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

Abstract

The embodiment of the invention relates to a content generation method and device, equipment, a storage medium and a program product, and relates to the technical field of generative models, large models, large language models, agents, artificial intelligence and the like. The method comprises the steps of displaying a first content generation page in response to a trigger operation acting on a content generation function, and displaying a first description input control and a content generation control in the first content generation page; in response to an input operation on the first description input control, determining content description information; in response to a trigger operation acting on the content generation control, calling a content generation model based on the content description information, and generating target content; in response to a trigger operation acting on the editing control of the target content, displaying an editing page of the target content; and in response to an interaction operation acting on the editing page, editing the target content based on interaction content corresponding to the interaction operation to generate edited target content. Therefore, a secondary editing function is provided, and the generation efficiency and quality of the target content are improved.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, and in particular to a content generation method, apparatus, device, storage medium, and program product. Background Art

[0002] With the development of computer technology, generative models of a certain scale have emerged. People can call generative models by inputting their own needs, and output the various types of content they want (such as text, pictures, audio, video, or a combination of various types of multimedia files, etc.) to improve content generation efficiency and reduce costs. However, due to various reasons such as inaccurate user input requirements, limited training data, or weak model generalization ability, the content output by the model cannot meet user expectations well, resulting in poor quality of the content generated by the model or even unusable problems. Summary of the invention

[0003] In order to solve the above technical problems, the embodiments of the present disclosure provide a content generation method, apparatus, device, storage medium and program product.

[0004] In a first aspect, an embodiment of the present disclosure provides a content generation method, the method comprising:

[0005] In response to a triggering operation acting on a content generation function corresponding to the target content, displaying a first content generation page corresponding to the content generation function, and displaying a first description input control and a content generation control in the first content generation page;

[0006] In response to an input operation acting on the first description input control, determining content description information;

[0007] In response to a trigger operation acting on the content generation control, based on the content description information, a content generation model corresponding to the content generation function is called to generate the target content;

[0008] In response to a triggering operation on an editing control corresponding to the target content, an editing page of the target content is displayed;

[0009] In response to an interactive operation on the editing page, the target content is edited based on interactive content corresponding to the interactive operation to generate edited target content.

[0010] In a second aspect, an embodiment of the present disclosure further provides a content generation device, the device comprising:

[0011] A first content generation page display module, configured to respond to a triggering operation on a content generation function corresponding to a target content, display a first content generation page corresponding to the content generation function, and display a first description input control and a content generation control in the first content generation page;

[0012] a content description information determination module, configured to determine content description information in response to an input operation acting on the first description input control;

[0013] A target content generation module, configured to respond to a trigger operation acting on the content generation control, call a content generation model corresponding to the content generation function based on the content description information, and generate the target content;

[0014] An editing page display module, configured to respond to a triggering operation on an editing control corresponding to the target content and display an editing page of the target content;

[0015] The edited target content generation module is used to respond to the interactive operation acting on the editing page, edit the target content based on the interactive content corresponding to the interactive operation, and generate the edited target content.

[0016] In a third aspect, an embodiment of the present disclosure further provides an electronic device, the electronic device comprising:

[0017] processor;

[0018] A memory for storing executable instructions;

[0019] The processor is used to read executable instructions from the memory and execute the executable instructions to implement the content generation method described in any embodiment of the present disclosure.

[0020] In a fourth aspect, an embodiment of the present disclosure further provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the processor implements the content generation method described in any embodiment of the present disclosure.

[0021] In a fifth aspect, an embodiment of the present disclosure further provides a computer program product, which is used to execute the content generation method described in any embodiment of the present disclosure.

[0022] The content generation method, apparatus, device, storage medium and program product of the embodiments of the present disclosure can respond to a trigger operation on a content generation function corresponding to a target content, display a first content generation page corresponding to the content generation function, and display a first description input control and a content generation control in the first content generation page; respond to an input operation on the first description input control, determine content description information; respond to a trigger operation on the content generation control, call a content generation model corresponding to the content generation function based on the content description information, and generate target content; respond to a trigger operation on an edit control corresponding to the target content, display an edit page of the target content; respond to an interactive operation on the edit page, edit the target content based on the interactive content corresponding to the interactive operation, and generate edited target content; after the target content is generated by using a generative model, a secondary editing function for the target content is provided, so that the user can perform personalized editing on the target content through the edit page of the target content, and generate edited target content that better meets the user's expectations, thereby improving the generation efficiency of the target content by using the generative model, and improving the personalization of the target content and its degree of fit with business needs, thereby improving the generation quality of the target content.

[0023] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in the embodiments of the present disclosure are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards of relevant countries and regions, and provide corresponding operation entrances for users to choose to authorize or refuse. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and that components and elements are not necessarily drawn to scale.

[0025] Figure 1 A flowchart of a content generation method provided by an embodiment of the present disclosure;

[0026] Figure 2 A schematic diagram showing a first content generation page of a text content generation function provided by an embodiment of the present disclosure;

[0027] Figure 3 A flowchart of another content generation method provided by an embodiment of the present disclosure;

[0028] Figure 4A schematic diagram of a hotspot selection window provided in an embodiment of the present disclosure;

[0029] Figure 5 A schematic diagram showing a first content generation page of a picture content generation function provided in an embodiment of the present disclosure;

[0030] Figure 6 A schematic diagram of displaying an editing pop-up window for partial editing provided by an embodiment of the present disclosure;

[0031] Figure 7 A schematic diagram showing a first content generation page of another picture content generation function provided by an embodiment of the present disclosure;

[0032] Figure 8 A schematic diagram of displaying a global editing editor page provided by an embodiment of the present disclosure;

[0033] Fig. 9 A schematic diagram of displaying a first content generation page of a multimedia content generation function provided by an embodiment of the present disclosure;

[0034] Fig.10 A schematic diagram of displaying a second content generation page of a multi-size multimedia content provided by an embodiment of the present disclosure;

[0035] Fig.11 A schematic diagram of the structure of a content generation device provided by an embodiment of the present disclosure;

[0036] Fig.12 A schematic diagram of the structure of an electronic device provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0037] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments described herein, which are instead provided for a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0038] It should be understood that the various steps described in the method embodiments of the present disclosure may be performed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present disclosure is not limited in this respect.

[0039] The term "including" and its variations used herein are open inclusions, i.e., "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0040] It should be noted that the concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0041] It should be noted that the modifications of "one" and "plurality" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0042] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0043] In the related art, when a generative model with a certain scale is used to generate various types of content (such as text, pictures, audio, video or a combination of various types of multimedia files, etc.), the user mainly inputs his demand for a certain type of content (such as text content describing the demand), then constructs the demand as a model prompt, and inputs the model prompt into the generative model, and outputs the corresponding type of content through the operation of the model. However, the generative model is obtained through training with limited training data, and its calling method is to construct the prompt based on the demand input by the user. Therefore, the generalization ability of the generative model may be insufficient due to the inaccurate description of the demand input by the user, the limitations of the training data (such as uneven distribution, not adapted to a certain type of content, not covering enough business scenarios, etc.), and the poor training process, resulting in different expressiveness of the output content for different business scenarios and business needs. This easily leads to poor content generated by the model and cannot meet user needs well, such as the generated content has strong versatility but cannot fit the business scenario, making it difficult to use the generated content in combination with the business, and unable to support the customization of the user's own style and materials, etc.

[0044] Based on the above situation, the embodiment of the present disclosure provides a content generation solution. On the one hand, before calling the generative model to generate the target content, more dimensional interactive controls including the demand description input by the user can be provided, so that the user can more conveniently improve the user demand from multiple dimensions through these interactive controls, so that the generative model can obtain the model prompt word prompt that is more in line with the user demand and the relevant elements (such as pictures, audio, video or access paths, etc.) to be embedded in the target content, thereby improving the degree of fit of the target content generated by the model to the business demand and business scenario, so that the target content can better meet the user's expectations; on the other hand, after calling the generative model to generate the target content, an interactive function for secondary editing of the target content can be provided, so that the user can perform secondary editing processing on the target content to different degrees through the interactive controls in the editing page to obtain the edited target content, so that the target content can be better combined with the business demand and business scenario, so as to better support the user to personalize the content style and materials; in short, the generative model can be optimized before and after the call, so as to improve the degree of fit of the target content finally generated to the business scenario and business demand, so that the target content can better meet the user's expectations, thereby improving the generation efficiency and generation quality of the target content.

[0045] The content generation method provided in the embodiments of the present disclosure can be applied to scenarios where various types of content such as text, pictures, multimedia files, etc. are generated. The method can be executed by a content generation device, which can be implemented by software and / or hardware, and the device can be integrated in an electronic device with a human-computer interaction function. The electronic device can include, but is not limited to, a smart phone, a personal digital assistant (PDA), a tablet computer (Tablet Personal Computer, Tablet PC), a laptop computer, or a desktop computer.

[0046] Figure 1 FIG. 1 is a flow chart of a content generation method provided by an embodiment of the present disclosure. Figure 1 As shown, the content generation method may include the following steps:

[0047] S110 . In response to a triggering operation acting on a content generation function corresponding to the target content, a first content generation page corresponding to the content generation function is displayed, and a first description input control and a content generation control are displayed in the first content generation page.

[0048] Wherein, the target content is the content to be automatically generated. Exemplarily, the target content includes at least one of text content, picture content and multimedia content. The multimedia content here is a file that can integrate various types of content such as text, picture, audio, video, access path, etc., for example, it can be a poster or HyperText Markup Language 5 (H5) page, etc. The content generation function is a client function that can generate and / or edit content, which is adapted to the content type. For example, when the target content is text content, the content generation function is a text content generation function that generates text and / or edits text. For another example, when the target content is picture content, the content generation function is a picture content generation function that generates text and / or edits pictures, etc. For another example, when the target content is multimedia content, the content generation function is a multimedia content generation function such as generating multimedia files with text, editing multimedia files, and expanding the size of multimedia files, etc. The first content generation page is a page for displaying information related to the content generation function. The first description input control is an interactive control for inputting a description of content requirements, such as a text box, a function button pointing to a text box, etc. The content generation control is an interactive control for triggering a function of generating content, such as a function button presenting content generation text or an image.

[0049] Specifically, the electronic device may provide an interactive entry for the content generation function. For example, a drop-down button is displayed on a page, and the drop-down options include content generation functions corresponding to various types of content; for another example, a page may display function buttons for content generation functions corresponding to various types of content. Users can trigger the corresponding content generation function according to the type of content they need. Figure 2 , a drop-down button 210 of the content generation function is displayed in a page such as a client / webpage / applet with a content generation function. If the target content is text content, the user can trigger the drop-down button 210 and select the drop-down option of the text content generation function. In response to the interactive operation, the electronic device displays a first content generation page corresponding to the text content generation function. At least a first description input control 220 and a content generation control 230 are displayed in the first content generation page.

[0050] S120: Determine content description information in response to an input operation applied to a first description input control.

[0051] The content description information is text content used to describe the demand for content, and may include at least one of the subject, main elements, additional elements, purpose, and style of the content, for example.

[0052] Specifically, the user can trigger the first description input control. For example, see Figure 2, the first description input control 220 may be a text box 221, and the user may trigger the text box to input the content description information. Here, the user may input the complete content description information through the text box; or the text box may be provided with some existing text content in advance, and the user may fill in the remaining text content (such as Figure 2 Medium grey italic text example). For another example, the first description input control 220 may be an interactive button 222 that displays "imitate text" in the area surrounding the text box. When the user triggers the interactive button 222, a pop-up window may be displayed to display relevant prompts for automatically generating content description information. After the user performs an interactive operation according to the prompt, the electronic device may call a relevant model (such as a pre-trained generative model of natural language) to automatically generate content description information according to the function corresponding to the interactive button, so as to assist the user in completing the input operation of the content description information and improve the input efficiency and accuracy of the content description information. In this way, the electronic device can determine the content description information according to the text content input by the user in response to the input operation of the first description input control.

[0053] S130 . In response to a trigger operation acting on a content generation control, based on content description information, a content generation model corresponding to the content generation function is called to generate target content.

[0054] Among them, the content generation model is a pre-trained generative model that has the function of generating a certain type of content based on text.

[0055] Specifically, after the user inputs the content description information, the content generation control can be triggered. In response to the trigger operation, the electronic device uses the content description information to generate a model prompt word. For example, the electronic device can directly use the content description information as the model prompt word. For another example, the electronic device can obtain the framework of the constructed model prompt word in advance, and fill the information in the content description information into the framework to obtain a complete model prompt word. Then, the electronic device can input the model prompt word into the content generation model corresponding to the content type to which the target content belongs, and output the target content after model processing. The target content can be displayed in a certain area of ​​the first content generation page, and can be stored according to a pre-set storage path.

[0056] S140 . In response to a triggering operation on an editing control corresponding to the target content, an editing page of the target content is displayed.

[0057] Among them, the editing control is an interactive control corresponding to the content editing function, for example, it can be a local editing control for editing local content of the content, a global editing control for editing the entire content of the content, an editing control for optimizing the model input content of the generated content, etc.

[0058] Specifically, in order to improve the degree of fit between the target content and the business, the disclosed embodiment may also provide the user with a function of secondary editing of the target content after the target content is generated. In this way, the electronic device may display editing controls for the target content. For example, the electronic device may display editing controls for the target content in a relevant area of ​​the display position of the target content (such as an adjacent peripheral area or an edge area of ​​the same display area). For another example, after displaying the target content, the electronic device may display editing controls for the target content in the peripheral area of ​​the target content in response to a triggering operation such as a long press or selection performed by the user on the generated target content.

[0059] Then, when the user performs a trigger operation on the editing control, the electronic device can respond to the trigger operation and display a display page (i.e., an editing page) presented by the secondary editing function corresponding to the triggered editing control. In the editing page, the electronic device can display an interactive control corresponding to at least one editing function to show the user the corresponding secondary editing function, thereby guiding the user to perform the corresponding secondary editing operation.

[0060] S150 . In response to an interactive operation on the editing page, based on interactive content corresponding to the interactive operation, edit the target content to generate edited target content.

[0061] Specifically, the user can follow the guidance of the interactive control in the editing page to perform the corresponding interactive operation on the interactive control corresponding to the secondary editing function that the user wants to implement. The electronic device can respond to the interactive operation and obtain the content obtained by the interaction as the interactive content. Then, the secondary editing function is performed using the interactive content to complete the editing process of the target content and obtain the edited target content.

[0062] It should be noted that some secondary editing functions may require a series of interactive operations to be performed continuously, and the electronic device can respond to each interactive operation and continuously edit the target content according to the interactive content obtained each time. For example, if the secondary editing function is to add a new element to the target content, then a series of interactive operations may be selecting an element, adjusting the size of the element at least once, and adjusting the layout position of the element at least once, etc. The electronic device can perform editing in response to each interactive operation in sequence, and display the results generated according to the interactive content in real time until all editing processes are completed and the final target content is obtained.

[0063] The content generation method provided by the embodiment of the present disclosure can respond to a trigger operation on a content generation function corresponding to a target content, display a first content generation page corresponding to the content generation function, and display a first description input control and a content generation control in the first content generation page; respond to an input operation on the first description input control, determine content description information; respond to a trigger operation on the content generation control, call a content generation model corresponding to the content generation function based on the content description information, and generate target content; respond to a trigger operation on an edit control corresponding to the target content, display an edit page of the target content; respond to an interactive operation on the edit page, edit the target content based on the interactive content corresponding to the interactive operation, and generate edited target content; after the target content is generated by using a generative model, a secondary editing function for the target content is provided, so that the user can perform personalized editing on the target content through the edit page of the target content, and generate edited target content that better meets the user's expectations, thereby improving the generation efficiency of the target content by using the generative model, and improving the personalization of the target content and its degree of fit with business needs, thereby improving the generation quality of the target content.

[0064] Figure 3 is a flowchart of another content generation method provided by an embodiment of the present disclosure. The content generation method adds relevant technical content of the interaction process before generating the target content. The explanation of the terms in the content generation method that are the same or corresponding to the above embodiments will not be repeated here. Figure 3 , the content generation method comprises:

[0065] S310: In response to a triggering operation on a content generation function corresponding to the target content, a first content generation page corresponding to the content generation function is displayed, and a first description input control, a display content attribute setting control, and a content generation control are displayed in the first content generation page.

[0066] The content attribute setting control refers to an interactive control for setting the attributes of the target content to be generated, and its control implementation form can be set according to the type of the attribute. For example, the content attribute setting control can be implemented as a function button, a drop-down box or a text box, etc.

[0067] Specifically, according to the foregoing description, in the process of implementing content generation using a generative model in the related art, only an input function of the requirement description is provided before the model is called, but there may be a problem of inaccurate description. Therefore, the embodiment of the present disclosure can provide at least one content attribute setting control in addition to the first description input control. Each content attribute setting control corresponds to a content-related attribute, which can be an attribute that constrains the model, or it can be a related element input to the model for embedding in the content (such as pictures, audio, video, access paths to content-related information, encoded images of the access paths, etc.). That is, each content attribute setting control corresponds to a content attribute field of an attribute dimension, and each attribute dimension can be set according to the content type and business scenario. For example, for the generation scenario of corporate promotional content and the content type of promotional copy, each attribute dimension may include the industry dimension of the industry to which the product to be promoted belongs, the scenario dimension of the product usage scenario, the crowd category dimension of the product applicable population, the channel dimension of the product promotion channel, the model dimension of the generative model used to generate the content, etc.; for another example, for the content type of the picture, each attribute dimension may include the style dimension of the picture style to be generated, the number dimension of the pictures to be generated, the size dimension of the pictures to be generated, etc.; for another example, for the content type of multimedia, each attribute dimension may include the product dimension of the multimedia content to promote the product, the title dimension, the attribution dimension of the product attribution party, the layout dimension, etc. In this way, when the electronic device displays the first content generation page, it can also display at least one content attribute setting control therein to guide the user to increase the amount of information input into the model through interaction with the content attribute setting control, thereby improving the accuracy of the model operation, and then improving the quality of the target content generated subsequently and its fit with business needs.

[0068] S320: Determine content description information in response to an input operation applied to a first description input control.

[0069] S330: In response to a trigger operation on a content attribute setting control, determine a content attribute value.

[0070] Specifically, the user can trigger at least one content attribute setting control to determine the content that the user wants to input according to the control interaction mode, that is, the electronic device determines the content attribute value of the content attribute field corresponding to the triggered content attribute setting control according to the content attribute setting control triggered by the user and the user's operation result. For example, the electronic device can determine the relevant identifier corresponding to the content attribute setting control selected by the user as the content attribute value; for another example, the electronic device can determine the information input / selected by the user through the content attribute setting control as the content attribute value, etc.

[0071] S340: In response to a trigger operation acting on the content generation control, based on the content description information and the content attribute value, a content generation model corresponding to the content generation function is called to generate target content.

[0072] Among them, the content generation model is obtained by training the generative model using the training samples corresponding to the content generation function. In this way, the content generation model is adapted to the content generation function. For example, for the text content generation function, the content generation model is obtained by training with many text samples corresponding to the business scenario; for the image content generation function, the content generation model is obtained by training with many image samples corresponding to the business scenario; for the multimedia content generation function, the content generation model is obtained by training with many multimedia samples corresponding to the business scenario (such as poster samples, H5 page samples, layout frameworks of many elements in multimedia, element color matching rules, etc.).

[0073] Specifically, the electronic device can respond to the user's triggering operation on the content generation control and combine the aforementioned content description information and content attribute values ​​to construct a model prompt word. Similarly, the electronic device can directly set the content description information and content attribute values ​​as the model prompt word; or fill the content description information and content attribute values ​​into the placeholder of the pre-constructed model prompt word to generate a complete model prompt word. Then, the electronic device inputs the model prompt word into the content generation model corresponding to the content generation function corresponding to the target content, and generates the target content through model calculation.

[0074] In some embodiments, after S340, the method further includes: displaying the target content in the first area of ​​the first content generation page, and displaying a content storage control in the first content generation page; and in response to a trigger operation acting on the content storage space, storing the target content in the storage space corresponding to the trigger operation.

[0075] The first area is a display area in the first content generation page. Since the first area is used to display the generated target content, the first area is a display area with a larger size. The content storage control is an interactive control for triggering storage of content, for example, it can be a function button pointing to the storage space.

[0076] Specifically, see Figure 2After generating the target content, the electronic device may display the generated target content in the first area 240 distributed on the right side of the first content generation page for the user to view. At the same time, the electronic device may display a content storage control 250 with a text label of "Download / Save" in a certain area (such as the display area on the top right) of the first content generation page. If the user triggers the content storage control 250, the electronic device stores the target content in the corresponding storage space according to the pre-set storage path. Alternatively, the electronic device may display an interactive page such as a pop-up window or dialog box for setting the storage path in response to the user's triggering operation, and the electronic device may store the target content in the corresponding storage space according to the storage path set by the user through the interactive page.

[0077] In some embodiments, after S340, the method further includes: displaying the target content in the first area of ​​the first content generation page, and displaying a historical content viewing control in the first area; and displaying content information of the historical content generated within the historical time period in the second area of ​​the first content generation page in response to a trigger operation acting on the historical content viewing control.

[0078] The historical content viewing control is an interactive control for triggering the display of content generated in a historical time period. The second area is a display area in the first content generation page. Since the second area is used to display relevant information of historically generated content, the second area is a display area with a larger size. Historical content is content generated in a historical time period. Content information is information related to content, such as the content itself, the generation date of the content, a brief introduction to the content, etc.

[0079] Specifically, see Figure 2 After generating the target content, the electronic device may display the generated target content in the first area 240 distributed on the right side of the first content generation page for the user to view. At the same time, the electronic device may display a historical content viewing control 260 with a text label of "history" in a certain area (such as the display area on the top right) of the first content generation page. If the user triggers the historical content viewing control 260, the electronic device, in response to the triggering operation, shrinks the display space of the first area 240 to the left, uses the vacant area as the second area 270, and displays content information of at least one historical content in the second area 270.

[0080] In some embodiments, after S340, the method further includes: displaying the target content in the first area of ​​the first content generation page, and displaying a content storage control and a historical content viewing control in the first content generation page; in response to a trigger operation acting on the content storage space, storing the target content in the storage space corresponding to the trigger operation; in response to a trigger operation acting on the historical content viewing control, displaying content information of historical content generated within the historical time period in the second area of ​​the first content generation page.

[0081] Specifically, the electronic device can simultaneously display the content storage control and the history content viewing control in the first content generation page. The user can perform a trigger operation on the content storage control or the history content viewing control as needed, and the electronic device can perform corresponding processing according to the description in the above embodiment.

[0082] S350: In response to a triggering operation on an edit control corresponding to the target content, an edit page of the target content is displayed.

[0083] S360: In response to the interactive operation on the editing page, based on the interactive content corresponding to the interactive operation, edit the target content to generate the edited target content.

[0084] The content generation method provided by the above-mentioned embodiments of the present disclosure displays a content attribute setting control in a first content generation page; determines a content attribute value in response to a trigger operation acting on the content attribute setting control; calls a content generation model corresponding to the content generation function based on content description information and content attribute values ​​in response to a trigger operation acting on the content generation control to generate target content; and provides a setting function for at least one content attribute before using the model to generate content, so that the user can provide richer and more comprehensive prompt words to the model through the interactive operation of the content attribute setting control, thereby improving the accuracy of the model operation, and further improving the quality of the generated target content and its compliance with business needs.

[0085] In some embodiments, see Figure 2 If the target content is text content, the content attribute setting control includes at least one of the industry setting control 280, the scene setting control 290, the crowd category setting control 2100, the channel setting control 2110, the hotspot setting control 2120, the model setting control 2130, the quantity setting control 2140 and the word count setting control 2150.

[0086] Based on the above embodiment, S330 can be implemented as at least one of the following steps A to F.

[0087] Step A: In response to a selection operation on an industry setting control, the selected industry identifier is determined as a content attribute value of the industry dimension.

[0088] Among them, the industry logo represents the industry to which the target content applies.

[0089] Specifically, different industries have different language preferences, and product characteristics of different industries may also be different. Therefore, in the embodiments of the present disclosure, Figure 2 The first content generation page shown displays multiple industry setting controls 280 in the form of button controls. When the user selects an industry setting control 280, the electronic device can determine the text label corresponding to the control as the industry identifier selected by the user. For example, Figure 2 In the example, if the user selects the "Finance" industry button, the "Finance" text is set as the industry identifier. The industry identifier can be used as an industry constraint in the prompt word so that the content generation model can combine the language characteristics and product characteristics of the industry, so that the generated text content is more in line with the industry habits of the corresponding industry, rather than a general text applicable to all industries, thereby improving the professionalism of the generated text content.

[0090] It should be noted that the industry setting control can also be implemented as an interactive control for preset candidate items such as a drop-down box, which is not limited in the embodiments of the present disclosure. The industry as the preset candidate item can be set according to the business scenario and business needs.

[0091] Step B: In response to a selection operation on a scene setting control, the selected scene identifier is determined as a content attribute value of the scene dimension.

[0092] The scenario identifier represents the usage scenario or purpose of the target content. The usage scenario may include product introduction or product recommendation.

[0093] Specifically, different usage scenarios have different emphases on product promotion. For example, when the usage scenario is product introduction, the text content required will pay more attention to the objectivity and comprehensiveness of the product introduction, so as to strive to provide professional and comprehensive explanations of the product from multiple key angles. In addition, the style of the text content is more rigorous and formal, the language is more accurate and standardized, and less exaggerated or emotional words are used. When the usage scenario is product promotion, the text content required will be more storytelling and experiential (such as sharing personal usage experience, scenario description, etc.) to highlight the benefits and value of the product. In addition, the style of the text content is more lively, vivid, and contagious, and the language may contain rich rhetoric, Internet buzzwords, etc. to enhance interest and appeal. Therefore, the embodiments of the present disclosure can be used in Figure 2 The first content generation page shown displays a scene setting control 290 in the form of a drop-down box (only an example, not a limitation). When the user selects a drop-down option, the electronic device can determine the corresponding text content as the scene identifier selected by the user. For example, Figure 2 If the user selects the drop-down option of "Product Introduction", the "Product Introduction" text is set as the scenario identifier. The scenario identifier can be used as a constraint for the product usage scenario in the prompt word, so that the content generation model can combine the text content focus and language style required by the usage scenario, so that the generated text content is more in line with the needs of the corresponding usage scenario, rather than a general text applicable to all usage scenarios, thereby improving the scenario applicability of the generated text content.

[0094] Step C: In response to a selection operation on a crowd category setting control, the selected crowd category identifier is determined as a content attribute value of an applicable object dimension.

[0095] The crowd category identifier represents the category of the crowd to which the target content is applicable. The crowd category can be divided according to business scenarios and business needs. For example, the crowd category can be multiple crowds divided by age, or crowds divided by occupational categories, etc.

[0096] Specifically, different groups of people have different concerns about the product. Therefore, the embodiments of the present disclosure can be Figure 2 The first content generation page shown displays a crowd category setting control 2100 in the form of a drop-down box (only an example, not a limitation). When the user selects a drop-down option, the electronic device can determine the corresponding text content as the crowd category identifier selected by the user. For example, if the user selects the drop-down option of "female white-collar worker", the "female white-collar worker" text is set as the crowd category identifier. The crowd category identifier can be used as a constraint on the user's focus in the prompt word, so that the content generation model can combine the crowd focus, so that the generated text content is not a universal text applicable to all crowds, but rather highlights the text content that the corresponding crowd may pay attention to.

[0097] Step D: In response to a selection operation on the channel setting control, the selected channel identifier is determined as a content attribute value of the channel dimension.

[0098] The channel identifier represents the channel for distributing the target content. The content distribution channel may be, for example, a text message channel, a client push channel, a mini-program push channel, a public account push channel, and the like.

[0099] Specifically, different content distribution channels correspond to different text content lengths and content emphases. Figure 2The first content generation page shown displays a channel setting control 2110 in the form of a drop-down box (only an example, not a limitation). When a user selects a drop-down option, the electronic device can determine the corresponding text content as the channel identifier of the content distribution channel selected by the user. For example, if the user selects the drop-down option of "SMS", the "SMS" text is set as the channel identifier. The channel identifier can be used as a constraint on the distribution channel in the prompt word, so that the content generation model can combine the text emphasis of the distribution channel to make the generated text content more suitable for the corresponding distribution channel, further improving the quality of the text content.

[0100] Step E: In response to a selection operation on the hotspot setting control, the selected first hotspot identifier is determined as a content attribute value of the heat dimension.

[0101] Among them, the hot spot identification is the identification of information whose information popularity meets the preset popularity within a preset time period, and the hot spot identification is used to enhance the popularity of the target content. For example, the hot spot identification can be the title of the hot spot information, the entity contained in the hot spot information, the core content or full text content of the hot spot information, etc. The hot spot information can be obtained by using relevant models to extract hot spots from many web page contents and their interactive content (such as comments), video content and its interactive content (such as comments) within a certain time range. Hot spot information can also be obtained through other platforms with full authorization. The first hot spot identification is the hot spot identification selected by the user for the first interaction.

[0102] Specifically, the text content generated in the related art is relatively general and popular, lacking creative inspiration. Therefore, in the disclosed embodiment, the hot information of the current period is combined with the text content generation process to inspire the creativity of text content generation, so as to increase the topic and popularity of the text content through the hot information, and increase its attractiveness to users.

[0103] In a specific implementation, the electronic device can Figure 2 In the first content generation page shown, a hotspot setting control 2120 for triggering the hotspot information selection function is displayed in the form of a selection control pointing to a hotspot link / page. When the user triggers the hotspot setting control 2120, the electronic device can respond to the trigger operation and display a hotspot selection window containing multiple hotspot information in the content. The hotspot selection window can be a new page, a pop-up window, a floating window, etc. Figure 4As shown, the electronic device can display the hotspot selection window 400 corresponding to the industry hotspot in the form of a pop-up window, a floating window or a floating layer, and display multiple industry controls 410 implemented in the form of function buttons or navigation forms. Each industry control 410 is linked to a group of hotspot information 420 arranged in descending order of popularity. The user can select the hotspot information of the industry he wants to join. The electronic device can then determine the hotspot identifier corresponding to the hotspot information as the first hotspot identifier and return to the first content generation page. The first hotspot identifier can be used as supplementary information in the prompt word so that the content generation model can add content related to the hotspot information corresponding to the first hotspot identifier in the generated text content.

[0104] Step F: In response to a selection operation on a model setting control, the selected model identifier is determined as a content attribute value of the model dimension.

[0105] The model identifier is an identifier of a content generation model used to generate the target content.

[0106] Specifically, in addition to the interactive settings of the aforementioned attribute dimensions, the embodiments of the present disclosure can also add interactive controls for selecting models to control the generation of text content from a model perspective. The multiple candidate models provided here can be obtained by training text samples based on different business needs, for example, they can be general generative models, or they can be generative models that adapt to product recommendation needs, etc. Figure 2 , the electronic device can display the model setting control 2130 in the form of a radio button (only an example, not a limitation) in the first content generation page. When the user selects a certain model setting control, the electronic device can determine its corresponding text content as the model identifier of the content generation model selected by the user. For example, if the user selects the button of "General Model", the "General Model" text is set as the model identifier. In this way, when it is detected that the user has triggered the content generation control 230 later, the content generation model corresponding to the model identifier of the "General Model" can be called to generate text content.

[0107] Based on the above embodiments, continue to refer to Figure 2 , the electronic device can display the quantity setting control 2140 in the first content generation page. In this way, the quantity value input by the user can be determined in response to the user's triggering operation on the quantity setting control 2140. The content generation model can then batch-produce multiple differentiated text contents around the same topic according to the model prompt words constructed in the above embodiments, further improving the efficiency of text content generation.

[0108] Based on the above embodiments, continue to refer to Figure 2, the electronic device can display the word count setting control 2150 in the first content generation page. In this way, in response to the user's triggering operation on the word count setting control 2150, the word value input by the user can be determined, the word count of the text content output by the model can be constrained, and the text content with appropriate length can be output.

[0109] In some embodiments, if the target content is text content, S350 includes: in response to a triggering operation on a text optimization control corresponding to the target content, displaying a second description input control corresponding to the content description information of the target content.

[0110] The text optimization control is an interactive control for triggering optimization or secondary editing of the generated text content. The second description input control is another interactive control for inputting a description of the content requirements.

[0111] Specifically, see Figure 2 After the electronic device displays the target content in the first area of ​​the first content generation page, it can display a text optimization control 2160 in the surrounding area of ​​the target content (such as the area below). When the user triggers the text optimization control 2160, the electronic device can display a second description input control 2170 with a text input function in the form of a pop-up window, a floating window, or a floating layer, as an editing page for optimizing the content description information.

[0112] Based on the above embodiment, S360 can be implemented as: in response to the input operation acting on the second description input control, the input content is determined as interactive content; based on the interactive content, the content description information is updated, and based on the updated content description information, the content generation model is re-called to generate the edited target text.

[0113] Specifically, if the user is dissatisfied with the generated text content, he or she may input new content description information through the second description input control 2170 to supplement or correct the previously input content description information. In response to the above input operation, the electronic device obtains new content description information as the interactive content obtained this time. Then, the electronic device may adjust the previous model prompt words according to the new content description information to obtain new model prompt words; or regenerate new model prompt words using the new content description information and the original content description information. Afterwards, the new model prompt words are input into the content generation model to output new text content, i.e., the edited target text.

[0114] After generating the edited target text, the electronic device may display it in the first area of ​​the first content generation page. Figure 2, the electronic device can continue to display the edited target text below the previously generated text content. In this way, the text content before and after optimization can be displayed at the same time for the user to compare and view.

[0115] Through the above settings, the secondary editing of the text content can be achieved by optimizing the model input data, further improving the quality of the target text and its fit with user expectations.

[0116] In some embodiments, if the target content is text content, S350 includes: responding to a trigger operation on a hotspot change control corresponding to the target content, displaying a hotspot selection window, and displaying a plurality of candidate hotspot information in the hotspot selection window.

[0117] Specifically, see Figure 2 After the electronic device displays the target content in the first area of ​​the first content generation page, it can display a hotspot change control 2180 for triggering the hotspot information selection function in the surrounding area of ​​the target content (such as the area below). When the user triggers the hotspot change control 2180, the electronic device can display the hotspot information selection function in the form of a pop-up window, a floating window, or a floating layer. Figure 4 The hotspot selection window 400 shown serves as an editing page for optimizing the generated text content.

[0118] Based on the above embodiment, S360 can be implemented as: in response to a selection operation on the hotspot selection window, identifying the selected second hotspot as interactive content; based on the interactive content and content description information, re-calling the content generation model to generate the edited target text.

[0119] Specifically, referring to the above description of the hotspot setting control section, the electronic device can use the hotspot information selected by the user as interactive content. Then, a new model prompt word is generated using the selected hotspot information. For example, when the user has not selected the hotspot information through the hotspot setting control, the electronic device can add the hotspot information selected here to the model prompt word; for another example, when the user has selected the hotspot information through the hotspot setting control, the electronic device can use the hotspot information selected this time to replace the previously selected hotspot information, or add hotspot information, etc. Afterwards, the new model prompt word is input into the content generation model, and the new text content, that is, the edited target text, is output. In this way, the generated text content can be optimized through the hotspot information, further enhancing the topicality and popularity of the text content.

[0120] It should be noted that, in the process of the above-mentioned secondary editing to generate the text content, the user can repeat the above process until a satisfactory target text is obtained.

[0121] In some embodiments, see Figure 5If the target content is image content, that is, after the user selects the image content generation function, the electronic device can display a first description input control 510 and a content generation control 520 in the first content generation page. In addition, the content attribute setting controls that can be displayed by the electronic device include at least one of an image processing function control 530, an image setting control 540, a style setting control 550, a size setting control 560, and a quantity setting control 570.

[0122] On the basis of the above embodiment, S330 can be implemented as at least one of the following steps A′ to E′.

[0123] Step A′: In response to a selection operation on the image processing function control, the selected image processing function identifier is determined as a content attribute value of the model dimension.

[0124] Among them, the image processing function control is an interactive control pointing to a preset image processing function, such as a radio button, a drop-down box, or other selective interactive controls. Each image processing function control corresponds to an image processing function. The image processing function identifier is the identification information of the image processing function, such as the name, abbreviation, or code of the image processing function, which is used to determine the content generation model used to generate the target content. For example, for the image processing function that generates images based on text content (referred to as Wensheng images), its image processing function identifier corresponds to the general generative model of Wensheng images; for the image processing function that replaces the background of an image (referred to as background replacement), its image processing function identifier corresponds to the generative model of Wensheng images that retains the main body of the image; for the image processing function that expands the image (referred to as image extension), its image processing function identifier corresponds to the generative model that generates images based on images and texts, etc.

[0125] Specifically, see Figure 5 , the disclosed embodiment provides a variety of image processing functions, and different image processing functions can be implemented using different content generation models, that is, in the image content generation function mode, each content generation model is pre-trained using the image samples corresponding to the image processing function. In this way, after the user performs a selection operation on a certain image processing function control 530, the electronic device determines the corresponding image processing function identifier as the content attribute value of the model dimension, and when the subsequent model is called, the corresponding content generation model can be located through the image processing function identifier. In this way, different image processing function controls can be used to guide users to select a more suitable content generation model without being aware of the model, thereby further improving the quality of subsequent generated images.

[0126] Step B′: In response to a trigger operation acting on the picture setting control, the determined reference picture is determined as the content attribute value of the picture dimension.

[0127] The picture setting control is an interactive control for selecting pictures, which can point to a picture material library on the network side or to a local picture library. The reference picture is used to provide at least one of a main object, a contour of the main object, and a depth of field parameter.

[0128] Specifically, the process of generating pictures in the related art mainly involves calling a model to generate pictures according to text descriptions, but the quality of the generated pictures is greatly affected by the accuracy of the text descriptions. Therefore, the picture setting function is provided in the embodiment of the present disclosure to guide users to select existing pictures as references for generating picture content (referred to as reference pictures) to make up for the defects of inaccurate text descriptions.

[0129] For specific implementation, see Figure 5 , the electronic device may display a picture setting control 540 with text labels such as "Picture Library" and "Local Upload" in the first content generation page. In response to the user's triggering operation on the picture setting control 540, the electronic device may display a picture selection window and display at least one existing picture as a reference picture in the picture selection window. In this way, the subsequent content generation model can constrain the generation process of the picture content with the help of at least one of the main object in the reference picture, the outline of the main object, and the depth of field parameter of the picture, so as to improve the quality of the picture content.

[0130] Step C′: In response to a selection operation on the style setting control, the selected style identifier is determined as a content attribute value of the style dimension.

[0131] The style setting control is an interactive control for selecting the picture style of a picture. The style identifier is an identifier of a certain picture style, which represents the picture style of the generated target content.

[0132] Specifically, considering that the picture style of a picture has an important influence on the generated picture content and is also highly correlated with user needs, and the content description information input by the user may not accurately describe the picture style, a style setting control is provided in the embodiment of the present disclosure so that the user can determine a more accurate picture style by selecting the style setting control. In order to facilitate the user to understand the differences between different picture styles and thus assist them in selecting a suitable style setting control, the style setting control can be implemented in the form of an interactive control of a graphic label.

[0133] For specific implementation, see Figure 5 , the electronic device may display a plurality of preset style setting controls 550 in the first content generation page. In response to a user triggering operation on a certain style setting control 550, the electronic device may determine the text label of the triggered style setting control 550 as the style identifier selected by the user. For example, Figure 5If the user selects the radio button of "Style 1", the text of "Style 1" is set as the style identifier. The style identifier can be used as a constraint on the picture style in the prompt word, so that the content generation model can generate picture content according to the picture style, thereby improving the degree of fit between the picture content and the user's expectations.

[0134] Step D′: In response to a selection operation on the size setting control, the selected size is determined as a content attribute value of the size dimension.

[0135] The size represents the horizontal and vertical ratios of the generated target content.

[0136] Specifically, see Figure 5 , the electronic device may display multiple size setting controls 560 in the first content generation page. When the user selects a size setting control 560 (such as 16:9), the electronic device may determine the image size of the image content to be generated. In this way, accurate image size and ratio constraints may be provided to the content generation model, thereby further improving the degree of fit between the image content and user expectations / business needs / business scenarios.

[0137] Step E′: in response to a trigger operation acting on the quantity setting control, the determined quantity is determined as a content attribute value of the quantity dimension.

[0138] The quantity represents the amount of generated target content.

[0139] Specifically, see Figure 5 , the electronic device may display a quantity setting control 570 in the first content generation page. The user may input the quantity of the image content to be generated by triggering the quantity setting control 570. In this way, an accurate constraint on the number of images may be provided for the content generation model, so that the model may batch output multiple image contents with certain differences under the same model prompt word, thereby further improving the generation efficiency of the image content.

[0140] In some embodiments, if the target content is picture content, S350 may be implemented as: in response to a triggering operation on a local editing control corresponding to the target content, displaying a picture editing window corresponding to the target content.

[0141] Specifically, after generating the image content, the electronic device may display the selected number of image contents in the first area of ​​the first content generation page. Then, when the user performs a trigger operation on a certain image content, the electronic device may display an editing control that can trigger secondary editing of the image in the surrounding area of ​​the image content. If the user performs a trigger operation on the editing control, the electronic device may display the image editing window corresponding to the editing control in response to the trigger operation.

[0142] Continue to see Figure 5 For the general image processing function of the text map, the electronic device can display four image contents in the first area 580 of the first content generation page. Then, when the user performs a trigger operation such as selecting or long pressing a certain image content, the electronic device can display a local editing control 581 that can edit some elements in the image content and a global editing control 582 that can edit all elements of the image content in the surrounding area of ​​the image content. If the user triggers the local editing control 581, the electronic device can display its corresponding image editing window, such as Figure 6 The editing pop-up window 600 shown in FIG. 600 may display the triggered picture content 610, the main body editing control 620 for editing the main elements of the picture, and the background editing control 630 for editing the background of the picture.

[0143] Based on the above embodiments, continue to refer to Figure 5 S360 can be implemented as follows: responding to a triggering operation on a main editing control in a picture editing window, determining interactive content for editing a main object of a target content; performing editing processing on the target content based on the interactive content, and generating an edited target picture.

[0144] Specifically, if the user performs a trigger operation on the main editing control 620, the electronic device can perform a cutout process through a pre-set image processing algorithm to obtain the main object in the image content. In addition, the electronic device responds to the user's replacement operation, move operation, and other interactive operations on the main object of the content image 610, and determines the result after the operation (such as the main object after replacement, the position of the main object after movement, etc.) as the interactive content. Then, the interactive content is combined with the remaining image elements to complete the partial editing of the image body and generate the edited target image.

[0145] Based on the above embodiment, S360 can be implemented as: responding to the trigger operation of the background editing control in the picture editing window, determining the interactive content for editing the background of the target content; performing editing processing on the target content based on the interactive content, and generating an edited target picture.

[0146] Specifically, if the user performs a trigger operation on the background editing control 630, the electronic device can respond to the user's interactive operations such as filter operation, replacement operation, etc. on the background of the content image 610, and determine the result after the operation (such as the background after filtering, the background after replacement, etc.) as the interactive content. Then, the interactive content is combined with the remaining image elements to complete the partial editing of the image background and generate the edited target image.

[0147] See also Figure 7For the general background replacement image processing function, the electronic device may display a picture content that replaces the background of the reference picture (for example, a diagonal fill) in the first area 710 of the first content generation page. Then, when the user performs a triggering operation such as selecting or long pressing the picture content, the electronic device may display a local editing control in the surrounding area of ​​the picture content that can edit the main elements (also called main objects) in the picture content. If the user triggers the local editing control, the electronic device may use the first area as the corresponding picture editing window, and in the picture editing window, the main object 720 (such as Figure 7 The bottle in the image is set to an active state that can be moved and scaled.

[0148] Based on the above embodiments, continue to refer to Figure 7 S360 can be implemented as follows: in response to an interactive operation on a main object of the image content, determining the interactive content; and performing editing processing on the target content based on the interactive content to generate an edited target image.

[0149] Specifically, see Figure 7 , the user can perform interactive operations of moving and / or scaling the active subject object 720, and the result is interactive content, such as the moved position corresponding to the move operation, the position and range of the subject object after being enlarged or reduced corresponding to the scaling operation, etc. Then, the interactive content is combined with the remaining image elements to complete the partial editing of the subject object under the image processing function of background replacement, and generate the edited target image 730.

[0150] Continue to see Figure 7 For the general image processing function of image extension, the electronic device may display an image content 740 that expands the background of the reference image (using a diagonal fill example) in the first area 710 of the first content generation page. The range enclosed by the dotted line of the image content 740 is the original size of the reference image, and the range outside the dotted line range is the expanded background image. When the user performs a triggering operation such as selecting or long pressing the image content, the electronic device may display a local editing control in the surrounding area of ​​the image content that can edit the original reference image in the image content (i.e., the image within the dotted line range). If the user triggers the local editing control, the electronic device may use the first area as the corresponding image editing window, and set the main image part 741 within the range of the original reference image in the image content to a movable active state in the image editing window.

[0151] Based on the above embodiments, continue to refer to Figure 7S360 can be implemented as follows: in response to an interactive operation on a main image portion of the image content, determining interactive content; and performing editing processing on the target content based on the interactive content to generate an edited target image.

[0152] Specifically, see Figure 7 , the user can perform an interactive operation of moving the main image part 741 in an active state, and determine the position after the move corresponding to the move operation as the interactive content. Then, the corresponding content generation model is re-called according to the interactive content, so that the model regenerates the extended image background part according to the moved main image part 741 to obtain the edited target image. In this way, the background part of the existing reference image can be extended and drawn in a specified direction by moving the main image part 741, so as to quickly obtain new image content.

[0153] In some embodiments, if the target content is picture content, S350 may also be implemented as: in response to a triggering operation on a global editing control corresponding to the target content, an editor page corresponding to the target content is displayed.

[0154] Specifically, see Figure 5 When the user performs a triggering operation such as selecting or long pressing a certain image content, the electronic device can display a global editing control 582 in the surrounding area of ​​the image content. If the user triggers the global editing control 582, the electronic device can display its corresponding editor page, such as Figure 8 The editor page 800 shown in FIG. The editor page 800 may display triggered image content 810, a layer control 820 for editing the layer of the image content, a component control 830 for adding a new component to the image content, an element control 840 for adding a new element to the image content, a template control 850 for changing the overall layout of the image content, and an image processing tool 860 for automatically optimizing the image content (e.g., automatically making the overall image clearer).

[0155] Based on the above embodiments, S360 can be implemented as follows: responding to the triggering operation of the editing function control in the editor page, determining the interactive content corresponding to the triggered editing function; performing editing processing on the target content based on each interactive content, and generating an edited target image.

[0156] Specifically, the user can perform a trigger operation on at least one interactive control in the editor page 800, and perform the corresponding interactive operation according to the guidance of the triggered interactive control, and use the result obtained after the interactive operation ends as the interactive content of this editing. For example, the user can trigger the component control 830, and the electronic device will display multiple components corresponding to text, picture and other categories; then after the user triggers a component under the text category, the electronic device can add the triggered component to the preset position of the picture content 810 (such as the middle position above the picture). The newly added component at the preset position is used as the interactive content of this editing. Then, the electronic device can combine the interactive content with the remaining picture elements to generate the edited target picture.

[0157] It should be noted that the above interactive operation on the editor page 800 can be a single editing process or a series of editing processes that repeat the above process. After each editing, the edited target image can be obtained. After all editing is completed, the final target image can be obtained. In this way, the editor can be used to implement in-depth secondary editing of the generated image content, further improving the degree of fit between the image content and the user's expectations, thereby further improving the quality of the image content and the generation efficiency.

[0158] In some embodiments, see Fig. 9 If the target content is multimedia content, that is, after the user selects the multimedia content generation function, the electronic device can display a first description input control 910 (i.e., a title setting control for setting the main and subtitles of the multimedia content) and a content generation control 920 in the first content generation page. In addition, the content attribute setting controls that can be displayed by the electronic device include at least one of a product type setting control 930, a product attribution party information setting control 940, a layout setting control 950, and a quantity setting control 960.

[0159] It should be noted that see Fig. 9 , the first description input control 910 can be manually input by the user into the main and subtitles; or the interactive control of "automatically generate titles" can be triggered to display the interactive window 970 for automatically generating titles. The user can input his / her requirements for titles through the interactive window 970, and then the electronic device can be triggered to call the natural language model to automatically generate multiple groups of main and subtitles based on the title requirements input by the user. A "use" button is displayed behind each group of main and subtitles, so that the user can select the main and subtitles by triggering the button, and the selected main and subtitles are filled in the corresponding position of the first description input control 910 for display.

[0160] Based on the above embodiment, S330 can be implemented as at least one of the following steps A" to D".

[0161] Step A″: in response to a selection operation on a product type setting control, the selected product type identifier is determined as a content attribute value of the product dimension.

[0162] Among them, the product type identifier represents the product to which the target content is applicable.

[0163] Specifically, different products have different characteristics, and the corresponding multimedia content has different content focuses, and the overall framework and overall style of the multimedia content required by them may also be different. Fig. 9 The first content generation page shown displays a product type setting control 930 in the form of a radio button or a drop-down box or other selective control. When the user selects a product type setting control, the electronic device can determine the corresponding text content as the product type identifier selected by the user. Fig. 9 If the user selects the radio button of "makeup", the "makeup" text is set as the product type identifier. The product type identifier allows the content generation model to combine the content focus and overall style required by the product type, so that the generated multimedia content is more in line with the needs of the corresponding product and its usage scenarios.

[0164] For instance, taking the poster as an example, the content generation model can filter out poster subjects that are suitable for the product type and its style from a number of poster subjects based on the product type identifier; and, based on the filtered poster subjects, the learned element layout framework and element color matching rules, etc., a poster background can be generated; then, according to the poster subjects and poster background, and based on the learned element layout framework and element color matching rules, the obtained main and sub-titles, promotional texts, promotional pictures and subsequent product ownership information can be filled into the corresponding positions of the poster to generate the final poster.

[0165] Step B″: in response to a trigger operation acting on the information setting control, the determined product ownership information is determined as a content attribute value of the product ownership dimension.

[0166] Among them, the product owner information includes the icon of the product owner (such as a brand logo), the first access path of the product owner (such as the brand's official website) or the first encoded image corresponding to the first access path (such as a QR code storing the brand's official website), the second access path of the target product corresponding to the target content (such as the access website of the target product) or the second encoded image corresponding to the second access path (such as a QR code storing the access website of the target product), and at least one of the audio corresponding to the target product and the video corresponding to the target product.

[0167] Specifically, multimedia content can be used to promote target products. In addition to the promotional text and pictures related to the target products, multimedia content can also include product ownership information, so that users can obtain more information related to the target products through the product ownership information. Fig. 9 The first content generation page shown displays an information setting control 940 in the form of a control with a selection function, which may be at least one of an icon selection control, an access path selection control, and other information selection controls. The user may select the attribution party information that he or she wants to add to the multimedia content by triggering the information setting control 940. In this way, the electronic device may use the content generation model to embed the obtained product attribution party information as filler material into the multimedia content, thereby enriching the information type and amount of the multimedia content.

[0168] Step C″, in response to the selection operation on the layout setting control, determining the selected content layout identifier as the content attribute value of the element layout dimension,

[0169] Among them, the content layout identifier represents the horizontal and vertical ratios, size and element layout of the target content.

[0170] Specifically, multimedia content also has attributes such as content size and horizontal and vertical ratio. Therefore, the embodiments of the present disclosure can be used in Fig. 9 The first content generation page shown displays a layout setting control 950 in the form of a radio button or a drop-down box. The user can select the desired layout of the multimedia content by triggering the layout setting control 950, and determine the text content corresponding to the triggered layout setting control 950 as the content layout identifier. The content layout identifier can constrain the horizontal and vertical ratios, size, and layout of the internal elements of the content in the content generation model, thereby improving the quality and naturalness of the multimedia content.

[0171] Step D″: in response to a trigger operation acting on the quantity setting control, the determined quantity is determined as a content attribute value of the quantity dimension.

[0172] The quantity represents the amount of generated target content.

[0173] Specifically, see Fig. 9 , the electronic device may display a quantity setting control 960 in the first content generation page. The user may input the quantity of multimedia content that needs to be generated by triggering the quantity setting control 960. In this way, an accurate quantity constraint may be provided for the content generation model, so that the model may batch output multiple multimedia contents with certain differences under the same model prompt word, thereby further improving the generation efficiency of multimedia content.

[0174] In some embodiments, if the target content is multimedia content, S350 may be implemented as: in response to a triggering operation acting on a global editing control corresponding to the target content, displaying an editor page corresponding to the target content.

[0175] Specifically, after generating multimedia content, the electronic device may display a selected amount of multimedia content in the first area of ​​the first content generation page. Then, when the user performs a trigger operation on a certain multimedia content, the electronic device may display an editing control that can trigger secondary editing of the multimedia in the surrounding area of ​​the multimedia content. If the user performs a trigger operation on the editing control, the electronic device may display a multimedia editing window corresponding to the editing control in response to the trigger operation.

[0176] Continue to see Fig. 9 , the electronic device may display two posters in the first area 900 of the first content generation page. Then, when the user performs a triggering operation such as selecting or long pressing a poster, the electronic device may display a global editing control 980 in the surrounding area of ​​the multimedia content. If the user triggers the global editing control 980, the electronic device may display its corresponding editor page, such as Figure 8 The editor page 800 shown is just that the picture content 810 displayed in the editor page 800 is replaced by the triggered poster, and the remaining displayed controls remain unchanged.

[0177] Based on the above embodiments, continue to refer to Fig. 9 , S360 can be implemented as: responding to the triggering operation of the editing function control in the editor page, determining the interactive content corresponding to the triggered editing function; performing editing processing on the target content based on each interactive content, and generating edited target multimedia content.

[0178] Specifically, according to the above-mentioned instructions for performing global editing on the image content, secondary editing of the multimedia content can be implemented to generate edited target multimedia content.

[0179] In some embodiments, if the target content is multimedia content, S350 can also be implemented as: in response to a trigger operation acting on a multi-size editing control corresponding to the target content, a second content generation page corresponding to the target content is displayed, and at least one content size selection control is displayed in the second content generation page.

[0180] The multi-size editing control is an interactive control for triggering a multi-size editing function for generating multimedia contents of different sizes.

[0181] Specifically, see Fig. 9In response to the user performing a triggering operation such as selecting or long pressing a poster, the electronic device may also display a multi-size editing control 990 in the surrounding area of ​​the multimedia content. If the user triggers the multi-size editing control 990, the electronic device may display its corresponding editing interaction window. Given that the multi-size editing function generates the same multimedia content of different sizes based on the generated multimedia content, the editing interaction window may be referred to as the second content generation page, such as Fig.10 As shown. In the second content generation page 1000, a selection control of different sizes is displayed, namely, a content size selection control 1010. The content size selection control 1010 can display size information such as "900×383", or a check box to indicate that multiple sizes can be selected at the same time to generate multiple multimedia contents. In addition, a custom size control 1020 can also be displayed in the second content generation page 1000 to guide the user to enter custom size information other than the preset size.

[0182] Based on the above embodiments, continue to refer to Fig.10 S360 can be implemented as follows: responding to a selection operation on a content size selection control, determining a selected content size identifier, and responding to a trigger operation on a size expansion control, re-calling a content generation model based on the content size identifier and target content to generate a target multimedia content with an expanded size.

[0183] Specifically, if the user selects at least one content size selection control 1010 or a custom size control 1020, the electronic device can use its size information as a content size identifier. Then, the previously generated model prompt word is updated using the content size identifier, and the model prompt word and the generated target content (such as Fig.10 The selected poster shown in the figure is input into the content generation model, so that the content generation model changes the size and element layout of the content while retaining the main object, background, text elements, picture elements, audio and video elements, etc. of the target multimedia content, and generates the target multimedia content corresponding to the content size identifier. Fig.10 In the example, the content generation model can generate a poster 1030 corresponding to the first size "900×383", a poster 1040 corresponding to the second size "800×800", and a poster 1050 corresponding to the third size "900×500" based on the selected poster, and display them in the second content generation page. In this way, the interactive function of generating multimedia content of multiple sizes with one click can be supported through the setting of the multi-size editing control, thereby further improving the efficiency of the interactive operation of generating multimedia content and the efficiency of generating multimedia content.

[0184] Based on the above embodiment, if the user performs a trigger operation on a target of a certain size on the media content, the electronic device can highlight the triggered target multimedia content and display the global editing controls around it. Fig.10 If the poster 1030 corresponding to the first size "900×383" in the second content generation page 1000 is displayed, the electronic device highlights the poster 1030 in the middle area of ​​the second content generation page 1000, and displays a global editing control and a download control to provide secondary editing and storage capabilities.

[0185] The following is an embodiment of a content generating device provided by an embodiment of the present invention. The device and the content generating methods of the above-mentioned embodiments belong to the same inventive concept. For details not described in detail in the embodiment of the content generating device, reference can be made to the embodiment of the above-mentioned content generating method.

[0186] Fig.11 FIG. 1 is a schematic diagram showing the structure of a content generation device provided by an embodiment of the present disclosure. Fig.11 As shown, the content generating device 1100 may include:

[0187] A first content generation page display module 1110, configured to respond to a triggering operation on a content generation function corresponding to a target content, display a first content generation page corresponding to the content generation function, and display a first description input control and a content generation control in the first content generation page;

[0188] A content description information determination module 1120, configured to determine content description information in response to an input operation applied to a first description input control;

[0189] The target content generation module 1130 is used to respond to the trigger operation acting on the content generation control, call the content generation model corresponding to the content generation function based on the content description information, and generate the target content;

[0190] The editing page display module 1140 is used to respond to the triggering operation of the editing control corresponding to the target content and display the editing page of the target content;

[0191] The edited target content generating module 1150 is used to respond to the interactive operation acting on the editing page, edit the target content based on the interactive content corresponding to the interactive operation, and generate the edited target content.

[0192] The content generation device provided by the embodiment of the present disclosure can respond to a trigger operation on a content generation function corresponding to a target content, display a first content generation page corresponding to the content generation function, and display a first description input control and a content generation control in the first content generation page; respond to an input operation on the first description input control, determine content description information; respond to a trigger operation on the content generation control, call a content generation model corresponding to the content generation function based on the content description information, and generate target content; respond to a trigger operation on an edit control corresponding to the target content, display an edit page of the target content; respond to an interactive operation on the edit page, edit the target content based on the interactive content corresponding to the interactive operation, and generate edited target content; after the target content is generated by using a generative model, a secondary editing function for the target content is provided, so that the user can perform personalized editing on the target content through the edit page of the target content, and generate edited target content that better meets the user's expectations, thereby improving the generation efficiency of the target content by using the generative model, and improving the personalization of the target content and its degree of fit with business needs, thereby improving the generation quality of the target content.

[0193] In some embodiments, the first content generation page display module 1110 is further configured to:

[0194] A content attribute setting control is also displayed in the first content generation page;

[0195] Accordingly, the content generating device 1100 further includes a content attribute value determining module, which is used to:

[0196] In response to a triggering operation acting on a content attribute setting control, determining a content attribute value;

[0197] Accordingly, the target content generation module 1130 is specifically used for:

[0198] In response to a trigger operation acting on a content generation control, based on content description information and content attribute values, a content generation model corresponding to the content generation function is called to generate target content; wherein the content generation model is obtained by training a generative model using training samples corresponding to the content generation function.

[0199] In some embodiments, the target content includes at least one of text content, picture content, and multimedia content.

[0200] In some embodiments, if the target content is text-type content, the content attribute setting control includes at least one of an industry setting control, a scene setting control, a crowd category setting control, a channel setting control, a hotspot setting control, and a model setting control;

[0201] Accordingly, the content attribute value determination module is specifically configured to determine the content attribute value in response to a trigger operation acting on the content attribute setting control by at least one of the following:

[0202] In response to a selection operation on an industry setting control, a selected industry identifier is determined as a content attribute value of an industry dimension; the industry identifier represents an industry to which the target content is applicable;

[0203] In response to a selection operation acting on a scene setting control, determining the selected scene identifier as a content attribute value of the scene dimension; the scene identifier represents a usage scenario of the target content;

[0204] In response to a selection operation on a crowd category setting control, the selected crowd category identifier is determined as a content attribute value of an applicable object dimension; the crowd category identifier represents the category of the crowd to which the target content is applicable;

[0205] In response to a selection operation acting on a channel setting control, determining a selected channel identifier as a content attribute value of a channel dimension; the channel identifier represents a channel for distributing the target content;

[0206] In response to a selection operation on a hotspot setting control, the selected first hotspot identifier is determined as a content attribute value of a heat dimension; the hotspot identifier is an identifier of information whose information heat meets a preset heat within a preset time period, and the hotspot identifier is used to enhance the heat of the target content;

[0207] In response to a selection operation on a model setting control, a selected model identifier is determined as a content attribute value of a model dimension; the model identifier is an identifier of a content generation model used to generate target content.

[0208] In some embodiments, the edit page display module 1140 is specifically used to:

[0209] In response to a triggering operation on a text optimization control corresponding to the target content, a second description input control corresponding to the content description information of the target content is displayed;

[0210] Alternatively, in response to a triggering operation on a hotspot change control corresponding to the target content, a hotspot selection window is displayed, and a plurality of candidate hotspot information is displayed in the hotspot selection window.

[0211] Furthermore, the edited target content generation module 1150 is specifically used for:

[0212] If the editing page includes a second description input control, in response to an input operation acting on the second description input control, determining the input content as interactive content;

[0213] updating the content description information based on the interactive content, and re-calling the content generation model based on the updated content description information to generate an edited target text;

[0214] Alternatively, if the editing page includes a hotspot selection window, in response to a selection operation acting on the hotspot selection window, the selected second hotspot identifier is determined as the interactive content;

[0215] Based on the interactive content and content description information, the content generation model is re-called to generate the edited target text.

[0216] In some embodiments, if the target content is picture-type content, the content attribute setting control includes at least one of a picture processing function control, a picture setting control, a style setting control, a size setting control, and a quantity setting control;

[0217] Accordingly, the content attribute value determination module is specifically configured to determine the content attribute value in response to a trigger operation acting on the content attribute setting control by at least one of the following:

[0218] In response to a selection operation on the image processing function control, the selected image processing function identifier is determined as a content attribute value of the model dimension; the image processing function identifier is used to determine a content generation model used to generate the target content;

[0219] In response to a trigger operation acting on a picture setting control, determining a determined reference picture as a content attribute value of a picture dimension; the reference picture is used to provide at least one of a subject object, an outline of the subject object, and a depth of field parameter;

[0220] In response to a selection operation on the style setting control, the selected style identifier is determined as a content attribute value of the style dimension; the style identifier represents the screen style of the generated target content;

[0221] In response to a selection operation on a size setting control, the selected size is determined as a content attribute value of the size dimension; the size represents a horizontal and vertical ratio of the generated target content;

[0222] In response to a trigger operation acting on a quantity setting control, the determined quantity is determined as a content attribute value of a quantity dimension; the quantity represents the quantity of the generated target content.

[0223] In some embodiments, the edit page display module 1140 is specifically used to:

[0224] In response to a triggering operation on a local editing control corresponding to the target content, displaying a picture editing window corresponding to the target content;

[0225] Alternatively, in response to a triggering operation acting on a global editing control corresponding to the target content, an editor page corresponding to the target content is displayed.

[0226] Furthermore, the edited target content generation module 1150 is specifically used for:

[0227] If the editing page includes a picture editing window, then in response to a triggering operation of a body editing control in the picture editing window, determining interactive content for editing a body object of the target content, or in response to a triggering operation of a background editing control in the picture editing window, determining interactive content for editing a background of the target content;

[0228] Performing editing processing on the target content based on the interactive content to generate an edited target image;

[0229] Alternatively, if the edit page includes an editor page, in response to a triggering operation acting on an edit function control in the editor page, determining interactive content corresponding to the triggered edit function;

[0230] The target content is edited based on each interactive content to generate an edited target image.

[0231] In some embodiments, if the target content is multimedia content, the content attribute setting control includes at least one of a product type setting control, a product attribution party information setting control, a layout setting control, and a quantity setting control;

[0232] Accordingly, the content attribute value determination module is specifically configured to determine the content attribute value in response to a trigger operation acting on the content attribute setting control by at least one of the following:

[0233] In response to a selection operation on a product type setting control, determining a selected product type identifier as a content attribute value of a product dimension; the product type identifier represents a product to which the target content applies;

[0234] In response to a trigger operation acting on an information setting control, the determined product attribution party information is determined as a content attribute value of a product attribution party dimension; the product attribution party information includes at least one of an icon of the product attribution party, a first access path of the product attribution party or a first encoded image corresponding to the first access path, a second access path of a target product corresponding to the target content or a second encoded image corresponding to the second access path, audio corresponding to the target product, and video corresponding to the target product;

[0235] In response to a selection operation on a layout setting control, the selected content layout identifier is determined as a content attribute value of an element layout dimension; the content layout identifier represents the horizontal and vertical ratios, dimensions, and element layout of the target content.

[0236] In some embodiments, the edit page display module 1140 is specifically used to:

[0237] In response to a triggering operation on a global editing control corresponding to the target content, an editor page corresponding to the target content is displayed;

[0238] Alternatively, in response to a triggering operation acting on a multi-size editing control corresponding to the target content, a second content generation page corresponding to the target content is displayed, and at least one content size selection control is displayed in the second content generation page.

[0239] Furthermore, the edited target content generation module 1150 is specifically used for:

[0240] If the edit page includes an editor page, responding to a triggering operation on an edit function control in the editor page, determining interactive content corresponding to the triggered edit function;

[0241] Performing editing processing on target content based on each interactive content to generate edited target multimedia content;

[0242] If the editing page includes a second content generation page, then in response to a selection operation on a content size selection control, the selected content size identifier is determined, and in response to a trigger operation on a size expansion control, based on the content size identifier and the target content, the content generation model is re-called to generate the target multimedia content of the expanded size.

[0243] In some embodiments, the content generation device 1100 further includes:

[0244] A control display module, configured to, in response to a trigger operation acting on a content generation control, call a content generation model corresponding to the content generation function based on content description information, generate target content, display the target content in a first area of ​​a first content generation page, and display a content storage control and / or a historical content viewing control on the first content generation page;

[0245] A content storage module, configured to respond to a trigger operation acting on a content storage space and store the target content into a storage space corresponding to the trigger operation;

[0246] And / or, a content information display module is used to display content information of historical content generated within a historical time period in a second area of ​​the first content generation page in response to a trigger operation acting on the historical content viewing control.

[0247] The content generating device provided in the embodiment of the present invention can execute the content generating method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0248] It is worth noting that in the embodiment of the above-mentioned content generation device, the various units and modules included are only divided according to functional logic, but are not limited to the above-mentioned division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present disclosure.

[0249] The disclosed embodiment also provides an electronic device, which may include a processor and a memory, wherein the memory may be used to store executable instructions, wherein the processor may be used to read the executable instructions from the memory and execute the executable instructions to implement the content generation method in the above embodiment.

[0250] Fig.12 A schematic structural diagram of an electronic device provided by an embodiment of the present disclosure is shown.

[0251] like Fig.12 As shown, the electronic device 1200 may include a processing device 1201 (e.g., a central processing unit, a graphics processor, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1202 or a program loaded from a storage device 1208 to a random access memory (RAM) 1203. In the RAM 1203, various programs and data required for the operation of the electronic device 1200 are also stored. The processing device 1201, the ROM 1202, and the RAM 1203 are connected to each other via a bus 1204. An input / output interface (I / O interface) 1205 is also connected to the bus 1204.

[0252] Typically, the following devices may be connected to the I / O interface 1205: an input device 1206 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 1207 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 1208 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1209. The communication device 1209 may allow the electronic device 1200 to communicate with other devices wirelessly or by wire to exchange data.

[0253] It should be noted that Fig.12 The electronic device 1200 shown is only an example and should not limit the functions and scope of use of the embodiments of the present disclosure. Fig.12 The electronic device 1200 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead.

[0254] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program contains program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device 1209, or installed from a storage device 1208, or installed from a ROM 1202. When the computer program is executed by the processing device 1201, the above-mentioned functions defined in the content generation method of any embodiment of the present disclosure are executed.

[0255] The embodiments of the present disclosure further provide a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the processor implements the content generation method in any embodiment of the present disclosure.

[0256] It should be noted that the computer-readable medium disclosed above may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, device or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0257] In some embodiments, the client and server may communicate using any currently known or future developed network protocol such as HTTP, and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.

[0258] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0259] The computer-readable medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device executes the steps of the content generation method described in any embodiment of the present disclosure.

[0260] In embodiments of the present disclosure, computer program code for performing the operations of the present disclosure may be written in one or more programming languages ​​or a combination thereof, including but not limited to object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on a user's computer, partially on a user's computer, as a separate software package, partially on a user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0261] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the equipment, method and computer program product according to various embodiments of the present disclosure. In this regard, each box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0262] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chip (SOCs), complex programmable logic devices (CPLDs), and the like.

[0263] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0264] The above description is only a preferred embodiment of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of disclosure involved in the present disclosure is not limited to the technical solutions formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in the present disclosure (but not limited to) by each other to form a technical solution.

[0265] In addition, although each operation is described in a specific order, this should not be understood as requiring these operations to be performed in the specific order shown or in a sequential order. Under certain circumstances, multitasking and parallel processing may be advantageous. Similarly, although some specific implementation details are included in the above discussion, these should not be interpreted as limiting the scope of the present disclosure. Some features described in the context of a separate embodiment can also be implemented in a single embodiment in combination. On the contrary, the various features described in the context of a single embodiment can also be implemented in multiple embodiments individually or in any suitable sub-combination mode.

[0266] Although the subject matter has been described in language specific to structural features and / or methodological logical actions, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. On the contrary, the specific features and actions described above are merely example forms of implementing the claims.

Claims

1. A content generation method, characterized in that: include: In response to a triggering operation acting on a content generation function corresponding to the target content, displaying a first content generation page corresponding to the content generation function, and displaying a first description input control and a content generation control in the first content generation page; In response to an input operation acting on the first description input control, determining content description information; In response to a trigger operation acting on the content generation control, based on the content description information, a content generation model corresponding to the content generation function is called to generate the target content; In response to a triggering operation on an editing control corresponding to the target content, an editing page of the target content is displayed; In response to an interactive operation on the editing page, the target content is edited based on interactive content corresponding to the interactive operation to generate edited target content.

2. The method according to claim 1, characterized in that: The method further comprises: A content attribute setting control is also displayed in the first content generation page; In response to a triggering operation acting on the content attribute setting control, determining a content attribute value; The step of responding to the triggering operation of the content generation control, calling the content generation model corresponding to the content generation function based on the content description information, and generating the target content includes: In response to a trigger operation acting on the content generation control, based on the content description information and the content attribute value, a content generation model corresponding to the content generation function is called to generate the target content; wherein the content generation model is obtained by training a generative model using training samples corresponding to the content generation function.

3. The method according to claim 1 or 2, characterized in that: The target content includes at least one of text content, picture content and multimedia content.

4. The method according to claim 2, characterized in that: If the target content is text content, the content attribute setting control includes at least one of an industry setting control, a scene setting control, a crowd category setting control, a channel setting control, a hotspot setting control, and a model setting control; The determining of the content attribute value in response to the triggering operation acting on the content attribute setting control includes at least one of the following: In response to a selection operation on the industry setting control, determining a selected industry identifier as the content attribute value of the industry dimension; the industry identifier represents the industry to which the target content is applicable; In response to a selection operation on the scene setting control, determining a selected scene identifier as the content attribute value of the scene dimension; the scene identifier represents a usage scenario of the target content; In response to a selection operation on the crowd category setting control, a selected crowd category identifier is determined as the content attribute value of the applicable object dimension; the crowd category identifier represents the category of the crowd to which the target content is applicable; In response to a selection operation on the channel setting control, determining a selected channel identifier as the content attribute value of the channel dimension; the channel identifier represents a channel for distributing the target content; In response to a selection operation on the hotspot setting control, the selected first hotspot identifier is determined as the content attribute value of the heat dimension; the hotspot identifier is an identifier of information whose information heat meets a preset heat within a preset time period, and the hotspot identifier is used to enhance the heat of the target content; In response to a selection operation on the model setting control, a selected model identifier is determined as the content attribute value of the model dimension; the model identifier is an identifier of the content generation model used to generate the target content.

5. The method according to claim 3, characterized in that: If the target content is the text content, the step of responding to the triggering operation of the editing control corresponding to the target content to display the editing page of the target content includes: In response to a triggering operation on a text optimization control corresponding to the target content, displaying a second description input control corresponding to the content description information of the target content; Alternatively, in response to a triggering operation on a hotspot change control corresponding to the target content, a hotspot selection window is displayed, and a plurality of candidate hotspot information is displayed in the hotspot selection window.

6. The method according to claim 5, characterized in that The step of responding to the interactive operation on the editing page and editing the target content based on the interactive content corresponding to the interactive operation to generate the edited target content includes: If the editing page includes the second description input control, in response to an input operation acting on the second description input control, determining the input content as the interactive content; updating the content description information based on the interactive content, and re-calling the content generation model based on the updated content description information to generate an edited target text; Alternatively, if the editing page includes the hotspot selection window, in response to a selection operation on the hotspot selection window, the selected second hotspot identifier is determined as the interactive content; Based on the interactive content and the content description information, the content generation model is recalled to generate an edited target text.

7. The method according to claim 2, characterized in that If the target content is picture content, the content attribute setting control includes at least one of a picture processing function control, a picture setting control, a style setting control, a size setting control, and a quantity setting control; The determining of the content attribute value in response to the triggering operation acting on the content attribute setting control includes at least one of the following: In response to a selection operation on the image processing function control, determining the selected image processing function identifier as the content attribute value of the model dimension; the image processing function identifier is used to determine the content generation model used to generate the target content; In response to a trigger operation acting on the picture setting control, determining a determined reference picture as the content attribute value of the picture dimension; the reference picture is used to provide at least one of a subject object, a contour of the subject object, and a depth of field parameter; In response to a selection operation on the style setting control, determining a selected style identifier as the content attribute value of the style dimension; the style identifier represents the screen style of the generated target content; In response to a selection operation on the size setting control, determining a selected size as the content attribute value of the size dimension; the size represents a horizontal and vertical ratio of the generated target content; In response to a trigger operation acting on the quantity setting control, the determined quantity is determined as the content attribute value of the quantity dimension; the quantity represents the quantity of the generated target content.

8. The method according to claim 3, characterized in that If the target content is the image content, the step of responding to the triggering operation of the editing control corresponding to the target content to display the editing page of the target content includes: In response to a triggering operation on a local editing control corresponding to the target content, displaying a picture editing window corresponding to the target content; Alternatively, in response to a triggering operation acting on a global editing control corresponding to the target content, an editor page corresponding to the target content is displayed.

9. The method according to claim 8, characterized in that The step of responding to the interactive operation on the editing page and editing the target content based on the interactive content corresponding to the interactive operation to generate the edited target content includes: If the editing page includes the picture editing window, then in response to a triggering operation of a main body editing control in the picture editing window, determining the interactive content for editing the main body object of the target content, or in response to a triggering operation of a background editing control in the picture editing window, determining the interactive content for editing the background of the target content; Performing editing processing on the target content based on the interactive content to generate an edited target image; Alternatively, if the editing page includes the editor page, in response to a triggering operation acting on an editing function control in the editor page, determining the interactive content corresponding to the triggered editing function; The target content is edited based on each of the interactive contents to generate an edited target image.

10. The method according to claim 2, characterized in that If the target content is multimedia content, the content attribute setting control includes at least one of a product type setting control, a product owner information setting control, a layout setting control, and a quantity setting control; The step of determining the content attribute value in response to a triggering operation acting on the content attribute setting control includes: In response to a selection operation on the product type setting control, determining the selected product type identifier as the content attribute value of the product dimension; The product type identifier represents the product to which the target content is applicable; In response to a trigger operation acting on the information setting control, the determined product attribution party information is determined as the content attribute value of the product attribution party dimension; the product attribution party information includes at least one of an icon of the product attribution party, a first access path of the product attribution party or a first encoded image corresponding to the first access path, a second access path of a target product corresponding to the target content or a second encoded image corresponding to the second access path, audio corresponding to the target product, and video corresponding to the target product; In response to a selection operation on the layout setting control, the selected content layout identifier is determined as the content attribute value of the element layout dimension; the content layout identifier represents the horizontal and vertical ratio, size and element layout of the target content.

11. The method according to claim 3, characterized in that If the target content is the multimedia content, the step of responding to the triggering operation of the editing control corresponding to the target content to display an editing page of the target content includes: In response to a triggering operation on a global editing control corresponding to the target content, displaying an editor page corresponding to the target content; Alternatively, in response to a triggering operation on a multi-size editing control corresponding to the target content, a second content generation page corresponding to the target content is displayed, and at least one content size selection control is displayed in the second content generation page.

12. The method according to claim 11, characterized in that The step of responding to the interactive operation on the editing page and editing the target content based on the interactive content corresponding to the interactive operation to generate the edited target content includes: If the editing page includes the editor page, in response to a triggering operation acting on an editing function control in the editor page, determining the interactive content corresponding to the triggered editing function; Performing editing processing on the target content based on each of the interactive contents to generate edited target multimedia content; If the editing page includes the second content generation page, then in response to a selection operation on the content size selection control, the selected content size identifier is determined, and in response to a trigger operation on the size expansion control, based on the content size identifier and the target content, the content generation model is re-called to generate the target multimedia content with an expanded size.

13. The method according to claim 1, characterized in that After the response to the triggering operation acting on the content generation control, based on the content description information, calling the content generation model corresponding to the content generation function to generate the target content, the method further includes: Displaying the target content in a first area of ​​the first content generation page, and displaying a content storage control and / or a historical content viewing control in the first content generation page; In response to a triggering operation acting on the content storage space, storing the target content in the storage space corresponding to the triggering operation; And / or, in response to a triggering operation acting on the historical content viewing control, content information of historical content generated within a historical time period is displayed in a second area of ​​the first content generation page.

14. A content generating device, characterized in that: include: A first content generation page display module, configured to respond to a triggering operation on a content generation function corresponding to a target content, display a first content generation page corresponding to the content generation function, and display a first description input control and a content generation control in the first content generation page; a content description information determination module, configured to determine content description information in response to an input operation acting on the first description input control; A target content generation module, configured to respond to a trigger operation acting on the content generation control, call a content generation model corresponding to the content generation function based on the content description information, and generate the target content; An editing page display module, configured to display an editing page of the target content in response to a triggering operation on an editing control corresponding to the target content; The edited target content generation module is used to respond to the interactive operation acting on the editing page, edit the target content based on the interactive content corresponding to the interactive operation, and generate the edited target content.

15. An electronic device, characterized in that: include: processor; A memory for storing executable instructions; The processor is used to read the executable instructions from the memory and execute the executable instructions to implement the content generation method described in any one of claims 1 to 13.

16. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the processor implements the content generation method according to any one of claims 1 to 13.

17. A computer program product, characterized in that The computer program product is used to implement the content generation method according to any one of claims 1 to 13.

Images