Novel generation method and device based on text storage box, electronic equipment and medium

Abstract

The invention provides a novel generation method and device based on a text storage box, electronic equipment and a medium, and the method comprises the steps: firstly displaying a first novel generation page which comprises a first text display area and a first text storage area; the user respectively selects N groups of candidate historical texts in the first text display area and respectively drags the N groups of candidate historical texts to a blank text storage box of the first text storage area to obtain N candidate text storage boxes, and then a second novel generation page is displayed and comprises a second text display area and a second text storage area, the N candidate text storage boxes are displayed in a second text storage area, a user selects a target text storage box in the second text storage area and drags a target group historical text in the target text storage box to a target position in a second text display area, and the target group historical text can be moved to the target position; and finally, the novel generated and displayed in the second text display area contains the historical text of the target group. According to the invention, the novel generation efficiency is effectively improved.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, and particularly to a novel generation method, device, electronic device, and medium based on a text storage box. Background Art

[0002] With the continuous development of artificial intelligence technology, the novel generation technology based on artificial intelligence has gradually emerged. It can help people improve writing efficiency, so the importance of this technology has become increasingly prominent. During the novel generation process, for the large amount of content generated by artificial intelligence, there may be some content that meets the user's preferences. The user not only needs to use this content at the current position but also wants to use it at other positions in the novel. The existing technologies mainly use two methods when using this content at other positions. One is to manually flip to the original position of this content by memory, copy the content, and then flip back to the current position for pasting. The other is to search the full text or within a specified range by setting keywords, and then search through all the found content one by one to find the required content. Finally, the content also needs to be copied and then flipped back to the current position for pasting.

[0003] However, for a long novel generated by artificial intelligence, it is very difficult to find the content that meets the user's preferences from a large number of paragraphs through manual searching. Although the method of searching by keywords can relatively narrow the range, after each search, there will still be many paragraphs that appear the keyword but are actually irrelevant. It still requires manual screening of the found paragraphs one by one, and it is difficult to achieve accurate searching. Moreover, no matter which of the above methods, it depends on the user's memory. If the user doesn't think of which positions might have the content that meets the preferences for a moment, or doesn't think of the appropriate keywords for a moment, or has a memory error and thinks of the wrong keywords, it is very difficult to quickly and accurately find the content that meets the preferences. Finally, even if these contents are found, the traditional copy-paste method is also relatively cumbersome to operate. Especially when the length is long, there are often many paragraphs between the original position and the current position of the desired paragraph, and sometimes it even requires flipping many pages to achieve copying and pasting from the original position to the current position. The above factors all greatly reduce the generation efficiency of the novel.

[0004] Therefore, there is a technical problem of low generation efficiency in the current method of generating novels by artificial intelligence, which needs to be improved. Summary of the Invention

[0005] Embodiments of this application provide a novel generation method, device, electronic device, and medium based on a text storage box to alleviate the technical problem of low generation efficiency existing in the current novel generation method.

[0006] To solve the above technical problems, the embodiments of this application provide the following technical solutions:

[0007] This application provides a novel generation method based on a text storage box. The method includes:

[0008] Display a first novel generation page, which includes a first text display area and a first text storage area. The first text display area displays multiple segments of historical texts, and the first text storage area displays at least one blank text storage box;

[0009] In response to N selection operations on N groups of different candidate historical texts in the first text display area and the drag operation of dragging each selected group of candidate historical texts to one of the blank text storage boxes in the first text storage area, the N groups of candidate historical texts are respectively stored in N candidate text storage boxes. Each group of candidate historical texts includes at least one segment of candidate historical text, and N is a positive integer;

[0010] Display a second novel generation page, which includes a second text display area and a second text storage area. The second text display area displays multiple segments of current texts, and the second text storage area displays the N candidate text storage boxes;

[0011] In response to the selection operation on the target text storage box in the second text storage area and the drag operation of dragging the target group of historical texts in the target text storage box to the target position in the second text display area, move the target group of historical texts to the target position;

[0012] Generate and display a novel containing the target group of historical texts in the second text display area.

[0013] Meanwhile, an embodiment of this application also provides a novel generation device based on a text storage box. The device includes:

[0014] A first display module for displaying a first novel generation page, which includes a first text display area and a first text storage area. The first text display area displays multiple segments of historical texts, and the first text storage area displays at least one blank text storage box;

[0015] A first response module for, in response to N selection operations on N groups of different candidate historical texts in the first text display area and the drag operation of dragging each selected group of candidate historical texts to one of the blank text storage boxes in the first text storage area, storing the N groups of candidate historical texts in N candidate text storage boxes respectively. Each group of candidate historical texts includes at least one segment of candidate historical text, and N is a positive integer;

[0016] The second display module is used to display a second novel generation page, which includes a second text display area and a second text storage area. Multiple current texts are displayed in the second text display area, and the N candidate text storage boxes are displayed in the second text storage area;

[0017] The second response module is used to respond to the selection operation of the target text storage box in the second text storage area and the drag operation of dragging the target group of historical texts in the target text storage box to the target position in the second text display area, and move the target group of historical texts to the target position;

[0018] The third display module is used to generate and display a novel containing the target group of historical texts in the second text display area.

[0019] This application also provides an electronic device, including a memory and a processor; the memory stores an application program, and the processor is used to run the application program in the memory to execute the steps in the novel generation method described in any one of the above.

[0020] An embodiment of this application provides a computer-readable storage medium, which stores multiple instructions suitable for a processor to load and execute the steps in the above novel generation method.

[0021] Beneficial effects: This application provides a novel generation method, device, electronic device and medium based on text storage boxes. The method first displays a first novel generation page, which includes a first text display area and a first text storage area. The user respectively selects N groups of candidate historical texts in the first text display area and drags them to the blank text storage boxes in the first text storage area to obtain N candidate text storage boxes. Then, a second novel generation page is displayed, which includes a second text display area and a second text storage area. The above N candidate text storage boxes are displayed in the second text storage area. The user selects the target text storage box in the second text storage area and drags the target group of historical texts in it to the target position in the second text display area, then the target group of historical texts can be moved to the target position. Finally, the novel generated and displayed in the second text display area will contain the target group of historical texts. By setting up a text storage area and text storage boxes, this application can first store the desired historical texts in the text storage boxes. If needed later, only need to search among the N text storage boxes, and the search range is small, so the target group of historical texts required can be found quickly and accurately. In addition, when storing and applying historical texts, only simple selection and drag operations need to be performed between different areas of the same page, without the need to switch back and forth between the location of the historical text and the current location, which is simple and easy to operate. Considering the above factors, the novel generation efficiency has been effectively improved. Brief Description of the Drawings

[0022] The following, in conjunction with the drawings, through a detailed description of the specific embodiments of the present application, will make the technical solutions and other beneficial effects of the present application obvious.

[0023] Figure 1 It is a schematic diagram of the scenario of the novel generation method based on a text storage box provided by an embodiment of the present application.

[0024] Figure 2 It is a schematic flow diagram of the novel generation method based on a text storage box provided by an embodiment of the present application.

[0025] Figure 3 It is the first schematic diagram of the first novel generation page in an embodiment of the present application.

[0026] Figure 4 It is the second schematic diagram of the first novel generation page in an embodiment of the present application.

[0027] Figure 5 It is the third schematic diagram of the first novel generation page in an embodiment of the present application.

[0028] Figure 6 It is the first schematic diagram of the second novel generation page in an embodiment of the present application.

[0029] Figure 7 It is the second schematic diagram of the second novel generation page in an embodiment of the present application.

[0030] Figure 8 It is the third schematic diagram of the second novel generation page in an embodiment of the present application.

[0031] Figure 9 It is the fourth schematic diagram of the second novel generation page in an embodiment of the present application.

[0032] Figure 10 It is the fifth schematic diagram of the second novel generation page in an embodiment of the present application.

[0033] Figure 11 It is the sixth schematic diagram of the second novel generation page in an embodiment of the present application.

[0034] Figure 12 It is a schematic structural diagram of the novel generation device based on a text storage box provided by an embodiment of the present application.

[0035] Figure 13 It is a schematic structural diagram of the electronic device provided by an embodiment of the present application.

[0036] Description of the reference numerals:

[0037] Generating device 11; large language model 12; first text display area 100; first text storage area 200; blank text storage box 30; blank label 31; candidate text storage box 40; first feature label 41; second text display area 300; second text storage area 400; label display area 500; storage box display area 600; first display module 121; first response module 122; second display module 123; second response module 124; third display module 125; radio frequency circuit 101; memory 102; input unit 103; display unit 104; sensor 105; audio circuit 106; WiFi module 107; processor 108. Detailed implementation manners

[0038] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of protection of the present application.

[0039] Please refer to Figure 1 , Figure 1 The scene schematic diagram of the novel generation method applied based on the text storage box provided in the embodiment of the present application. This scene includes a generating device 11 and a large language model 12. A novel generation application / website runs in the generating device 11. The user issues various instructions for generating a novel in the relevant interface of the novel generation application / website. The generating device 11 responds to each instruction, constructs various prompt words and calls the large language model 12 to process the prompt words, and sequentially returns the generated content after processing, and finally obtains the entire novel and displays it in the generating device 11.

[0040] During the novel generation process, the novel generation application / website first displays the first novel generation page, which includes a first text display area and a first text storage area. The large language model 12 generates a large amount of historical texts, which are displayed to the user through the first text display area. The user can select multiple groups of favorite historical texts from them and drag and drop them into the blank text storage boxes in the first text storage area. After N selection and drag-and-drop operations, N candidate text storage baskets can be obtained, and each candidate text storage box contains a selected group of candidate historical texts. When continuing to generate a novel, the novel generation application / website will display the second novel generation page, which includes a second text display area and a second text storage area. The current text generated by the large language model 12 is displayed to the user through the second text display area, and at the same time, the N candidate text storage boxes are displayed to the user through the second text storage area. The user can select a target text storage box required at present from the second text storage area and move the target group of historical texts it contains to the target position in the second text display area by dragging. After the movement, the target group of historical texts can be displayed in the second text display area and become part of the finally generated novel.

[0041] In the above way, the user can quickly and accurately find the required target group of historical texts, and only need to perform simple selection and drag-and-drop operations between different areas of the same page to generate a novel containing the target group of historical texts. The whole process is simple and easy to operate, effectively improving the novel generation efficiency. In the following embodiments, the above novel generation process will be described in detail.

[0042] Please refer to Figure 2 , Figure 2 which is a schematic flowchart of the novel generation method based on text storage boxes provided by the embodiments of the present application. The method specifically includes:

[0043] S1: Display the first novel generation page, which includes a first text display area and a first text storage area. The first text display area displays multiple paragraphs of historical texts, and the first text storage area displays at least one blank text storage box.

[0044] A large language model is a language model with a relatively large number of parameters, aiming to understand and generate human language. Trained with a large amount of text data, it can perform a wide range of tasks including text summarization, translation, analysis, generation, etc. In the process of generating novels with the help of large language models, the text of the novel is gradually generated. For the generated historical text, including the main text content and outline content, the page where it is located is the first novel generation page. The first novel generation page includes a first text display area and a first text storage area. Among them, the first text display area is used to display multiple segments of historical text, and the first text storage area initially stores at least one blank text storage box in its initial state. The blank text storage box only has a storage shape, but no valid content is filled. The inside of the blank text storage box can be completely blank, or "drag and store content" and other similar prompt messages can be set according to needs. This prompt message can disappear automatically after the user performs a drag operation. The number of blank text storage boxes can be set or created according to needs.

[0045] The first text storage area can have two states: collapsed and expanded. If the user does not need to use the text storage function currently, it can be collapsed, and only the first text display area is retained on the first novel generation page to facilitate the user to browse or edit the historical text, etc.; if the user needs to use the text storage function currently, then both the first text display area and the first text storage area can be retained on the first novel generation page, and the two areas can be displayed side by side. For example, the left side is the first text display area, and the right side is the first text storage area, to facilitate selection and drag operations.

[0046] Exemplarily, as Figure 3 shown, the first novel generation page includes a first text display area 100 and a first text storage area 200. For the convenience of representation, a dotted line is used to separate the two. The dotted line is only used to divide the area and does not represent the actual existence of the dotted line in the real scene. Historical text is displayed in the first text display area 100, which can specifically include outlines at all levels and main text content. The outlines at all levels can include partial outlines, chapter outlines, and paragraph outlines, and the main text content can include the main text of each chapter. Multiple blank text storage boxes 30 are displayed in the first text storage area 200. Similar prompt messages such as "drag and store content" can be displayed in the blank positions of each blank text storage box 30. In addition, each blank text storage box 30 is also provided with a blank label 31, and the part about the label will be further described in the subsequent embodiments.

[0047] S2: In response to N selection operations on N groups of different candidate historical texts in the first text display area and the drag - and - drop operation of dragging each selected group of candidate historical texts to one of the blank text storage boxes in the first text storage area, the N groups of candidate historical texts are respectively stored in N candidate text storage boxes. Each group of candidate historical texts includes at least one segment of candidate historical text, and N is a positive integer.

[0048] After the user selects a certain group of historical texts in the first text display area, the generating device, in response to this operation, takes this group of historical texts as a group of candidate historical texts. The term "candidate" indicates that this group of historical texts is available for the user to select in subsequent steps. Each group of candidate historical texts includes at least one segment of candidate historical text, which means that when the user selects these contents each time, the user can select only one segment of historical text, or can also select two or more segments of historical text at the same time. For the case of only selecting one segment of historical text, the user can select an entire segment of historical text, or can also select only a part of a certain segment of historical text. That is, when selecting, the user can flexibly select the content of historical texts according to their own needs.

[0049] After the selection, the user drags this group of candidate historical texts to one of the blank text storage boxes in the first text storage area. The generating device, in response to this operation, stores this group of candidate historical texts in the corresponding blank text storage box, and this blank text storage box serves as a candidate text storage box. After the drag - and - drop operation, this group of candidate historical texts is still displayed in the first text display area, that is, the drag - and - drop operation has no impact on it.

[0050] Exemplarily, as Figure 4 shown, the user performs a selection and drag - and - drop operation on a certain group of historical texts such as "Text Plot One" displayed in the first text display area 100 and drags it to one of the blank text storage boxes 30. As Figure 5 shown, after the drag - and - drop operation is completed, "Text Plot One" is stored in the above - mentioned blank text storage box 30, and a candidate text storage box 40 is obtained.

[0051] After the above - mentioned selection operation and drag - and - drop operation, a candidate text storage box is obtained in the first text storage area, and a group of candidate historical texts is stored in the box. When the user performs the above - mentioned selection operation and drag - and - drop operation on another group of historical texts again, another candidate text storage box can be obtained, and another group of candidate historical texts is stored in the box. By performing the selection operation and drag - and - drop operation N times respectively, N candidate text storage boxes can be obtained, where N is a positive integer, that is, there is at least one candidate text storage box.

[0052] In one embodiment, S2 specifically includes:

[0053] S21: Input each selected group of candidate historical texts into a large - language model to obtain at least one first feature label.

[0054] S22: Incorporate each group of candidate historical texts and their associated first feature tags into the corresponding candidate text storage box.

[0055] After each selection of a group of candidate historical texts, in addition to being able to incorporate them to obtain a candidate text storage box, feature extraction can also be performed on them to obtain first feature tags, and the group of candidate historical texts and the associated first feature tags are jointly incorporated and displayed in the candidate text storage box. During subsequent searches, the first feature tags displayed in each candidate text storage box can be viewed first. After selecting several desired first feature tags from all the first feature tags, the specific text content in the candidate text storage boxes corresponding to these tags can be viewed, and finally the target text storage box required can be determined. Compared with the method of directly viewing the text content, the method of viewing the tags first and then the text content can reduce the user's viewing workload, enabling the user to quickly locate the required content and further improving the search efficiency.

[0056] After each selection of a group of candidate historical texts, the background constructs corresponding prompt words based on the content of the group of candidate historical texts and inputs the prompt words to the large language model. The prompt words are used to prompt the large language model to perform feature extraction on the group of candidate historical texts, select or summarize several words that best match the features of the group of texts from all the text content and output them. The large language model processes the prompt words and outputs several words, and these words serve as the first feature tags. The number of first feature tags can be set as needed, such as 2, 3, or more.

[0057] Specifically, an example of the prompt word is as follows:

[0058] Please analyze the following novel passage, extract key information and make annotations, and output in JSON format.

[0059] Passage content: [Novel passage excerpted by the user]

[0060] JSON output format: { "All words": ["word1", "word2", "word3", "..." ], "Final selection": [ "Most characteristic word1", "Most characteristic word2", "Most characteristic word3"]}

[0061] Precautions:

[0062] 1. All words: Extract all nouns, verbs, adjectives, etc. that appear in the passage, including characters, locations, things, emotions, plots, etc. These words should be able to comprehensively summarize the content of the passage as much as possible.

[0063] 2. Proper Nouns: Among "all words", there must be at least one proper noun (e.g., person's name, place name, book title, organization name, etc.). If there is no obvious proper noun in the paragraph, try to dig out implicit proper nouns or proper nouns related to the paragraph theme.

[0064] 3. Final Selection: From "all words", select 3 words with the highest information content and most able to reflect the text characteristics as the final annotation result. At least one of these 3 words must be a proper noun. The most characteristic words are those that can best reflect the paragraph theme, emotion, style or meaning. The selected words should be representative and generalizable, and can help readers quickly grasp the core content of the paragraph.

[0065] 4. JSON Format: Please ensure that the output JSON format is correct and avoid syntax errors. "All words" and "Final Selection" should both be placed in an array.

[0066] Please be sure to carefully read the above precautions to ensure that the annotation results you provide are accurate and complete.

[0067] Of course, the above is only one example of the prompt content. This application is not limited to this. Those skilled in the art can reasonably set the prompt content according to needs, as long as it can enable the large language model to understand and output the corresponding first feature label.

[0068] Specifically, assume that the content of the novel paragraph (a certain set of candidate historical texts) excerpted by the user is as follows:

[0069] The sun is setting in the west, and the heartbroken man is at the end of the world.” He rides a thin and bony horse and slowly walks on the desolate ancient road. The cold wind blows, rolling up the yellow sand on the ground and blinding his eyes. He feels a rush of loneliness and despair welling up in his heart.

[0070] After being processed by the large language model based on the above prompt content, the "all words" obtained include: "sunset", "heartbroken man", "end of the world", "thin horse", "ancient road", "cold wind", "yellow sand", "loneliness", "despair", "traveler", "desolate", and the "Final Selection" includes: "sunset", "loneliness", "heartbroken man". Then the obtained first feature labels include: "sunset", "loneliness", "heartbroken man".

[0071] In the existing solution for searching by keywords, the user has to come up with the keywords actively by themselves. If they can't think of the appropriate keywords for a moment, or if they think of the wrong keywords due to memory errors, it is very difficult to quickly and accurately find the desired historical text. Therefore, the implementation of this solution is difficult and the search efficiency is low. In the embodiments of the present application, the first feature label is the most representative word automatically generated by the large language model after feature recognition of each group of candidate historical texts. Searching with this as a reference can achieve precise positioning, and searching within the existing label range will also reduce the search difficulty and effectively improve the search efficiency.

[0072] In one embodiment, the first feature label can also be modified or added through a custom method. By combining the two methods of customization and model generation, the first feature label can more accurately reflect the characteristics of each group of historical texts, providing an effective reference for subsequent searches.

[0073] Exemplarily, as Figure 4 and Figure 5 shown, the specific content of the generated first feature label 41 is represented by "label 1" and "label 2", and the specific content of these first feature labels 41 will be filled in the blank label 31. The user can also modify the content filled in the blank label 31, or add a new blank label 31 and fill in the content by themselves to obtain the final first feature label 41.

[0074] S3: Display the second novel generation page. The second novel generation page includes a second text display area and a second text storage area. The second text display area displays multiple segments of current text, and the second text storage area displays N candidate text storage boxes.

[0075] For the currently generated text, including the outline content and the body content, its page is the second novel generation page. The second novel generation page includes a second text display area and a second text storage area. Among them, the second text display area is used to display multiple segments of current text, and the second text storage area is used to display the N candidate text storage boxes obtained in the above steps.

[0076] Similarly, the second text storage area can have two states: collapsed and expanded. If the user doesn't need to use the candidate historical text currently, it can be collapsed, and only the second text display area is retained on the second novel generation page to facilitate the user to browse or edit the current text, etc.; if the user needs to use the candidate historical text currently, the second text display area and the second text storage area can be retained on the second novel generation page at the same time, and the two areas can be displayed side by side. For example, the second text display area is on the left and the second text storage area is on the right, so as to facilitate subsequent selection and dragging operations.

[0077] Exemplarily, as Figure 6As shown, the second novel generation page includes a second text display area 300 and a second text storage area 400. The current text is displayed in the second text display area 300, which may specifically include outlines at all levels and the main text content. Multiple candidate text storage boxes 40 are displayed in the second text storage area 400, and each candidate text storage box 40 stores a group of candidate historical texts selected and dragged in the above steps, such as "Main text plot one", "Main text plot twelve", "Main text plot fifteen", "Main text plot twenty", etc. At the same time, each candidate text storage box 40 displays a corresponding first feature label 41, Figure 6 which are respectively represented by "Label 1" to "Label 8" in it.

[0078] S4: In response to the selection operation of the target text storage box in the second text storage area and the drag operation of dragging the target group of historical texts in the target text storage box to the target position in the second text display area, move the target group of historical texts to the target position.

[0079] After the user selects a certain candidate text storage box in the second text storage area, the generation device responds to this operation, takes this candidate text storage box as the target text storage box, and a group of historical texts stored in the target text storage box as the target group of historical texts. The user drags the target group of historical texts to the target position in the second text display area, and the generation device responds to this operation and moves the target group of historical texts to the target position.

[0080] In one embodiment, S4 specifically includes:

[0081] S41: If there is already generated text at the target position, replace the already generated text with the target group of historical texts.

[0082] S42: If there is no already generated text at the target position, insert the target group of historical texts into the target position.

[0083] The target position can be any position specified by the user. According to the specific situation of the target position, different processing can be performed on the target group of historical texts. If there is already generated text at the target position, that is, the last position of the drag operation is directly above a certain paragraph of the current text. At this time, the already generated text of this paragraph can be replaced with the target group of historical texts. After replacement, only the target group of historical texts remains at the target position, and the already generated text of this paragraph is overwritten. If there is no already generated text at the target position, that is, the last position of the drag operation is at a blank position above or below a certain paragraph of the current text. At this time, the target group of historical texts can be inserted into this blank position. After insertion, the target group of historical texts remains at the target position, and the already generated text of this paragraph still exists.

[0084] In the above - mentioned manner, users can very conveniently complete the insertion or replacement of the historical text of the target group. Of course, this application is not limited to this. For insertion or replacement, users can also set it by themselves. For example, after the dragging action ends, a pop - up window can be shown to the user. The pop - up window includes options for insertion and replacement. After the user makes a selection for one of the options, regardless of whether there is already generated text at the target position, a result corresponding to the option will be generated. Or before dragging, set the insertion or replacement first. After the user performs the dragging operation, regardless of whether there is already generated text at the target position, a result corresponding to the setting will be generated. Those skilled in the art can reasonably set the insertion and replacement methods according to needs, as long as the corresponding effects can be achieved simply and quickly.

[0085] In one embodiment, after S4, it further includes:

[0086] S4a: Remove the target text storage box from the second text storage area.

[0087] For the same group of historical texts, if it has been dragged once, it is very likely that the user will not need to use it again at a new position. If the group of historical texts still remains in the second text storage area after dragging, it will occupy the space of the second text storage area, increasing the workload when searching for the required historical texts later. In addition, if the user forgets that the group of historical texts has been dragged, it is possible that the group of historical texts will be selected by the user again and appear in the novel again. And excessive occurrences of the same or similar content will affect the quality of the novel. Therefore, after the target group of historical texts is moved to the target position, the target text storage box can be removed from the second text storage area to free up space in the second text storage area and also avoid the decline in the quality of the novel caused by the user's mis - selection.

[0088] Exemplarily, as Figure 7 and Figure 9 shown, the user performs a selection and dragging operation on a certain group of candidate historical main texts such as "Main Plot Twelve" shown in the second text storage area 400, and drags it to the target position in the second text display area 300. As Figure 8 and Figure 10 shown, after the dragging ends, "Main Plot Twelve" is moved to the target position, and at the same time, the corresponding target text storage box in the second text storage area 400 is removed. Among them, Figure 7 if there is already generated text "Main Plot Thirty - Two" at the target position, then after the movement, "Main Plot Twelve" will replace "Main Plot Thirty - Two". Figure 9 if there is no already generated text at the target position, then after the movement, "Main Plot Twelve" will be inserted between "Main Plot Thirty - One" and "Main Plot Thirty - Two".

[0089] In one embodiment, the second text storage area includes a label display area and a storage box display area. After S22, it further includes:

[0090] S23: Display all candidate text storage boxes in the storage box display area, and display all first feature labels in the label display area.

[0091] Before S4, it further includes:

[0092] S4a: In response to a selection operation on a target first feature label in the label display area, preferentially display the candidate text storage box associated with the target first feature label in the storage box display area.

[0093] Each candidate text storage box is displayed in the second text storage area. In addition to storing a group of candidate historical texts, each candidate text storage box also stores the first feature labels generated in the above steps. If only the candidate text storage boxes are displayed in the second text storage area, since the candidate text storage boxes have a certain size and the content displayed on the page of the generating device is limited, when there are many candidate text storage boxes, usually only a part of the candidate text storage boxes are displayed, and the other part of the candidate text storage boxes are not displayed first and will be displayed only when the user performs a page-turning operation. For this situation, when the user wants to view the first feature labels, not only does the user need to view several candidate text storage boxes displayed on the current page, but also needs to continuously operate to turn out the hidden candidate text storage boxes, which will result in low viewing efficiency. In addition, sometimes the user may need to view the text content corresponding to several specific labels, and the way of manually selecting labels one by one will also make it difficult to view.

[0094] Therefore, in this embodiment, the second text storage area is divided into a label display area and a storage box display area. The storage box display area is used to display all candidate text storage boxes, and these candidate text storage boxes have two forms: direct display and display after page turning. The label display area is used to display all first feature labels without displaying the specific text content. Similarly, the first feature labels also have two forms: direct display and display after page turning. The label display area can be expanded or collapsed based on the user's selection. When expanded, the label display area and the storage box display area can be arranged side by side. For example, the label display area can be set above the storage box display area.

[0095] After the user performs a selection operation on one or some of the first feature tags in the tag display area, the generation device responds to this operation, takes these first feature tags as target first feature tags, and then determines this part of the candidate text storage boxes associated with the target first feature tags from all candidate text storage boxes, and preferentially displays these candidate text storage boxes in the storage box display area. There are various ways of preferential display. For example, only these candidate text storage boxes are displayed, and the others are hidden; or these candidate text storage boxes are displayed at the top; or these candidate text storage boxes are distinguished from other parts by color, etc. Those skilled in the art can set the way of preferential display according to needs, so that users can intuitively see these candidate text storage boxes and the specific text content they store.

[0096] Since the number of characters of the first feature tags is small, through the above method, users do not need to rummage through all candidate text storage boxes to select their favorite first feature tags, but only need to select in the tag display area. Subsequently, they only need to view this part of the candidate text storage boxes preferentially displayed, so the viewing difficulty is reduced and the work efficiency is improved.

[0097] Exemplarily, as Figure 11 shown, the second text storage area 400 includes a tag display area 500 and a storage box display area 600. The storage box display area 600 is used to display all candidate text storage boxes 40, and the tag display area 500 is used to display all first feature tags 41. When there are many display objects, both can be paged to display more objects. After the user selects one or several first feature tags 41 in the tag display area 500, the storage box display area 600 will preferentially display the corresponding one or several candidate text storage boxes 40. In addition, a search box can also be set in the storage box display area 600. Users can either find the required candidate text storage boxes 40 by selecting the first feature tags 41 or enter relevant information in the search box to find them. The combination of the two further improves the efficiency.

[0098] In one embodiment, before S4, it further includes:

[0099] S4b: In response to the selection operation on the target current text in the second text display area, input the target current text into the large language model to obtain at least one second feature tag.

[0100] S4c: Match the second feature tags with all the first feature tags associated with the N candidate text storage boxes.

[0101] S4d: Preferentially display the candidate text storage boxes associated with the first feature tags that match successfully in the second text storage area.

[0102] In the above embodiments, the user first manually selects the target text storage box in the second text storage area and then drags it to the second text display area. For the manual selection method, the efficiency still needs to be improved. In this embodiment, after the second novel generation page is displayed, the user first selects one or several segments of the current text in the second text display area. These current texts are usually the texts that need to be replaced later, or these current texts are near the positions where they need to be inserted later. In response to this operation, the generating device uses these current texts as the target current texts, and the background constructs corresponding prompt words according to the content of the target current texts and inputs the prompt words into the large language model. The prompt words are used to prompt the large language model to extract features from the target current texts, select or refine several words that best match the target current texts from all the text contents as the second feature tags. Different from the above embodiments, the second feature tags do not need to be displayed.

[0103] After obtaining the second feature tags, the background matches the second feature tags with all the first feature tags associated with the N candidate text storage boxes. If a certain first feature tag is the same as any one of the second feature tags, it means that the first feature tag matches the second feature tag successfully; otherwise, it means the match is unsuccessful. The number of second feature tags and the number of first feature tags can be equal or not equal. Those skilled in the art can reasonably set the numbers of both according to needs to make the matching effect better.

[0104] For the case of successful matching, determine some of the candidate text storage boxes associated with this first feature tag from all the candidate text storage boxes and preferentially display them in the second text storage area. Similarly, there are various ways of preferential display. For example, only display these candidate text storage boxes and hide the others; or display these candidate text storage boxes at the top; or distinguish these candidate text storage boxes from the others with colors, etc. Those skilled in the art can set the way of preferential display according to needs so that users can intuitively see these candidate text storage boxes and the specific text contents they store.

[0105] For example, if the second feature tags obtained from the target current text selected by the user include "character a", and among the N candidate text storage boxes, the first feature tags of 3 candidate text storage boxes all include "character a", then these 3 candidate text storage boxes will all be preferentially displayed to the user, and then the user can select the most suitable target text storage box from them.

[0106] In this application, a large language model is used to generate a first feature tag and a second feature tag respectively. By matching the two types of tags, one or several candidate text storage boxes with the highest content relevance to the target current text can be accurately found from N candidate text storage boxes. This part of the text content conforms to the creation logic better than other content and is more suitable for inserting or replacing the target current text. Since the search range is reduced from N to one or several, the viewing difficulty is reduced and the work efficiency is improved.

[0107] In one embodiment, S2 specifically further includes:

[0108] S24: Obtain the original position information of each group of candidate historical texts.

[0109] S25: Collect each group of candidate historical texts and the associated original position information into the corresponding candidate text storage boxes.

[0110] After S3, it further includes:

[0111] S3a: In response to the selection operation of the target original position information associated with the target group of historical texts, jump to display the target original position where the target group of historical texts is located in the second text display area.

[0112] When the user selects and drags a certain group of candidate historical texts, it is to directly use this group of candidate historical texts. However, in some cases, the user may also need to know where this group of candidate historical texts is excerpted from and what specific content is written in the context of the excerpt position, so as to select the most suitable target group of candidate historical texts or better edit this group of candidate historical texts after moving. For this situation, when the user performs the selection and drag operation on a certain group of historical texts, the background can obtain the original position information of this group of candidate historical texts. The original position information is used to record the position of this group of candidate historical texts in the story when it was initially generated, such as paragraph 5, chapter 2. After obtaining it, each group of candidate historical texts and the associated original position information are collected into the corresponding candidate text storage boxes. After being displayed to the user on the second story generation page, in addition to performing the selection and drag operation in S4, the user can also perform a selection operation on the target original position information associated with the target group of historical texts. The generating device can respond to this operation and jump the second text display area from the current position to the target original position where the target group of historical texts is located, such as jumping from paragraph 20, chapter 2 to paragraph 5, chapter 2, so that the user can make a decision after understanding the context information to improve the generation quality of the novel.

[0113] S5: Generate and display a novel containing the target group of historical texts in the second text display area.

[0114] After moving the target group of historical texts to the target position, multiple segments of current texts and the target group of historical texts will be displayed in the second text display area. Users can directly perform operations such as acceptance or saving on these contents, or perform acceptance or saving operations after modification, so that these contents jointly serve as the latest generated content of the novel.

[0115] As can be seen from the above embodiments, for the novel generation method provided by the present application, by setting up a text storage area and text storage boxes, favorite historical texts can be first stored in the text storage boxes. Subsequently, if they are needed, only need to search in N text storage boxes, and the search range is relatively small. Therefore, the target group of historical texts required can be found quickly and accurately. In addition, when storing and applying historical texts, only simple selection and dragging operations need to be performed between different areas on the same page, without the need to switch back and forth between the location of the historical text and the current location, which is simple and easy to operate. Considering the above factors, the novel generation efficiency has been effectively improved.

[0116] Based on the method described in the above embodiments, this embodiment will further describe from the perspective of a novel generation device based on text storage boxes. Please refer to Figure 12 , the novel generation device may include:

[0117] A first display module 121, configured to display a first novel generation page, the first novel generation page includes a first text display area and a first text storage area, the first text display area displays multiple segments of historical texts, and the first text storage area displays at least one blank text storage box;

[0118] A first response module 122, configured to respond to N selection operations on N groups of different candidate historical texts in the first text display area, and a dragging operation of dragging each selected group of candidate historical texts to one of the blank text storage boxes in the first text storage area, and store the N groups of candidate historical texts in N candidate text storage boxes respectively, each group of candidate historical texts includes at least one segment of candidate historical text, and N is a positive integer;

[0119] A second display module 123, configured to display a second novel generation page, the second novel generation page includes a second text display area and a second text storage area, the second text display area displays multiple segments of current texts, and the second text storage area displays the N candidate text storage boxes;

[0120] A second response module 124, configured to respond to a selection operation on a target text storage box in the second text storage area, and a dragging operation of dragging the target group of historical texts in the target text storage box to a target position in the second text display area, and move the target group of historical texts to the target position;

[0121] The third display module 125 is configured to generate and display a novel containing the historical text of the target group in the second text display area.

[0122] In one embodiment, the first response module 122 includes:

[0123] An input unit, configured to input a selected group of candidate historical texts into a large language model each time to obtain at least one first feature label;

[0124] A first storage unit, configured to jointly store each group of candidate historical texts and the associated first feature labels into corresponding candidate text storage boxes.

[0125] In one embodiment, the second text storage area includes a label display area and a storage box display area. The novel generation device further includes a fourth display module that works after the display unit, and a third response module that works before the second response module 124;

[0126] The fourth display module is configured to display all candidate text storage boxes in the storage box display area and display all first feature labels in the label display area;

[0127] The third response module is configured to, in response to a selection operation on a target first feature label, preferentially display the candidate text storage box associated with the target first feature label in the second text storage area.

[0128] In one embodiment, the novel generation device further includes an input module, a matching module, and a fifth display module that work before the second response module 124;

[0129] The input module is configured to, in response to a selection operation on a target current text in the second text display area, input the target current text into a large language model to obtain at least one second feature label;

[0130] The matching module is configured to match the second feature label with all first feature labels associated with N candidate text storage boxes;

[0131] The fifth display module is configured to preferentially display the candidate text storage boxes associated with the successfully matched first feature labels in the second text storage area.

[0132] In one embodiment, the second response module 124 includes:

[0133] A replacement unit, configured to replace the generated text with the historical text of the target group if there is generated text at the target position;

[0134] An insertion unit, configured to insert the historical text of the target group into the target position if there is no generated text at the target position.

[0135] In one embodiment, the novel generation device further includes a removal module that operates after the second response module 124. The removal module is configured to remove the target text storage box from the second text storage area.

[0136] In one embodiment, the first response module 122 includes:

[0137] An acquisition unit, configured to acquire the original position information of each group of candidate historical texts;

[0138] A second storage unit, configured to jointly store each group of candidate historical texts and the associated original position information in the corresponding candidate text storage box;

[0139] The novel generation device further includes a fourth response module that operates after the second display module 123. The fourth response module is configured to, in response to a selection operation on the target original position information associated with the target group of historical texts, jump to and display the target original position where the target group of historical texts is located in the second text display area.

[0140] Different from the prior art, the novel generation device based on text storage boxes provided by the present application can, by setting a text storage area and text storage boxes, first store the desired historical texts in the text storage boxes. Subsequently, if needed, only a search needs to be performed among the N text storage boxes, and the search range is small. Therefore, the target group of historical texts required can be found quickly and accurately. In addition, when storing and applying historical texts, only simple selection and dragging operations need to be performed between different areas on the same page, without the need to switch back and forth between the location of the historical text and the current location, which is simple and easy to operate. Considering the above factors, the novel generation efficiency is effectively improved.

[0141] Correspondingly, an embodiment of the present application further provides an electronic device, as Figure 13 shown. The electronic device may include a radio frequency (RF) circuit 101, a memory 102 including one or more computer-readable storage media, an input unit 103, a display unit 104, a sensor 105, an audio circuit 106, a WiFi module 107, a processor 108 including one or more processing cores, and a power supply 109, etc. Those skilled in the art can understand that Figure 13 the structure of the electronic device shown in

[0142] The radio frequency circuit 101 can be used for receiving and transmitting information during information reception and call processes. Specifically, after receiving the downlink information from the base station, it is handed over to one or more processors 108 for processing. Additionally, data related to the uplink is sent to the base station. The memory 102 can be used to store software programs and modules. The processor 108 executes various functional applications and novel generation by running the software programs and modules stored in the memory 102. The input unit 103 can be used to receive input digital or character information, and generate keyboard, mouse, joystick, optical, or trackball signal inputs related to customer settings and function control.

[0143] The display unit 104 can be used to display information input by the customer or information provided to the customer, as well as various graphical user interfaces of the server. These graphical user interfaces can be composed of graphics, text, icons, videos, and any combination thereof.

[0144] The electronic device may further include at least one sensor 105, such as a light sensor, a motion sensor, and other sensors. The audio circuit 106 includes a speaker, and the speaker can provide an audio interface between the customer and the electronic device.

[0145] WiFi belongs to wireless transmission technology. The electronic device can help the customer send and receive emails, browse the web, and follow up on streaming media through the WiFi module 107. It provides the customer with wireless broadband Internet access. Although Figure 13 the WiFi module 107 is shown, it can be understood that it does not belong to an essential component of the electronic device and can be omitted entirely within the scope of not changing the essence of the application as needed.

[0146] The processor 108 is the control center of the electronic device. It connects various parts of the entire mobile phone using various interfaces and lines. By running or executing the software programs and / or modules stored in the memory 102, and by calling the data stored in the memory 102, it executes various functions of the electronic device and processes data, thereby providing overall monitoring of the mobile phone.

[0147] The electronic device further includes a power source 109 (such as a battery) for supplying power to each component. Preferably, the power source can be logically connected to the processor 108 through a power management system, thereby implementing functions such as management of charging, discharging, and power consumption management through the power management system.

[0148] Although not shown, the electronic device may further include a camera, a Bluetooth module, etc., which will not be elaborated here. Specifically, in this embodiment, the processor 108 in the server will, according to the following instructions, load the executable files corresponding to the processes of one or more application programs into the memory 102, and the processor 108 will run the application programs stored in the memory 102 to achieve the following functions:

[0149] Displays a first novel generation page, where the first novel generation page includes a first text display area and a first text storage area. The first text display area displays multiple segments of historical texts, and the first text storage area displays at least one blank text storage box;

[0150] In response to N selection operations on N groups of different candidate historical texts in the first text display area and drag operations of dragging each selected group of candidate historical texts to one of the blank text storage boxes in the first text storage area, the N groups of candidate historical texts are respectively stored in N candidate text storage boxes. Each group of candidate historical texts includes at least one segment of candidate historical text, and N is a positive integer;

[0151] Displays a second novel generation page, where the second novel generation page includes a second text display area and a second text storage area. The second text display area displays multiple segments of current texts, and the second text storage area displays the N candidate text storage boxes;

[0152] In response to a selection operation on a target text storage box in the second text storage area and a drag operation of dragging the target group of historical texts in the target text storage box to a target position in the second text display area, the target group of historical texts is moved to the target position;

[0153] Generates and displays a novel containing the target group of historical texts in the second text display area.

[0154] The electronic device provided by this application, by setting a text storage area and text storage boxes, can first store the desired historical texts in the text storage boxes. If they need to be used later, only need to search among the N text storage boxes, and the search range is relatively small. Therefore, the target group of historical texts required can be found quickly and accurately. In addition, when storing and applying historical texts, only simple selection and drag operations need to be performed between different areas of the same page, without the need to switch back and forth between the location of the historical text and the current location, which is simple and easy to operate. Considering the above factors, the novel generation efficiency has been effectively improved.

[0155] In the above embodiments, the descriptions of the embodiments each have their own focuses. For parts not detailed in a certain embodiment, reference can be made to the detailed descriptions above, and will not be elaborated here.

[0156] Those of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructions, or by controlling relevant hardware through instructions. The instructions can be stored in a computer-readable storage medium and loaded and executed by a processor.

[0157] To this end, an embodiment of the present application provides a computer-readable storage medium storing multiple instructions that can be loaded by a processor to implement the following functions:

[0158] Display a first novel generation page, where the first novel generation page includes a first text display area and a first text storage area. Multiple historical texts are displayed in the first text display area, and at least one blank text storage box is displayed in the first text storage area;

[0159] In response to N selection operations on N groups of different candidate historical texts in the first text display area and drag operations of dragging each selected group of candidate historical texts to one of the blank text storage boxes in the first text storage area, the N groups of candidate historical texts are respectively stored in N candidate text storage boxes. Each group of candidate historical texts includes at least one candidate historical text, and N is a positive integer;

[0160] Display a second novel generation page, where the second novel generation page includes a second text display area and a second text storage area. Multiple current texts are displayed in the second text display area, and the N candidate text storage boxes are displayed in the second text storage area;

[0161] In response to a selection operation on a target text storage box in the second text storage area and a drag operation of dragging the target group of historical texts in the target text storage box to a target position in the second text display area, move the target group of historical texts to the target position;

[0162] Generate and display a novel including the target group of historical texts in the second text display area.

[0163] By setting up a text storage area and text storage boxes, the computer-readable storage medium provided by the present application can first store the desired historical texts in the text storage boxes. If needed later, one only needs to search among the N text storage boxes, and the search range is small. Therefore, the target group of historical texts required can be found quickly and accurately. In addition, when storing and applying historical texts, one only needs to perform simple selection and drag operations between different areas on the same page, without having to switch back and forth between the location of the historical text and the current location. It is simple and easy to operate. Considering the above factors, the novel generation efficiency is effectively improved.

[0164] The above has introduced in detail a novel generation method, device, electronic device, and computer-readable storage medium based on a text storage box provided by the embodiments of the present application. Specific examples are used in this article to elaborate on the principle and implementation manner of the present application. The description of the above embodiments is only used to help understand the technical solution and its core idea of the present application; those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A novel generation method based on a text storage box, characterized in that, The method includes: Displaying a first novel generation page, the first novel generation page including a first text display area and a first text storage area, the first text display area displaying multiple historical texts, and the first text storage area displaying at least one blank text storage box; In response to N selection operations on N groups of different candidate historical texts in the first text display area and a drag operation of dragging each selected group of candidate historical texts to one of the blank text storage boxes in the first text storage area, storing the N groups of candidate historical texts in N candidate text storage boxes respectively, each group of candidate historical texts including at least one candidate historical text, and N being a positive integer; Displaying a second novel generation page, the second novel generation page including a second text display area and a second text storage area, the second text display area displaying multiple current texts, and the second text storage area displaying the N candidate text storage boxes; In response to a selection operation on a target text storage box in the second text storage area and a drag operation of dragging the target group of historical texts in the target text storage box to a target position in the second text display area, moving the target group of historical texts to the target position; Generating and displaying a novel including the target group of historical texts in the second text display area.

2. The novel generation method according to claim 1, characterized in that, The step of storing the N groups of candidate historical texts in N candidate text storage boxes respectively includes: Inputting each selected group of candidate historical texts into a large language model to obtain at least one first feature label; Storing each group of candidate historical texts and the associated first feature label together in the corresponding candidate text storage box.

3. The novel generation method according to claim 2, wherein, The second text storage area includes a label display area and a storage box display area. After the step of storing each group of candidate historical texts and the corresponding first feature label together in the corresponding candidate text storage box, it further includes: Displaying all the candidate text storage boxes in the storage box display area and displaying all the first feature labels in the label display area; Before the step of responding to a selection operation on a target text storage box in the second text storage area, it further includes: In response to a selection operation on a target first feature label in the label display area, preferentially displaying the candidate text storage box associated with the target first feature label in the storage box display area.

4. The novel generation method according to claim 2, wherein Before the step of responding to a selection operation on a target text storage box in the second text storage area, it further includes: In response to a selection operation on a target current text in the second text display area, inputting the target current text into a large language model to obtain at least one second feature label; Matching the second feature label with all the first feature labels associated with the N candidate text storage boxes; Preferentially displaying the candidate text storage box associated with the first feature label with a successful match in the second text storage area.

5. The novel generation method according to claim 1, wherein The step of moving the target group of historical texts to the target position includes: If there is already generated text at the target position, replacing the already generated text with the target group of historical texts; If there is no already generated text at the target position, inserting the target group of historical texts at the target position.

6. The novel generation method according to claim 1, characterized in that, After the step of moving the target group of historical texts to the target position, the following steps are further included: Remove the target text storage box from the second text storage area.

7. The novel generation method according to claim 1, wherein The step of storing N groups of candidate historical texts in N candidate text storage boxes respectively includes: Obtain the original position information of each group of candidate historical texts; Store each group of candidate historical texts and the associated original position information together in the corresponding candidate text storage box; After the step of displaying the second novel generation page, the following steps are further included: In response to a selection operation on the target original position information associated with the target group of historical texts, jump to and display the target original position where the target group of historical texts is located in the second text display area.

8. A novel generation device based on a text storage box, characterized in that, The device includes: A first display module, configured to display a first novel generation page, the first novel generation page including a first text display area and a first text storage area, the first text display area displaying multiple paragraphs of historical texts, and the first text storage area displaying at least one blank text storage box; A first response module, configured to, in response to N selection operations on N groups of different candidate historical texts in the first text display area and a drag operation of dragging each selected group of candidate historical texts to one of the blank text storage boxes in the first text storage area, store the N groups of candidate historical texts in N candidate text storage boxes respectively, each group of candidate historical texts including at least one paragraph of candidate historical text, and N being a positive integer; A second display module, configured to display a second novel generation page, the second novel generation page including a second text display area and a second text storage area, the second text display area displaying multiple paragraphs of current texts, and the second text storage area displaying the N candidate text storage boxes; A second response module, configured to, in response to a selection operation on the target text storage box in the second text storage area and a drag operation of dragging the target group of historical texts in the target text storage box to the target position in the second text display area, move the target group of historical texts to the target position; A third display module, configured to generate and display a novel including the target group of historical texts in the second text display area.

9. An electronic device, characterized in that, It includes a memory and a processor; the memory stores an application program, and the processor is configured to run the application program in the memory to execute the steps in the novel generation method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, A computer program is stored on the computer-readable storage medium, and the computer program is executed by the processor to implement the steps in the novel generation method according to any one of claims 1 to 7.

Images