Content generation method and device, equipment, storage medium and program product

Abstract

The invention provides a content generation method and device, equipment, a storage medium and a program product, and relates to the technical field of artificial intelligence such as natural language processing, long text generation and automatic writing. The method comprises the following steps: acquiring a generation request for indicating to generate at least two target content chapters; determining a first content chapter and a second content chapter from the target content chapters; respectively and correspondingly generating and outputting first generation content corresponding to the first content chapter and second generation content corresponding to the second content chapter by adopting a first mode and a second mode in parallel; in response to completion of generation of the first generation content, obtaining a generation progress of the second generation content; in response to the generation progress indicating that the second generation content is generated completely, the second generation content is output, and the first mode indicates that the generated first generation content is output in real time; the second mode indicates that the second generated content is output after the second generated content is generated. Therefore, the overall quality of the generated content can be improved while the real-time performance is considered.

Description

Technical Field

[0001] This disclosure relates to the field of computer technology, specifically to artificial intelligence technologies such as natural language processing, long text generation, and automated writing, and particularly to content generation methods, devices, electronic devices, computer-readable storage media, and computer program products. Background Technology

[0002] Against the backdrop of the rapid development of the Internet and artificial intelligence technologies and the explosive growth of digital content, automatic content generation technology has emerged. It uses computer algorithms and intelligent models to achieve intelligent creation, replacing or assisting manual work, and quickly generating texts that are coherent, reasonable, and targeted.

[0003] Therefore, how to provide content generation services efficiently and with high quality, and improve the user experience of using content generation services, is a matter of concern and urgent need. Summary of the Invention

[0004] This disclosure provides a content generation method, apparatus, electronic device, computer-readable storage medium, and computer program product.

[0005] In a first aspect, embodiments of this disclosure propose a content generation method, comprising: obtaining a generation request, the generation request indicating the generation of at least two target content chapters; determining a first content chapter and a second content chapter from the target content chapters; employing a first method and a second method in parallel to generate and output first generated content corresponding to the first content chapter and second generated content corresponding to the second content chapter, respectively; in response to the completion of generating the first generated content, obtaining the generation progress of the second generated content; and in response to the generation progress indicating the completion of generating the second generated content, outputting the second generated content; wherein, the first method indicates that the generated first generated content is output in real time during the generation of the first generated content; and the second method indicates that the second generated content is output after the generation of the second generated content is completed.

[0006] Secondly, embodiments of this disclosure propose a content generation apparatus, comprising: a generation request acquisition unit configured to acquire a generation request, the generation request indicating the generation of at least two target content chapters; a content chapter determination unit configured to determine a first content chapter and a second content chapter from the target content chapters; a content generation and output unit configured to concurrently employ a first method and a second method to generate and output first generated content corresponding to the first content chapter and second generated content corresponding to the second content chapter, respectively; a progress acquisition unit configured to acquire the generation progress of the second generated content in response to the completion of the generation of the first generated content; and a content output unit configured to output the second generated content in response to the generation progress indication that the second generated content has been generated; wherein the first method indicates that the generated first generated content is output in real time during the generation of the first generated content; and the second method indicates that the second generated content is output after the generation of the second generated content is completed.

[0007] Thirdly, embodiments of this disclosure provide an electronic device, the electronic device comprising: 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, the instructions being executed by the at least one processor to enable the at least one processor to implement the content generation method as described in any implementation of the first aspect.

[0008] Fourthly, embodiments of this disclosure provide a non-transitory computer-readable storage medium storing computer instructions that enable a computer to implement the content generation method described in any implementation of the first aspect when executed.

[0009] Fifthly, embodiments of this disclosure provide a computer program product including a computer program that, when executed by a processor, can implement the content generation method as described in any implementation of the first aspect.

[0010] The content generation method, apparatus, electronic device, computer-readable storage medium, and computer program product provided in this disclosure firstly obtain a generation request, the generation request indicating the generation of at least two target content chapters; then, determine a first content chapter and a second content chapter from the target content chapters; then, employ a first method and a second method in parallel to generate and output first generated content corresponding to the first content chapter and second generated content corresponding to the second content chapter, respectively; then, in response to the completion of generating the first generated content, obtain the generation progress of the second generated content; finally, in response to the generation progress indicating that the second generated content has been generated, output the second generated content; wherein, the first method indicates that the generated first generated content is output in real time during the generation of the first generated content; the second method indicates that the second generated content is output after it has been generated.

[0011] This disclosure enables the use of both a first and a second method in parallel, so that during the process of processing the generation request and the generation content, at least a portion of the generated content can be fed back to the user in real time using the first generated content to meet the user's requirements for the real-time presentation of the generated content, while the overall generation quality of the generated content can be ensured by using the second generated content, which is generated completely in an "overall generation method" and takes into account more of the context of the generated content.

[0012] 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

[0013] 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 content generation process provided in this embodiment of the disclosure; Figure 3 A flowchart illustrating a process for generating second generated content, provided as an embodiment of this disclosure; Figure 4 A flowchart illustrating a specific implementation of a content generation process in an application scenario, provided by an embodiment of this disclosure; Figure 5 A structural block diagram of a content generation apparatus provided in an embodiment of this disclosure; Figure 6This is a schematic diagram of the structure of an electronic device suitable for performing a content generation method, provided as an embodiment of this disclosure. Detailed Implementation

[0014] 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.

[0015] Furthermore, the acquisition, storage, use, processing, transportation, provision, and disclosure of user personal information involved in the technical solutions disclosed herein comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0016] Figure 1 An exemplary system architecture 100 is shown, to which embodiments of the content generation methods, apparatus, electronic devices, and computer-readable storage media of this disclosure can be applied.

[0017] like Figure 1 As 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.

[0018] 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 content generation applications, online Q&A applications, and instant messaging applications.

[0019] 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.

[0020] Server 105 can provide various services through its built-in applications. Taking a content generation application as an example, which provides services to generate content based on user-provided generation requests, server 105 can achieve the following effects when running this application: First, server 105 can obtain generation requests from terminal devices 101, 102, and 103 via network 104. These generation requests indicate the generation of at least two target content chapters. Then, server 105 determines a first content chapter and a second content chapter from the target content chapters. Next, server 105 uses a first method and a second method in parallel to generate and output the first generated content corresponding to the first content chapter and the second generated content corresponding to the second content chapter, respectively. The first method indicates that the generated first content is output in real-time during its generation; the second method indicates that the second generated content is output after its generation is complete. Then, in response to the completion of the first generated content generation, server 105 obtains the generation progress of the second generated content. Finally, in response to the generation progress indicating that the second generated content has been generated, server 105 outputs the second generated content.

[0021] It should be noted that, in addition to being obtained from terminal devices 101, 102, and 103 via network 104, the generation requests can also be pre-stored locally on server 105 through various means. Therefore, when server 105 detects that this data is already stored locally (e.g., when starting to process a previously stored content generation task), it can choose to directly retrieve this data from locally. In this case, the exemplary system architecture 100 may also exclude terminal devices 101, 102, and 103 and network 104.

[0022] Since generating content chapters and generating content requires significant computing resources and strong computing power, the content 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 content generation device is also generally located in 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 perform the aforementioned calculations performed by the server 105 through content 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 content generation application determines that the terminal device it is on 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 content generation device can also be located in the terminal devices 101, 102, and 103. In this case, the exemplary system architecture 100 may also exclude the server 105 and the network 104.

[0023] 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.

[0024] Please refer to Figure 2 , Figure 2 A flowchart of a content generation process provided for an embodiment of this disclosure includes process 200.

[0025] Process 200 specifically includes the following steps: Step 201: Obtain the generation request; In embodiments of this disclosure, the execution entity of the content generation method (e.g. Figure 1 The server 105 shown can receive instructions to generate at least two target chapter contents in a generation request.

[0026] For example, the generation request can be described by the user in natural language through terminal devices 101, 102, and 103. For example, it may be to generate content that consists of at least two target content chapters and can meet the content requirements (e.g., to generate a document or image consisting of at least two target content chapters).

[0027] In some embodiments, content requirements may include the content type, language style, keywords, subject terms, and other information that should be included in the generated content, as well as requirements for the language form of the content.

[0028] Upon receiving the generation request, the executing entity can parse the request to plan a content outline that includes at least two chapters. For example, after obtaining the generation request, the executing entity can construct a content outline including at least two target content chapters by breaking down themes, determining arguments, logically grouping, and establishing a hierarchical structure.

[0029] In other words, the executing entity can use natural language processing technology to analyze key information in the user's intent, such as writing purpose and topic direction, while extracting core content and knowledge points from the document information. Then, based on this integrated information, a content outline is generated.

[0030] This content outline allows the implementing entity to generate content for each target content chapter by dividing it into chapters, and then combine the content from these target content chapters using the content outline to obtain the complete content that satisfies the generation request.

[0031] For example, the executing entity can divide the content into content chapter 1 and content chapter 2. After generating the content of content chapter 1 and content chapter 2 respectively through the subsequent steps of this article, the content of content chapter 1 and content chapter 2 are arranged according to the order and composition of the content chapters in the content outline to form complete content that satisfies the generation request.

[0032] It should be understood that in some scenarios, if the generation request corresponds to a low generation difficulty and a small amount of content, making it impossible to divide the content outline, or if the user instructs that they do not want to divide the content outline, at least two (target) content chapters, then in such cases, the executing entity can complete the generation action by utilizing one of the methods of generating the first or second generated content, as discussed later. This article will not discuss such situations in detail.

[0033] Correspondingly, in some scenarios, since the content generation process provided by the embodiments of this application depends on at least two target content chapters, the content generation process provided by the embodiments of this application can also be understood as a "long requirement" or "long text" generation scenario in some scenarios.

[0034] Step 202: Determine the first content chapter and the second content chapter from the target content chapters; In the embodiments of this disclosure, based on step 201, this step aims to have the aforementioned executing entity divide the content chapter into a first content chapter and a second content chapter, so as to use different methods and strategies to generate the first content chapter and the second content chapter in parallel, so that they can be generated in different ways and strategies to meet different generation preferences and needs. For example, the first content chapter can be used to meet the user's need for "real-time viewing of the presented content" during the generation process, while the second content chapter can be used to meet the user's need for polishing the overall generated content and ensuring overall quality.

[0035] In some embodiments, the executing entity may distinguish and divide the first content chapter and the second content chapter based on factors such as the importance of the content chapter and the amount of content it includes. For example, the executing entity may designate content chapters that are less related to the "core content" as the first content chapter, and content chapters that are more related to the "core content" as the second content chapter.

[0036] Step 203: Using the first method and the second method in parallel, generate and output the first generated content corresponding to the first content chapter and the second generated content corresponding to the second content chapter, respectively; In the embodiments of this disclosure, based on step 201, this step aims to divide the content chapters into a first content chapter and a second content chapter by the aforementioned executing entity, and then process the content generation tasks of these content chapters in parallel based on the division results. For example, in a scenario where content chapter 1 and content chapter 2 are divided, the executing entity can use content chapter 1 as the first content chapter and content chapter 2 as the second content chapter.

[0037] Then, the executing entity can generate specific content in the corresponding content chapter by calling, for example, a content generation model (e.g., a large language model, a pre-trained agent with generative capabilities).

[0038] A Large Language Model (LLM) is an artificial intelligence model designed to understand and generate human language. Based on its understanding, an LLM can perform processing operations to obtain corresponding results. For example, it can be invoked by an agent to perform the task of generating specific content within a given "content section" based on prompts about generating that specific content.

[0039] LLMs can be trained on large amounts of text data and perform a wide range of tasks, including text summarization, translation, sentiment analysis, and more. A key characteristic of LLMs is their large scale; they typically include a large number of parameters to help them learn complex patterns in language data. These models are often based on deep learning architectures, such as transformers, which contributes to their superior performance across various natural language processing tasks.

[0040] An intelligent agent is a system or entity capable of perceiving its environment and autonomously making decisions and executing tasks. It can make judgments and choices based on its own goals and changes in the external environment, and take actions to achieve a specific purpose. For example, an intelligent agent can be an aggregation of one or more different models, trained using training samples corresponding to its intended use. For instance, regarding the task of generating specific content within a given "content chapter" as input, the intelligent agent can be trained using sample content chapters and corresponding sample generated content. This allows it to, for example, process the "content chapter," determine the semantics of the sample question information and the content it should include, match relevant reference knowledge from a knowledge base related to that semantics and the content it should include, and generate specific content within the "content chapter" based on the combination and rewriting of that reference knowledge.

[0041] In this step, the division into a first content chapter and a second content chapter allows the executing entity to perform the content chapter generation actions in parallel via a "two-way branch." That is, the executing entity can generate the content of content chapter 1 while simultaneously performing the content generation actions for content chapter 2.

[0042] For example, the executing entity can call a generation thread A to generate the content of content chapter 1 (for example, the generation thread can be used to perform actions that call LLM or the agent), and use another generation thread B, which is different from the generation thread A, to generate the content of content chapter 2, so as to achieve the "parallelism" of the two processes, so that content chapter 1 and content chapter 2 can be executed simultaneously.

[0043] Accordingly, in this step, the distinction between the first and second content chapters not only allows them to operate in parallel, enabling faster content generation, but also allows them to implement different generation and display strategies. That is, the generation process of the first content chapter can be executed in a first manner, while the generation process of the second content chapter can be executed in a second manner. The first manner indicates that the generated first content is output in real time during its generation; the second manner indicates that the second content is output after its generation is complete.

[0044] That is, during the process of generating the first content using the first method, the executing entity can choose to synchronously and in real-time output and display the currently generated content elements that make up the first content. For example, if the first content is text, the content element can be a specific character used to compose the first content. For example, each time such a content element is generated, the executing entity can choose to present it directly to improve the efficiency of user feedback and response. For example, the executing entity can directly provide it to the decoding and presentation module, so that the decoding and presentation module can directly output the character or pixel block to the user (e.g., provide it to terminal devices 101, 102, and 103, so that terminal devices 101, 102, and 103 can present the currently generated character or pixel block to the user in real-time and continuously).

[0045] In some embodiments, during the generation of the first generated content, a cyclical "predict the next token" mechanism may be used, so that after each generation of some content elements (e.g., content fragments of a certain number of characters), these contents can be provided to the user immediately, while the background continues to process the remaining contents, thereby ensuring that the user can quickly obtain these content elements and at least part of the results of the first generated content.

[0046] Regarding the process of generating the second generated content, the executing entity may choose to present and display the complete second generated content to the user only once after it has been fully generated, rather than continuously providing and presenting newly generated content elements during the generation process.

[0047] In some embodiments, the processes of generating the first generated content and generating the second generated content can have independent execution environments and resource allocation mechanisms, and can be run in parallel through multi-threading or multi-processing techniques. During the setup process, corresponding system resources, such as memory and CPU time slices, can be allocated to each generation process to ensure that the two sequences can run efficiently and stably without interfering with each other.

[0048] In this way, the executing entity can generate the first content so that the user can receive the feedback content generated by the executing entity in response to the generation request earlier. This not only improves the response and feedback speed of the executing entity, but also improves the overall content generation quality through the second generated content, which is allowed to be fed back more slowly and completely (because such second generated content can be fed back completely, so the overall quality of the second generated content can be improved by proofreading and polishing it as a whole within and between content chapters after the generation of each content element).

[0049] It should be understood that, based on different scenario requirements, the executing entity can selectively choose the length of the "content elements" to be presented. For example, if the generation speed is sufficient, the executing entity can choose to present them after generating X characters. This allows users to continuously obtain content in sentences or paragraphs, while also ensuring a certain degree of error correction and continuity adjustment capabilities between these characters.

[0050] Step 204: In response to the completion of generating the first content, obtain the generation progress of the second content; In the embodiments of this disclosure, based on step 203 above, if the first generated content has been generated, that is, the first generated content has been generated completely, for example, the content of content chapter 1 has been generated completely, then in this case, the executing entity can obtain the generation progress of the second generated content to determine whether it is necessary to continue to use the new content chapter as the first content chapter and continue to generate the first generated content in the first manner, so as to keep the user able to continuously receive feedback.

[0051] Next, if the execution entity determines and obtains the generation progress indicator that the second generated content has been generated completely, that is, it has also been generated completely, the execution entity can respond to this and choose to continue executing step 205.

[0052] Step 205: In response to the generation progress indication that the second generated content has been generated, output the second generated content.

[0053] In the embodiments of this disclosure, if the execution entity determines and obtains the generation progress indication in step 204 above, indicating that the second generated content has been generated completely, that is, it has also been generated completely, the execution entity may respond to this by choosing to output the second generated content after outputting the first generated content, so as to present and provide the user with the complete content corresponding to his / her generation request (for example, after outputting and presenting the first generated content, the second generated content is output and presented sequentially and successively).

[0054] The content generation method provided in this embodiment first obtains a generation request, which instructs the generation of at least two target content chapters; then, a first content chapter and a second content chapter are determined from the target content chapters; then, a first method and a second method are used in parallel to generate and output the first generated content corresponding to the first content chapter and the second generated content corresponding to the second content chapter, respectively; then, in response to the completion of the generation of the first generated content, the generation progress of the second generated content is obtained; finally, in response to the generation progress indication that the second generated content has been generated, the second generated content is output; wherein, the first method instructs that the generated first generated content is output in real time during the generation of the first generated content; the second method instructs that the second generated content is output after the generation of the second generated content is completed. Thus, by using the first method and the second method in parallel, in the process of processing the generation request and generating content, the first generated content can be used to provide real-time feedback to the user on at least part of the generated content, satisfying the user's requirements for real-time presentation of the generated content, while the second generated content, generated in a "holistic generation method" and which takes into account more of the context of the generated content, can also balance the overall generation quality of the generated content.

[0055] In some embodiments, if the generation progress indicator determined and obtained in step 204 above indicates that the second generated content has not been generated in full, that is, the current second generated content cannot be provided or presented to the user, the executing entity may respond by choosing to continue generating and providing the first generated content using a new first content chapter, so that while waiting for the second generated content, the user can still continuously and in real time obtain some content feedback, that is, as discussed above, maintain the real-time feedback status for the user.

[0056] For example, if the executing entity only makes the above division of some content chapters during the process of dividing and determining the first content chapter and the second content chapter, then in such a case, the executing entity can divide a new first content chapter from those content chapters that were not previously specifically divided into first and second content chapters, in order to continue to output and provide the first generated content.

[0057] In some embodiments, to avoid omissions caused by chapter division (e.g., because a specific content chapter was not divided into a first content chapter or a second content chapter, resulting in the corresponding content not being generated), the division of all content chapters may be completed at once during chapter division, so that there will be no "content chapters that were not specifically divided into a first content chapter or a second content chapter".

[0058] In this case, if the second content is not fully generated, the executing entity can choose to first read the generation status of each second content chapter.

[0059] Regarding the occupied state, if a specific second content chapter has been provided and input into a generation thread for task processing, it can be understood as occupied. If it has not been provided or input into a generation thread, or even if it has not been assigned a corresponding generation thread, its occupied state can be correspondingly unoccupied.

[0060] Next, if there are target second content chapters in these second content chapters that are not occupied, the executing entity can determine a new first content chapter by re-dividing the target second content chapter and converting it into a first content chapter, and then use the new first content chapter to continue generating the first generated content.

[0061] In this way, after the first content chapter has been used and generated, the user's "instant feedback experience" can be maintained and improved without affecting the second content generation process or causing duplicate generation. This is achieved by transforming and re-dividing the second content chapter that has not yet been processed.

[0062] In some embodiments, due to reasons such as computing resource usage and configuration difficulty, the number of actual generation threads that can be used (i.e., available generation threads) can be limited during the content generation process to constrain the usage of computing resources during content generation.

[0063] In this case, the executing entity can also complete the content generation process in batches during the generation of the second content, thereby reducing the configuration requirements and difficulty of the generation thread.

[0064] For ease of understanding, the following will be referenced. Figure 3 Let's discuss them together. Figure 3 A flowchart of a process for generating second generated content is provided for an embodiment of this disclosure, including process 300.

[0065] Process 300 specifically includes the following steps: Step 301: Read the total number of chapters in the second content section; Specifically, as discussed above, the executing entity can first read the total number of the divided second content chapters to determine whether the currently available generation threads can meet the requirement of processing these second content chapters simultaneously.

[0066] It should be understood that, for ease of understanding, the “available generation threads” can actually be the “available generation threads” for the second content chapter. For example, it can be the result of minus 1 the total number of generation threads that can be used by the execution entity currently configured (because at least one thread needs to be allocated to the first generation content).

[0067] Next, the executing entity can compare the total number of the second content chapters with the target number of available generation threads to determine whether the current second content chapter can be simultaneously allocated the corresponding available generation threads for processing.

[0068] If the total number is less than or equal to the target number of available generation threads, it means that the current number of available generation threads is sufficient to process these second content chapters simultaneously. In this case, the execution entity can allocate an available generation thread for each second content chapter and then use the allocated available generation thread to complete the processing of the second content chapter and generate the corresponding (sub)generated content. This action is not shown in process 300.

[0069] If the total number exceeds the target number of available generating threads, the execution entity can choose to continue executing step 302.

[0070] Step 302: In response to the total number being greater than the target number of available generation threads, divide the second content chapter into at least two generation batches; Specifically, if the total number is greater than the target number of available generation threads, it means that it is currently impossible to use these available generation threads to complete the processing of all the second content chapters at once. In such a case, as discussed above, the executing entity can divide the second content chapters into at least two generation batches and process them in batches.

[0071] Typically, the executing entity can generate at least two batches based on the "target quantity". That is, except for the last batch in which the number of second content chapters is less than or equal to the "target quantity", the number of second content chapters in the remaining batches is equal to the "target quantity".

[0072] Then, the executing entity can continue to execute step 303 to generate the sub-generated content corresponding to each of the second content chapters in batches based on the generation batch and utilizing the available generation threads. For example, as discussed above, in each generation batch, the executing entity can allocate available generation threads for the second content chapters in that batch and utilize these available generation threads to generate the sub-generated content corresponding to each of the second content chapters.

[0073] Step 303: Based on the generation batch, generate the sub-generated content corresponding to each of the second content chapters in batches using the available generation threads; Step 304: In response to the completion of the generation of corresponding sub-generated content for each second content chapter, the second generated content corresponding to the second content chapter is generated by combining the sub-generated content.

[0074] Specifically, after generating the sub-generated content corresponding to each second content chapter by generating batches, the executing entity can respond by combining the sub-generated content to generate the second generated content corresponding to the second content chapter.

[0075] In some embodiments, during the process of combining the various sub-generated contents, the executing entity may, as discussed above, not only directly splice and arrange the formal combinations of the various sub-contents, but also "polish" the content of the formal combination after completing the formal combination (e.g., unify the wording for the unified object, adjust content with logical conflicts, etc.) to form a coherent "second generated content".

[0076] In some embodiments, as discussed above, because the content elements of the first generated content are presented to the user in real time during the generation process, the quality of the generated content is usually weaker than that of the second generated content (because the second generated content is allowed more time and space for overall polishing and adjustment). Therefore, in the process of dividing the first content chapters and the second content chapters, there is a general tendency to divide more content chapters into the second content chapters in order to improve the overall content generation quality. In such cases, the first content chapters can be selected and divided as little as possible to improve the overall content generation quality.

[0077] In some embodiments, the executing entity can select the first content chapter in the content chapter order as the first content chapter. This not only allows users to obtain an intuitive experience of generating feedback from the beginning of the content, but also allows content chapters with weaker "contextual connections" to be used as the first content chapter, so that content chapters with stronger contextual connections can be better "polished" and the overall content generation quality can be improved.

[0078] Accordingly, in some embodiments, if the executing entity allows the continuation of the already generated first content chapter by changing the target second content chapter to the first content chapter, and continues to output and provide the first generated content, and also chooses to generate the sub-generated content of the second content chapter in batches, then when dividing the generation batches, the executing entity can simultaneously refer to the case where the first content chapter in the sorting order is selected as the first content chapter. In the process of dividing the second content chapter into at least two generation batches, it can choose to divide the second content chapter into at least two generation batches according to the chapter sorting, starting from the last content chapter in the sorting order and proceeding in reverse chapter order. Furthermore, the execution order of these generation batches can correspond to their respective divisions (i.e., the generation batch that is divided first is executed first).

[0079] For example, in the case of content chapters 1, 2, 3, 4, and 5, if the number of available generation threads for the second content chapter is 2, the execution entity can take the content chapter 1, which is the first content chapter, as the first content chapter, and take the content chapters 2, 3, 4, and 5 as the second content chapters, dividing them into two generation batches: (content chapters 5 and 4) and (content chapters 3 and 2). The available generation threads are then called in the order of (content chapters 5 and 4) first and (content chapters 3 and 2) last.

[0080] Therefore, when supplementing or modifying the first content chapter using the target second content chapter, it is possible to prioritize the use of target second content chapters that are far from those that have already been used and generated second content, in order to avoid the splits or breaks in the second generated content caused by the target second content chapter being converted into the first content chapter, and to reduce the negative impact of the target second content chapter being converted into the first content chapter on the second generated content.

[0081] In some optional implementations of this embodiment, when the executing entity supplements the first content chapter using the target second content chapter, it can also choose to supplement it in a "single chapter" manner to try to control the total number of first content chapters, so that more content chapters can be generated as second content chapters, thereby improving the overall content generation quality. That is, when the executing entity needs to change or supplement the first content chapter using the target second content chapter, the executing entity can first detect the number of target second content chapters that are not occupied.

[0082] If at least two such target second content chapters exist, the executing entity can respond by selecting the target second content chapter with the higher ranking as the first content chapter, and continue to generate the first generated content using the first method. That is, if at least two target second content chapters exist, the executing entity can preferentially select the target second content chapter that is closer to the previously selected first content chapter and further from the second content chapter in the previously executed generation batch (divided by reverse ranking) as a supplement to the first content chapter, and continue to generate the first generated content using the first method. Alternatively, the executing entity can select the target second content chapter with the higher ranking as the first content chapter based on the chapter ranking, and continue to generate the first generated content.

[0083] In this way, not only can the first content chapter be supplemented by a target second content chapter that is closer to the first generated content and does not cause the second generated content to be fragmented, but the first content chapter can also be supplemented by a "separate chapter" supplementation method based on "chapter" to minimize the number of target second content chapters actually used as supplementation, so that the target second content chapters can still be more inclined to generate content as "second content chapters".

[0084] In some embodiments, if the target second content chapter is converted into a first content chapter for processing and at least a portion of the first generated content is generated, then in such a case, the executing entity may, in response, choose to delete the target second content chapter from the target generation batch that includes the target second content chapter.

[0085] This is to avoid mistakenly and repeatedly generating content as a second content chapter when the target second content chapter has already been used as a first content chapter, thus avoiding wasting computing resources.

[0086] In some embodiments, if the executing entity references related content (e.g., expert knowledge, references, etc.) during the generation of the first generated content and / or the second generated content, the executing entity may respond by presenting descriptive information about the related content (e.g., the name of the related content, a link to access it, etc.) after referencing the related content. For example, the executing entity may present this descriptive information in a specific area of ​​the user feedback interface.

[0087] This allows users to better understand the generated content through these descriptive details (for example, they can learn about the source of the generated content through these related details).

[0088] It should be noted that the associated content can be obtained directly from the local storage device by the aforementioned executing entity, or it can be obtained from a non-local storage device (e.g., Figure 1 The information can be obtained from the terminal devices 101, 102, and 103 shown. The local storage device can be a data storage module located within the aforementioned execution entity, such as a server hard drive. In this case, the associated content can be quickly read locally. The non-local storage device can also be any other electronic device configured to store data, such as user terminals. In this case, the aforementioned execution entity can obtain the required associated content by sending an acquisition command to the electronic device.

[0089] In some optional implementations of this embodiment, the executing entity may also choose to adjust the order of the description information by referring to the chapter order of the content chapters corresponding to the related content. For example, the related content corresponding to content chapter 1 may be presented before the related content corresponding to content chapter 2. This ensures that the order of the description information corresponds to the order of the content chapters, avoiding the situation where the description information is presented in a cross-chapter or disordered order due to sorting based on reference or citation time, which would affect the user's reading experience.

[0090] In some embodiments, to avoid situations where a content chapter is not expected to be used as both the first and second content chapters simultaneously due to reasons such as failure to update the generation batch in a timely manner after the target second content chapter is converted, resulting in the simultaneous generation of first and second generated content, the executing entity may, after generating the second generated content, first select to check whether there is such an overlapping content chapter (i.e., an overlapping content chapter is used as the first content chapter to generate the first generated content, and at the same time as the second content chapter to generate the second generated content).

[0091] Accordingly, if such overlapping content sections exist, the executing entity can respond by deleting the portions of the second generated content associated with the overlapping content sections, resulting in updated second generated content. Then, the executing entity actually selects to output the updated second generated content to avoid the overlapping content sections interfering with the first generated content already presented to the user, thus preventing user confusion and improving the user's overall reading experience of the generated content.

[0092] To enhance understanding, this disclosure also provides a specific implementation scheme for a particular application scenario. Please refer to the following for details. Figure 4 . Figure 4 The flowchart of the content generation process specifically implemented in an application scenario provided in this disclosure includes process 400.

[0093] For ease of understanding, process 400 will also utilize... Figure 1The exemplary system architecture 100 shown will be discussed.

[0094] For example, in process 400, terminal device 101 can be used by a user (not shown) to obtain generated content services provided by server 415. For example, terminal device 101 can use interface 410 to collect generation requests from the user and provide generated content.

[0095] In process 400, if a user has a generation request, they can send a generation request 421 to the server 105 by executing S401 through the terminal device 101.

[0096] Accordingly, if server 105 receives the generation request 421, it can generate the (target) content chapter corresponding to the generation request 421 by executing S402. For example, content chapters 431, 432...43N are generated (where N is a positive integer).

[0097] Next, server 105 can determine the first content chapter and the second content chapter from the target content chapter by executing S403.

[0098] For example, in process 400, in S403, server 105 can select the "first" content chapter 431 as the first content chapter, and the remaining content chapters as the second content chapters.

[0099] Next, server 105 can execute S404 and S405 in parallel to generate the first generated content 441 corresponding to the first content chapter by executing S404, and continuously output the first generated content corresponding to the first content chapter in real time by executing S406 (i.e., generate the first generated content in the first way) by executing S406, and generate the second generated content 442 corresponding to the second content chapter by executing S405 (i.e., generate the second generated content in the second way).

[0100] S406 can be "repeatedly executed". That is, during the execution of S404, i.e. the generation of the first generated content 441, the server 105 can respond by executing S406 once whenever a content element (e.g., character 451) that is used to compose the first generated content 441 is generated, so as to provide or output the currently generated new content element (e.g., character 451) to the terminal device 101.

[0101] Accordingly, upon receiving the components of these first generated content 441, the terminal device 101 can present them to the user accordingly. For example, after the characters 451 are generated, the terminal device 101 can present them on the interface 410 after the server 105 provides the characters 451 by executing S406, so that the user can view and access them.

[0102] Accordingly, after the first content 441 is generated, the interface 410 can also present the complete content of the first generated 441 as a whole (that is, the complete content is presented as a whole by presenting one content element after another).

[0103] During the execution of S405, specifically the generation of the second generated content 442, server 105 can utilize available generation threads (not shown in the diagram) to generate the sub-generated content corresponding to each content chapter that serves as the second content chapter. For example, sub-generated content 462 corresponding to content chapter 432, sub-generated content 46N corresponding to content chapter 43N, etc.

[0104] Then, server 105 obtains the second generated content 442 by combining sub-generated content 462 …46N.

[0105] Furthermore, after the first content 441 is fully generated, the server 105 can execute S407 to obtain the generation progress of S405, that is, whether the second generated content 442 has been generated by executing S405.

[0106] For example, if the second generated content 442 has been generated at this time, then the server 105 can further output the second generated content 442 by executing S408, for example, outputting it to the terminal device 101, so that the terminal device 101 can present the second generated content 442 to the user through the interface 410, so as to complete the complete content generation process.

[0107] Further reference Figure 5 As an implementation of the methods shown in the above figures, this disclosure provides an embodiment of a content generation apparatus, which is similar to... Figure 2 Corresponding to the method embodiments shown, this device can be specifically applied to various electronic devices.

[0108] like Figure 5As shown, the content generation apparatus 500 of this embodiment may include: a generation request acquisition unit 501, a content chapter determination unit 502, a content generation and output unit 503, a progress acquisition unit 504, and a content output unit 505. The generation request acquisition unit 501 is configured to acquire a generation request, which indicates the generation of at least two target content chapters; the content chapter determination unit 502 is configured to determine a first content chapter and a second content chapter from the target content chapters; the content generation and output unit 503 is configured to use a first method and a second method in parallel to generate and output first generated content corresponding to the first content chapter and second generated content corresponding to the second content chapter, respectively; the progress acquisition unit 504 is configured to acquire the generation progress of the second generated content in response to the completion of the first generated content generation; the content output unit 505 is configured to output the second generated content in response to the generation progress indication that the second generated content has been generated; wherein the first method indicates that the generated first generated content is output in real time during the generation of the first generated content; the second method indicates that the second generated content is output after the generation of the second generated content is completed.

[0109] In this embodiment, the specific processing and technical effects of the content generation device 500, including the generation request acquisition unit 501, the content chapter determination unit 502, the content generation and output unit 503, the progress acquisition unit 504, and the content output unit 505, can be found in references to [the relevant documentation]. Figure 2 The relevant descriptions of steps 201-205 in the corresponding embodiments will not be repeated here.

[0110] In some optional implementations of this embodiment, the device 500 further includes: an occupation reading unit configured to read the occupation status of each second content chapter in response to a generation progress indication that the second generated content has not been completed; and a chapter changing unit configured to, in response to the existence of a target second content chapter whose occupation status is not occupied, use the target second content chapter as the first content chapter and continue to generate the first generated content using the first method.

[0111] In some optional implementations of this embodiment, generating the second generated content corresponding to the second content chapter includes: reading the total number of second content chapters; in response to the total number being greater than the target number of available generation threads, dividing the second content chapters into at least two generation batches; based on the generation batches, using available generation threads to generate the sub-generated content corresponding to each of the second content chapters in batches; in response to each second content chapter having completed the generation of its corresponding sub-generated content, generating the second generated content corresponding to the second content chapter by combining the various sub-generated content.

[0112] In some optional implementations of this embodiment, the apparatus 500 further includes: a repeating chapter adjustment unit, configured to delete the target second content chapter in a target generation batch that includes the target second content chapter in response to the target second content chapter having been generated as at least a part of the first generated content chapter as the first content chapter.

[0113] In some optional implementations of this embodiment, the device 500 further includes: a first content chapter selection unit, configured to select the first content chapter in the sorting order as the first content chapter based on the chapter order of the content chapters.

[0114] In some optional implementations of this embodiment, dividing the second content chapter into at least two generation batches includes: dividing the second content chapter into at least two generation batches in reverse order of chapter sorting, starting from the last content chapter in the sorting order, wherein the order of generation batch division corresponds to the order of execution.

[0115] In some optional implementations of this embodiment, the repeated chapter adjustment unit is further configured to, in response to the existence of at least two target second content chapters whose generation occupancy status is unoccupied, select the target second content chapter with the earlier ranking as the first content chapter based on the chapter ranking, and continue to generate the first generated content using the first method.

[0116] In some optional implementations of this embodiment, the apparatus 500 further includes: a description information presentation unit, configured to present description information of the associated content in response to referencing associated content during the generation of the first generated content and / or the second generated content, after referencing the associated content.

[0117] In some optional implementations of this embodiment, the apparatus 500 further includes: a content update unit configured to, in response to the existence of overlapping content chapters, delete the portion of the second generated content associated with the overlapping content chapters to obtain updated second generated content, wherein the overlapping content chapters are used as first content chapters to generate the first generated content and also as second content chapters to generate the second generated content; and an update content output unit configured to output the updated second generated content.

[0118] This embodiment exists as a device embodiment corresponding to the above method embodiment. The content generation device provided in this embodiment, by using the first method and the second method in parallel, enables the user to receive real-time feedback of at least part of the generated content during the process of processing the generation request and generating content, thereby meeting the user's requirements for real-time presentation of the generated content. At the same time, the overall generation quality of the generated content can be taken into account by using the second generated content, which is generated completely in the "overall generation method" and can take into account more of the context of the generated content.

[0119] According to embodiments of this disclosure, this disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0120] Figure 6 A schematic block diagram of an example electronic device 600 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.

[0121] like Figure 6 As shown, device 600 includes a computing unit 601, which can perform various appropriate actions and processes based on a computer program stored in read-only memory (ROM) 602 or a computer program loaded into random access memory (RAM) 603 from storage unit 608. RAM 603 may also store various programs and data required for the operation of device 600. The computing unit 601, ROM 602, and RAM 603 are interconnected via bus 604. Input / output (I / O) interface 605 is also connected to bus 604.

[0122] Multiple components in device 600 are connected to I / O interface 605, including: input unit 606, such as keyboard, mouse, etc.; output unit 607, such as various types of monitors, speakers, etc.; storage unit 608, such as disk, optical disk, etc.; and communication unit 609, such as network card, modem, wireless transceiver, etc. Communication unit 609 allows device 600 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0123] The computing unit 601 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 601 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 601 performs the various methods and processes described above, such as content generation methods. For example, in some embodiments, the content generation method may be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 608. In some embodiments, part or all of the computer program may be loaded and / or installed on device 600 via ROM 602 and / or communication unit 609. When the computer program is loaded into RAM 603 and executed by the computing unit 601, one or more steps of the content generation method described above may be performed. Alternatively, in other embodiments, the computing unit 601 may be configured to perform the content generation method by any other suitable means (e.g., by means of firmware).

[0124] 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.

[0125] 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.

[0126] 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.

[0127] 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).

[0128] 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.

[0129] Computer systems can include clients and servers. Clients and servers are generally geographically separated and typically interact via communication networks. The client-server relationship is established by computer programs running on the respective computers and having a client-server relationship with each other. Servers can be cloud servers, also known as cloud computing servers or cloud hosts, which are hosting products within the cloud computing service ecosystem to address the management difficulties and weak business scalability inherent in traditional physical hosts and Virtual Private Servers (VPS) services. Servers can also be categorized as distributed system servers or servers incorporating blockchain technology.

[0130] According to the technical solution of the present disclosure, by using the first method and the second method in parallel, during the process of processing the generation request and the generation content, it is possible to use the first generated content to provide real-time feedback to the user on at least part of the generated content, thereby meeting the user's requirements for the real-time presentation of the generated content, and at the same time, to take into account the overall generation quality of the generated content by using the second generated content, which is generated completely in the "overall generation method" and can take into account more of the context of the generated content.

[0131] 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 provided in this disclosure can be achieved, and this is not limited herein.

[0132] 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 content generation method, comprising: Obtain a generation request, which instructs the generation of at least two target content chapters; From the target content chapters, determine the first content chapter and the second content chapter; The first and second methods are used in parallel to generate and output the first generated content corresponding to the first content chapter and the second generated content corresponding to the second content chapter, respectively. In response to the completion of generating the first generated content, obtain the generation progress of the second generated content; In response to the generation progress indicator indicating that the second generated content has been generated, the second generated content is output; Wherein, the first method indicates that the generated first content is output in real time during the process of generating the first generated content; The second method indicates that the second generated content is output after the second generated content has been generated.

2. The method according to claim 1, further comprising: In response to the generation progress indication that the second generated content has not been completed, the generation occupancy status of each chapter of the second content is read; In response to the existence of a target second content chapter whose generation occupancy status is unoccupied, the target second content chapter is used as the first content chapter, and the first generated content is generated again using the first method.

3. The method according to claim 2, wherein, Generate the second generated content corresponding to the second content chapter, including: Read the total number of chapters in the second content section; In response to the total number exceeding the target number of available generation threads, the second content chapter is divided into at least two generation batches; Based on the generation batch, the available generation threads are used to generate the sub-generated content corresponding to each of the second content chapters in batches; In response to the completion of the generation of corresponding sub-generated content for each of the second content chapters, the second generated content corresponding to the second content chapter is generated by combining the sub-generated content.

4. The method according to claim 3, further comprising: In response to the fact that the target second content chapter has been generated as at least part of the first content chapter in the first generated content, the target second content chapter is deleted from the target generation batch that includes the target second content chapter.

5. The method according to claim 3, further comprising: Based on the chapter order of the content chapters, the first content chapter at the top of the order is selected as the first content chapter.

6. The method according to claim 5, wherein, The process of dividing the second content chapter into at least two generation batches includes: Based on the chapter sorting, starting from the last content chapter in the sorting order, the second content chapter is divided into at least two generation batches in reverse order of the chapter sorting, wherein the order of the generation batches corresponds to the order of execution.

7. The method according to claim 6, wherein, In response to the existence of a target second content chapter whose generation occupancy status is unoccupied, the target second content chapter is used as the first content chapter, and the first generated content is generated using the first method, including: In response to the existence of at least two target second content chapters whose generation occupancy status is unoccupied, based on the chapter sorting, the target second content chapter with the higher sorting is selected as the first content chapter, and the first generated content is generated using the first method.

8. The method according to claim 1, further comprising: In response to referencing associated content during the generation of the first generated content and / or the second generated content, description information of the associated content is presented after the associated content is referenced.

9. The method according to any one of claims 1-8, wherein, The output of the second generated content includes: In response to the existence of overlapping content chapters, the portion of the second generated content associated with the overlapping content chapters is deleted to obtain updated second generated content, wherein the overlapping content chapters are used as the first content chapters to generate the first generated content, and at the same time, they are used as the second content chapters to generate the second generated content. Output the updated second generated content.

10. A content generation apparatus, comprising: A request acquisition unit is configured to acquire a request to generate a request that indicates the generation of at least two target content chapters. The content chapter determination unit is configured to determine a first content chapter and a second content chapter from the target content chapters; The content generation and output unit is configured to use a first method and a second method in parallel to generate and output the first generated content corresponding to the first content chapter and the second generated content corresponding to the second content chapter, respectively. Progress acquisition unit; configured to acquire the generation progress of the second generated content in response to the completion of the generation of the first generated content; The content output unit is configured to output the second generated content in response to the generation progress indication that the second generated content has been generated; Wherein, the first method indicates that the generated first content is output in real time during the process of generating the first generated content; The second method indicates that the second generated content is output after the second generated content has been generated.

11. 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 content generation method according to any one of claims 1-9.

12. A non-transitory computer-readable storage medium storing computer instructions for causing the computer to perform the content generation method of any one of claims 1-9.

13. A computer program product comprising a computer program that, when executed by a processor, implements the content generation method according to any one of claims 1-9.

Images