configured preset template in the content template library.

[0104] Specifically, the fifth prompt word includes at least one of worldview, character relationships, and narrative context. Through the third configuration operation of the structured content module in the preset template, diverse worldviews, complex character relationships, and / or diverse narrative contexts can be adjusted and / or set to adapt to different usage scenarios, improving the flexibility of the preset template. Saving the configured preset template enables template reuse, reducing repetitive labor costs. This allows the preset large model to output target content deeply adapted to the usage scenario based on the worldview, character relationships, and narrative context corresponding to the fifth prompt word in the configured preset template.

[0105] For example, please see Figure 12When the structured content module of the preset template is modified or adjusted, after triggering structured content module A, the graphical user interface displays the control corresponding to the fifth prompt word of structured content module A. By triggering the control corresponding to the fifth prompt word (gray fill), localized tag libraries corresponding to the worldview, character relationships, and narrative context can be displayed. Localized tags related to the worldview required by the preset template (such as the chaotic setting of the end of a dynasty) can be configured in the worldview tag library, localized tags related to character relationships required by the preset template (such as the Shu Han camp, the Three Visits to the Thatched Cottage, the master-disciple relationship, etc.) can be configured in the character relationship tag library, and localized tags related to the narrative context required by the preset template (such as national sentiment, loyalty and strategy, etc.) can be configured in the narrative context tag library.

[0106] For example, if the target template is a four-part character profile template with reconfigured fifth cue words, the preset large model is LLM., and the initial content of the creation is a description of Zhuge Liang, the four-part character profile template with reconfigured fifth cue words reconfigures the new worldview (the end of the dynasty), character relationships (relationship with Liu Bei, etc.), and narrative context (loyalty and strategy) with their respective localized tags.

[0107] The LLM, based on a reconfigured four-part character profile template with the fifth prompt and initial content, outputs target content comprising four parts. The first part is the character's introduction / biography: In the world of the late Eastern Han Dynasty, heroes rose up, and the world fell apart. The young Zhuge Liang lived in seclusion in Longzhong, his room filled with the sound of reading, earning him the title "Sleeping Dragon." Though living in the mountains, he had already foreseen the chaotic world. Liu Bei visited him three times before finally inviting him to join his cause. The second part is the character's deeds / worldly involvement: He forged deep friendships with Liu Bei, Guan Yu, and Zhang Fei, proposed the strategy of "dividing the world into three," and assisted in the establishment of the Shu Han state. From military planning to battlefield strategy, he always upheld loyalty and righteousness. The third part is landmark events / growth: He became famous for his ingenious "borrowing arrows with straw boats," his courage in the "empty city stratagem," and his perseverance in the "six expeditions to Qishan." Although he suffered repeated defeats and failed to unify the Central Plains, he earned the title of "Sage of Wisdom" through his dedication and tireless efforts. The fourth part is about his achievements and national sentiments: He died at Wuzhangyuan, where he transformed his loyalty and wisdom into eternity, becoming a model revered by later generations. Inventions such as the Eight Trigrams Formation and the Wooden Ox and Flowing Horse are closely linked to folk legends and have been continuously portrayed in literature and opera, forming the typical image of "wisdom and loyalty".

[0108] In one alternative embodiment, please refer to Figure 13 Step 018 includes: Step 0182: Select a target tag from the preset tag library as the fifth prompt word; and / or Step 0183: In response to the input action, generate a custom tag as the fifth prompt word.

[0109] Among these, target tags are localized tags adapted to creative needs. A preset database can store pre-configured localized tags. Input operations are interactive actions that generate custom localized tags. For example, the third configuration operation allows users to modify values ​​or input text using their fingers or by controlling a mouse, keyboard, or stylus.

[0110] Specifically, users can select preset tags from the preset tag library that match the creation scenario as target tags to help the preset large model enhance the localization of the output target content. If none of the preset tags in the preset tag library meet the creation scenario, users can add custom tags through input operations. Custom tags can more accurately adapt to the creation scenario, helping the preset large model output more precisely localized target content.

[0111] Optionally, the custom label generated in response to the input operation can be stored in the preset label library as a preset label so that the label can be reused later, reducing the cost of repetitive labor.

[0112] For example, please see Figure 14 When modifying or adjusting the structured content module of the preset template, after triggering structured content module A, the graphical user interface displays the control corresponding to the fifth prompt word for structured content module A. Triggering the control corresponding to the fifth prompt word displays a preset tag library (one or more, two in the example image) and custom tags. Triggering preset tag library 1 or 2 allows selection of the desired target tag as the fifth prompt word; triggering custom tags allows adding new localized tags as the fifth prompt word; and combining the selection and addition of custom tags from the preset tag library allows the resulting localized tag to be used as the fifth prompt word.

[0113] Optionally, please refer to Figure 15 Once the basic information and structured content template of the preset template have been configured (indicated by a thick border), the preview and publishing process of the configured preset template is performed (the preview and publishing controls are filled with gray). In this case, the graphical user interface displays the basic information, structural design (such as the content generated by each structured content module according to its weight), and composition of the structured content modules of the preset template, providing an overview of the preset template and allowing for a quick determination of whether to complete the creation or readjust the preset template. After the preset template is created, it is stored in the preset content template library for subsequent retrieval and retrieval.

[0114] All of the above technical solutions can be combined in any way to form optional embodiments of this application, and will not be described in detail here.

[0115] To facilitate better implementation of the content processing method of this application, this application also provides a content processing apparatus for executing the steps in the above-described content processing method. Please refer to... Figure 16 , Figure 16 This is a schematic diagram of the modules of the content processing apparatus 200 provided in an embodiment of this application. The content processing apparatus 200 may include: The acquisition module 201 is used to acquire the target template in the preset content template library. The content template library includes multiple preset templates. Each preset template includes at least one structured content module. The structured content module includes structured content and corresponding weights. The structured content is used to guide the content information of the output, and the weights corresponding to the structured content are used to guide the proportion of the output content to the total output content. The generation module 202 is used to generate metadata for the created content based on the initial content and the target template. The input / output module 203 is used to input the metadata of the created content into a preset large model, and output target content that matches the target template through the corresponding weights of the structured content module in the target template and the guidance of the structured content.

[0116] Each unit in the aforementioned content processing device 200 can be implemented entirely or partially through software, hardware, or a combination thereof. Each unit can be embedded in or independent of the processor in the electronic device in hardware form, or stored in the memory of the electronic device in software form, so that the processor can call and execute the operations corresponding to each unit.

[0117] The content processing device 200 can be integrated into a terminal or server that has storage and a processor and thus computing power, or the content processing device 200 can be the terminal or server.

[0118] This application also provides an electronic device, including a processor and a memory. The memory stores a computer program. The processor executes various processes of the above-described content processing method embodiments by calling the computer program stored in the memory, and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0119] In an optional embodiment, the electronic device further includes a display screen. The display screen can be used to display a graphical user interface (GUI) and receive operation commands generated by the user interacting with the GUI. The display screen may include a display panel and a touch panel. The display panel can be used to display information input by the user or information provided to the user, as well as various GUIs of the electronic device, which can be composed of graphics, text, icons, video, and any combination thereof. The touch panel can be used to collect touch operations performed by the user on or near it (e.g., operations performed by the user using a finger, stylus, or any suitable object or accessory on or near the touch panel), and generate corresponding operation commands, which execute corresponding programs. Optionally, the touch panel may include two parts: a touch detection device and a touch controller. The touch detection device detects the user's touch location and the signal generated by the touch operation, transmitting the signal to the touch controller; the touch controller receives touch information from the touch detection device, converts it into touch point coordinates, sends it to the processor, and can receive and execute commands from the processor. The touch panel can cover the display panel. When the touch panel detects a touch operation on or near it, it transmits the information to the processor to determine the type of touch event. Subsequently, the processor provides corresponding visual output on the display panel based on the type of touch event. In this embodiment, the touch panel and the display panel can be integrated into the display screen to achieve input and output functions. However, in some embodiments, the touch panel and the display panel can be implemented as two independent components to achieve input and output functions.

[0120] In one alternative embodiment, please refer to Figure 17 , Figure 17 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. The electronic device 300 includes a processor 301 and a memory 302. The memory 302 stores a computer program 303 that can run on the processor 301. When the computer program 303 is executed by the processor 301, it implements the various processes of the embodiments of the above-described content processing method and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0121] Please see Figure 18 , Figure 18 This is a schematic diagram of the hardware structure of an electronic device provided in an embodiment of this application. The electronic device can be a terminal or a server. Exemplarily, the electronic device 400 includes a central processing unit (CPU) 401, a system memory 404 including random access memory (RAM) 402 and read-only memory (ROM) 403, and a system bus 405 connecting the system memory 404 and the central processing unit 401.

[0122] In some embodiments, the electronic device 400 may also include a basic input / output system 406 that helps transmit information between various devices within the computer, and a mass storage device 407 for storing the operating system 413, the client 414, and other program modules 415.

[0123] In some embodiments, the basic input / output system 406 includes a display 408 for displaying information and an input device 409 for user input, such as a touch panel and other input devices. A touch panel is also called a touchscreen. A touch panel may include both a touch device and a touch controller. Other input devices may include, but are not limited to, physical keyboards, function keys (such as volume control buttons, power buttons, etc.), trackballs, mice, and joysticks, which will not be described further here.

[0124] Both the display 408 and the input device 409 are connected to the central processing unit 401 via an input / output controller 410 connected to the system bus 405. The basic input / output system 406 may also include the input / output controller 410 for receiving and processing input from touch panels, other input devices, etc. Optionally, the input / output system 406 may also include output devices, such as displays, printers, or other types of output devices.

[0125] Mass storage device 407 is connected to central processing unit 401 via a mass storage controller (not shown) connected to system bus 405. Mass storage device 407 and its associated computer-readable media provide non-volatile storage for electronic device 400. That is, mass storage device 407 may include computer-readable media (not shown) such as hard disk or compact disc read-only memory (CD-ROM) drive.

[0126] According to various embodiments of this application, the electronic device 400 can also be connected to a remote computer on a network, such as the Internet. That is, the electronic device 400 can be connected to a network 417 via a network interface unit 416 connected to a system bus 405, or the network interface unit 416 can be used to connect to other types of networks or remote computer systems (not shown).

[0127] This application also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the various processes of the above-described content processing method embodiments and achieves the same technical effect. To avoid repetition, it will not be described again here.

[0128] The processor can be the processor in the electronic device described in the above embodiments. The computer-readable storage medium can be a computer read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk, etc.

[0129] Computer-readable media can include computer storage media and communication media. Computer storage media includes volatile and non-volatile, removable and non-removable media implemented by any method or technology for storing information such as computer-readable instructions, data structures, program modules, or other data. Computer storage media include RAM, ROM, erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other solid-state storage technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape cassettes, magnetic tape, disk storage, or other magnetic storage devices. Of course, those skilled in the art will recognize that computer storage media are not limited to the above-mentioned types.

[0130] This application also provides a computer program product, including computer instructions that, when executed by a processor, implement the above-described content processing method. The processor may be a processor in the electronic device described above. When executed by the processor, the computer instructions implement various processes of the embodiments of the above-described content processing method and achieve the same technical effects; therefore, to avoid repetition, they will not be described again here.

[0131] It is understood that in the specific implementation of this application, data related to user identity or characteristics is involved. When the above embodiments of this application are applied to specific products or technologies, user permission or consent is required, and the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0132] In the description of this specification, the references to terms such as "certain embodiments," "an alternative embodiment," and "exemplarily" indicate that a specific feature, structure, material, or characteristic described in connection with an embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0133] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing a particular logical function or process, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order according to the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.

[0134] Although embodiments of this application have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of this application, the scope of which is defined by the claims and their equivalents.

Claims

1. A content processing method, characterized in that, include: Obtain a target template from a preset content template library. The content template library includes multiple preset templates. Each preset template includes at least one structured content module. Each structured content module includes structured content and corresponding weights. The structured content is used to guide the output content information, and the weights corresponding to the structured content are used to guide the proportion of the output content to the total output content. Metadata of the created content is generated based on the initial content of the created content and the target template; The metadata of the created content is input into a preset large model. Guided by the corresponding weights of the structured content modules in the target template and the structured content, the target content that matches the target template is output.

2. The content processing method according to claim 1, characterized in that, The structured content includes a first prompt word and a second prompt word. The first prompt word is used to constrain the semantic intent and generation logic of the content output by the corresponding structured content module. The second prompt word includes multiple preset prompt words. The method further includes: In response to a first configuration operation on the structured content module in the preset template, the weight, first prompt word, and second prompt word of the target structured content module are configured, and the configured preset template is stored in the content template library.

3. The content processing method according to claim 2, characterized in that, Also includes: Evaluate the matching degree between the target content and the first prompt word; If the matching degree is lower than the preset matching degree, the target content is re-output based on the metadata.

4. The content processing method according to claim 1 or 2, characterized in that, The structured content further includes a third prompt word and a fourth prompt word. The third prompt word includes a content output example matching the corresponding structured content module. The fourth prompt word is used to guide the style of the narration and / or subtitles of the content output by the corresponding structured content module. The method further includes: In response to a second configuration operation on the structured content module in the preset template, the third and fourth prompt words are configured, and the configured preset template is stored in the content template library.

5. The content processing method according to any one of claims 1-4, characterized in that, The structured content also includes a fifth prompt word, which guides the corresponding preset template to output content that matches the localized tags. The method further includes: In response to a third configuration operation on the structured content module in the preset template, the fifth prompt word is configured, and the configured preset template is stored in the content template library.

6. The content processing method according to claim 5, characterized in that, The fifth prompt word includes at least one of worldview, character relationships, and narrative context. The third configuration operation in response to the structured content module in the preset template, to configure the fifth prompt word and store the configured preset template in the content template library, includes: In response to a third configuration operation on the structured content module in the preset template, localized tags corresponding to at least one of the worldview, character relationships, and narrative context are configured, and the configured preset template is stored in the content template library.

7. The content processing method according to claim 5 or 6, characterized in that, The configuration of the fifth prompt word includes: Select a target tag from a preset tag library as the fifth prompt word; and / or In response to the input action, a custom tag is generated as the fifth prompt word.

8. The content processing method according to claim 7, characterized in that, The metadata also includes the regional information of the created content. The preset tag library matches the regional information, and different preset tag libraries are set for different regions.

9. The content processing method according to claim 5 or 6, characterized in that, The localized tags include at least one of the following: historical tags, style tags, color tags, and emotion tags.

10. The content processing method according to claim 1, characterized in that, The metadata also includes at least one of the following: identification information, type information, version information, and region information of the created content.

11. A content processing apparatus, characterized in that, include: The acquisition module is used to acquire a target template from a preset content template library. The content template library includes multiple preset templates. Each preset template includes at least one structured content module. The structured content module includes structured content and corresponding weights. The structured content is used to guide the output content information, and the weights corresponding to the structured content are used to guide the proportion of the output content to the total output content. A generation module is used to generate metadata for the created content based on the initial content of the created content and the target template; The input / output module is used to input the metadata of the created content into a preset large model, and output target content that matches the target template through the corresponding weights of the structured content module in the target template and the guidance of the structured content.

12. An electronic device, characterized in that, It includes a processor and a memory, wherein the memory stores a computer program, and the processor executes the content processing method as described in any one of claims 1-10 by calling the computer program stored in the memory.

13. A computer-readable storage medium, characterized in that, The device stores a computer program adapted for loading by a processor to perform the content processing method as described in any one of claims 1-10.

14. A computer program product, characterized in that, It includes computer instructions that, when executed by a processor, implement the content processing method as described in any one of claims 1-10.

Images