Video generation method and device, computing equipment, storage medium and program product

Abstract

Embodiments of the invention disclose a video generation method and apparatus. The method comprises the steps of obtaining a target story text; the target story text is subjected to content splitting processing, scene information, role information and subtitle information are obtained, and an incidence relation exists between the role information and the subtitle information; according to the role information, determining a role sticker corresponding to the subtitle information having an association relationship with the role information, and determining a scene material corresponding to the scene information; and generating a story video according to the subtitle information, the role sticker and the scene material. According to the technical scheme of the embodiment of the invention, the threshold of video production is reduced, the AI enabling creation is utilized, the user can be assisted to generate the video with the recarving modeling factor type playing method, the video uses some specific hot stem materials, the emotion feeling of the user in a certain scene is expressed in a picture mode, the video conversion of text stories is realized, and the user experience is improved. The user is helped to generate the video more quickly and more easily, the video production efficiency is improved, and the user experience is improved.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of Internet technology, and specifically to a video generation method, apparatus, computing device, storage medium, and program product. Background Art

[0002] With the rapid development of mobile Internet, users are increasingly fond of using short videos to express their emotions and opinions. However, due to the high threshold for short video production, many ordinary users are unable to quickly create short video works that meet their needs. Summary of the invention

[0003] In view of the above problems, the present application is proposed to provide a video generation method, apparatus, computing device, storage medium and program product that overcome the above problems or at least partially solve the above problems.

[0004] According to one aspect of an embodiment of the present application, a video generation method is provided, including:

[0005] Get the target story text;

[0006] The target story text is split into content to obtain scene information, character information and subtitle information, wherein there is a correlation between the character information and the subtitle information;

[0007] Determine, according to the character information, the character sticker corresponding to the subtitle information that is associated with the character information, and determine the scene material corresponding to the scene information;

[0008] Generate story videos based on subtitle information, character stickers, and scene materials.

[0009] Furthermore, character information includes emotional information;

[0010] Determining, according to the character information, the character sticker corresponding to the subtitle information associated with the character information further includes:

[0011] Query the emotional role library according to the emotional information to obtain the role stickers that match the emotional information;

[0012] The retrieved character stickers are determined as character stickers corresponding to the subtitle information that is associated with the emotion information.

[0013] Furthermore, the character information includes: action information;

[0014] Determining, according to the character information, the character sticker corresponding to the subtitle information associated with the character information further includes:

[0015] Query the action role library according to the action information to obtain the role stickers matching the action information;

[0016] The retrieved character sticker is determined as the character sticker corresponding to the subtitle information that is associated with the action information.

[0017] Furthermore, obtaining the target story text further includes:

[0018] Identify pending story proposals;

[0019] Using the AI ​​model, the story proposal to be processed is processed into a target story text in a preset format according to the preset first text prompt information.

[0020] Furthermore, using the AI ​​model to process the to-be-processed story proposal into a target story text in a preset format according to the preset first text prompt information further includes:

[0021] Using AI models, identify keywords in the story proposals to be processed, obtain meme elements that match the keywords, and generate meme story texts based on the meme elements and the story proposals to be processed;

[0022] The meme story text is rewritten into a target story text in a preset format according to the preset first text prompt information.

[0023] Furthermore, the method also includes: identifying sensitive words in the meme story text;

[0024] If there are sensitive words in the meme story text, a reminder message will be displayed in a preset manner to remind the user to make modifications.

[0025] Furthermore, generating a story video according to the subtitle information, character stickers, and scene materials further includes:

[0026] Lay subtitle information, character stickers, and scene materials on the timeline and render them to generate a story video.

[0027] Furthermore, the timeline includes: video track, audio track, subtitle track, sticker track; character stickers include: sticker images and sticker sound effects;

[0028] Laying subtitle information, character stickers, and scene materials to the timeline further includes:

[0029] Create video track, audio track, subtitle track, and sticker track in the video generation page;

[0030] Add subtitle information to the subtitle track, add sticker sound effects to the audio track, add sticker images to the sticker track, and add scene materials to the video track;

[0031] Record the mapping relationship between character stickers and subtitle information, set the display duration of subtitle information, sticker image display duration and sticker sound effect duration to be the same, set the start display time of sticker image and the start playback time of sticker sound effect according to the subtitle information display timestamp, and set the coordinate information corresponding to the subtitle information and sticker image.

[0032] Further, the method also includes: receiving a sticker batch editing request triggered by a user, wherein the sticker batch editing request carries an identifier of a character sticker to be edited;

[0033] In response to a sticker batch editing request, subtitle information and character stickers associated with the character sticker identifier to be edited are displayed on the sticker editing page to enable the user to perform batch editing.

[0034] According to another aspect of an embodiment of the present application, there is provided a video generating device, including:

[0035] An acquisition module, adapted to acquire a target story text;

[0036] A content splitting and processing module is adapted to perform content splitting and processing on the target story text to obtain scene information, role information and subtitle information, wherein there is a correlation relationship between the role information and the subtitle information;

[0037] A determination module, adapted to determine, based on the character information, a character sticker corresponding to the subtitle information associated with the character information, and to determine a scene material corresponding to the scene information;

[0038] The generation module is suitable for generating story videos based on subtitle information, character stickers, and scene materials.

[0039] According to another aspect of the embodiment of the present application, there is provided a computing device, including: a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus;

[0040] The memory is used to store at least one executable instruction, and the executable instruction enables the processor to execute operations corresponding to the above-mentioned video generation method.

[0041] According to another aspect of the embodiments of the present application, a computer storage medium is provided, in which at least one executable instruction is stored, and the executable instruction enables a processor to perform operations corresponding to the above-mentioned video generation method.

[0042] According to another aspect of the embodiments of the present application, a computer program product is provided, comprising at least one executable instruction, wherein the executable instruction enables a processor to perform operations corresponding to the above-mentioned video generation method.

[0043] According to the video generation method, device, computing device, storage medium and program product provided in the embodiment of the present application, the target story text is obtained; the target story text is split and processed to obtain scene information, character information and subtitle information, wherein there is a correlation between the character information and the subtitle information; according to the character information, the character stickers corresponding to the subtitle information that has a correlation with the character information are determined, and the scene material corresponding to the scene information is determined; according to the subtitle information, the character stickers and the scene material, a story video is generated. The technical solution of the embodiment of the present application lowers the threshold for video production. Users only need to provide story proposals and use AI to empower creation to assist users in generating videos with meme-like gameplay. The video uses certain specific hot-stalk materials to express the user's emotional feelings in a certain scenario through pictures, realizing the videoization of text stories. For users who lack image materials or only have picture materials, it also inspires user inspiration and helps users generate videos faster and easier, improves video production efficiency, and enhances user experience.

[0044] The above description is only an overview of the technical solution of the embodiment of the present application. In order to more clearly understand the technical means of the embodiment of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the embodiment of the present application more obvious and easy to understand, the specific implementation method of the embodiment of the present application is listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only used for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the embodiments of the present application. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0046] Figure 1 A schematic diagram of a process of a video generation method according to Embodiment 1 of the present application is shown;

[0047] Figure 2 This is a schematic diagram of the text editing page;

[0048] Figure 3 This is a schematic diagram of the corresponding scene materials inside the convenience store and next to the shelves;

[0049] Figure 4 This is a schematic diagram of the scene material corresponding to the cash register;

[0050] Figure 5 A schematic diagram of a sticker image in a character sticker corresponding to the character information;

[0051] Figure 6 A schematic diagram of adding some subtitle information, sticker images, and scene materials to the corresponding tracks;

[0052] Figure 7 A schematic diagram of some video frames of the generated story video;

[0053] Figure 8 An interactive diagram of a video generation method according to Embodiment 2 of the present application is shown;

[0054] Fig. 9 A timing diagram of a video generation method according to Embodiment 3 of the present application is shown;

[0055] Fig.10 A structural block diagram of a video generating device according to a fourth embodiment of the present application is shown;

[0056] Fig.11 A schematic diagram of the structure of a computing device according to Embodiment 5 of the present application is shown. DETAILED DESCRIPTION

[0057] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary 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 limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0058] It should be noted that the user-related data involved in the embodiments of the present application are all information and data that are authorized 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 choosing to authorize or reject.

[0059] First, the terms involved in one or more embodiments of the present application are explained.

[0060] meme: meme, Internet meme (a picture, video, etc. transmitted from one user to another on the Internet); imitative transmission behavior (behavior transmitted through non-genetic means such as imitation);

[0061] Large language model: A large language model is a commonly used model in the field of natural language processing. It is used to process a variety of natural language tasks, such as text classification, question answering, and dialogue, to generate natural language text or understand the meaning of language text. Large language model is also a general term for deep learning models trained with large amounts of text data. It can better train analysis, reasoning, and output key information of text;

[0062] Expansion: Use the large language template to generate text with the corresponding number of words and style from the given prompt;

[0063] Sensitive word check: Identify whether there are sensitive words in the text;

[0064] Intelligent sentence segmentation: Cut the given text into several sentences of a certain length as required;

[0065] Script sorting: Match the expanded text and materials so that the generated video can better match the story content;

[0066] Json: is an open standard file format and data exchange format that is easy for people to read and write, and easy for machines to parse and generate;

[0067] TimeLine: Timeline refers to the project data obtained by expanding the video in chronological order, including video track, audio track, text track, sticker track, etc.

[0068] Material positioning: refers to positioning the material frame according to the corresponding coordinate axis;

[0069] Sound Effect Stickers: Sticker material with a bound audio track;

[0070] Material library: a place to store and manage various types of materials, which can be pictures, videos, audio, text, etc.

[0071] Embodiment 1

[0072] Figure 1 FIG. 4 shows a flow chart of a video generation method according to Embodiment 1 of the present application, as shown in FIG. Figure 1 As shown, the method comprises the following steps:

[0073] Step S101, obtaining the target story text.

[0074] The target story text refers to the basic story content that is written and used to further generate story videos.

[0075] The user can directly provide the target story text. For example, the user inputs the target story text through the client or uploads a story text file that has been written and contains the target story text, thereby obtaining the target story text. In addition, the user can also only provide a story proposal and process the story proposal using the following method to obtain the target story text: determine the story proposal to be processed; use the AI ​​model to process the story proposal to be processed into a target story text in a preset format according to the preset first text prompt information.

[0076] A story proposal is a short text description that summarizes the main clues and key points of the story. It can accurately convey the core idea and theme of the whole story in one sentence, a few sentences or a paragraph. For example, a story proposal is "A little boy found a magical wand, and he created many wonderful things with this magic wand." Of course, a story proposal can also be composed of multiple paragraphs. There is no specific limitation here. The story proposal can be able to clearly and clearly describe the whole storyline.

[0077] In one embodiment of the present application, the client can provide the user with a "+" text entry, or other follow-up entry for non-plus sign scenes. After the user clicks the corresponding entry, a text editing page will be displayed to the user, such as Figure 2 As shown, the text editing page includes a text input box and a "one-click generation" control. The user can enter the story proposal to be processed in the text input box. After the user has entered the story proposal, he / she clicks the "one-click generation" control in the text editing page to determine the story proposal to be processed this time.

[0078] If the client resources are sufficient and large enough, the AI ​​model can be deployed on the client; of course, in order to save the client's computing resources, the AI ​​model can also be deployed on the AI ​​model server or video server. Among them, the AI ​​model can include a large language model.

[0079] In the case where the AI ​​model is deployed on the client, after determining the pending story proposal, the AI ​​model can be directly used to process the pending story proposal into a target story text in a preset format according to the preset first text prompt information. For example, the first text prompt information corresponding to the pending story proposal can be read from the database first, and then the pending story proposal can be input into the AI ​​model for processing to generate the target story text in a preset format. This processing method requires that the pending story proposal input by the user itself meets the requirements of the replica meme, that is, there is no need to rewrite the pending story proposal into a meme.

[0080] An example of the first text prompt information is given below. It should be noted that the first text prompt information can be changed according to the actual target story text to be output:

[0081] Please rewrite the story proposal into a dialogue script divided into scenes, and describe the character's actions before the dialogue to restore the story scene. If it is a first-person narration, use "I" as the character name. scene_info represents the scene, voiceover represents the narration, and conversation represents the dialogue. Each sentence is wrapped in the English symbol "", and multiple sentences are separated by English commas. Emotions are wrapped in %% and do not use () to separate. Remember to output in Chinese.

[0082] The output format is:

[0083]

[0084] The following is an example of a target story text in a preset format after being processed using the first text prompt information:

[0085] Scene 1:

[0086] "Convenience store"

[0087] %Daily% (I walked into a convenience store and prepared to buy a lunch box)

[0088] My question: Boss, which type of bento is more delicious?

[0089] Boss %busy%: (without raising his head) The ones on the far right are quite popular, you can choose.

[0090] Act II:

[0091] "Next to the shelf]

[0092] % Hesitation% (I walked to the far right, staring at the dazzling array of lunch boxes, hesitant)

[0093] Me% talking to myself%: They all look good, which one should I choose?

[0094] Act III:

[0095] "checkout counter]

[0096] %Decision% (Finally choose one, typing on the computer and walking towards the cashier)

[0097] Me: Boss, I want this.

[0098] Boss %impatient%: (looked up and quickly scanned the code) Five dollars and fifty cents, WeChat or cash?

[0099] It should be noted that the target story text is only schematically shown here, and those skilled in the art can set it to other formats according to actual needs.

[0100] In an optional implementation of the present application, it is very likely that the story proposal to be processed input by the user does not meet the requirements of replicating the meme. For such story proposal to be processed, an AI model can be used to process it in two steps to finally obtain a target story text in a preset format: first, the AI ​​model is used to identify keywords in the story proposal to be processed, and meme elements matching the keywords are obtained. A meme story text is generated based on the meme elements and the story proposal to be processed, and then, the meme story text is rewritten into a target story text in a preset format according to a preset first text prompt information.

[0101] Specifically, the AI ​​model has a semantic analysis function. For example, the AI ​​model can be a model such as OpenAI GPT, Wenxin Yiyan, etc., which can perform keyword recognition on a given text. Therefore, the AI ​​model can be used to perform semantic analysis on the story proposal to be processed, so as to identify the keywords in the story proposal to be processed. Here, the main purpose is to identify some words that can express emotional information, scene information, and character information. Then, combined with the identified keywords, meme elements matching the keywords are obtained. The meme elements here mainly include: Internet buzzwords, stalks, etc. The meme elements are integrated into the story proposal to be processed, and the story content is enriched and expanded to generate a meme story text. For example, the second text prompt information can be obtained from the database. The second text prompt information describes the requirements that the AI ​​model needs to complete, such as introducing meme elements, continuing to expand the description of the story, etc. Then, the meme story text is generated according to the second text prompt information and the story proposal to be processed. The meme story text is a descriptive story text that meets the requirements of replicating the meme. In order to facilitate the subsequent generation of videos, it is also necessary to use the AI ​​model to process the meme story text, read the first text prompt information corresponding to the story proposal to be processed from the database, and then input the story proposal to be processed into the AI ​​model for processing to generate the target story text in a preset format.

[0102] In an optional implementation, the AI ​​model is deployed on the model server, so the processing of the pending story proposals can be achieved by the following method:

[0103] The client generates a story proposal processing request, which carries a story proposal to be processed, and sends the story proposal processing request to the video server. After receiving the story proposal processing request, the video server calls the AI ​​model to process the story proposal to be processed, mainly processing the story proposal to be processed into a target story text in a preset format according to a pre-set first text prompt information.

[0104] Continue to combine Figure 2 To explain, users Figure 2 Enter your story proposal in the text input box and click Figure 2 After clicking the "One-click Generate" control in the text box, the client generates a story proposal processing request, which carries the story proposal entered by the user in the text box, and sends the story proposal processing request to the video server. The video server responds to the story proposal processing request, calls the AI ​​model, and processes the story proposal to be processed, mainly processing the story proposal to be processed into a target story text in a preset format according to the preset first text prompt information. The specific implementation process is similar to that described above and will not be repeated here.

[0105] In an alternative embodiment, Figure 2 The text editing page in the text editing page can also include an "AI rewrite" control. When the user is not satisfied with the story proposal filled in, or the user simply writes the story proposal, the user can use Figure 2 In the Chinese text editing page, the "AI expansion" control is used to rewrite the story. In the case where the AI ​​model is deployed on the model server, the following methods can be used to process the story proposals:

[0106] The client generates a story proposal expansion request, which carries a pending story proposal, and sends the story proposal expansion request to the video server. After receiving the story proposal expansion request, the video server calls the AI ​​model to expand the pending story proposal, mainly to identify keywords in the pending story proposal, obtain meme elements matching the keywords, and generate a meme story text based on the meme elements and the pending story proposal.

[0107] After using the AI ​​model to generate the meme story text, the video server returns the meme story text to the client, and the client can display the meme story text in the text editing page. For example, Figure 2 The pending story proposals in the Chinese text editing page are updated to meme story text.

[0108] When the user is satisfied with the meme story text displayed on the text editing page and has no modification suggestions, he can click Figure 2 After clicking the "One-click Generate" control in the text box, the client generates a story proposal processing request, which carries the story proposal entered by the user in the text box, and sends the story proposal processing request to the video server. The video server responds to the story proposal processing request, calls the AI ​​model, and processes the story proposal to be processed, mainly processing the story proposal to be processed into a target story text in a preset format according to the preset first text prompt information. The specific implementation process is similar to that described above and will not be repeated here.

[0109] In an optional implementation, after the user enters a story proposal in the text input box, the story proposal can be risk-controlled, mainly by using a pre-set sensitive word library to identify whether there are sensitive words in the story proposal to be processed, for example, Figure 2 To explain, Figure 2The text editing page in can also include a "risk control" control. When a user clicks the "risk control" control, it is deemed that a risk control processing request has been triggered. The pending story proposal is segmented and the segmentation result is matched with the sensitive word library to determine whether there are sensitive words in the pending story proposal. If there are sensitive words in the pending story proposal, a reminder message is displayed in a preset manner. For example, a pop-up window can be displayed, which displays the existence of sensitive words: xxx, aaa, etc., please modify.

[0110] In addition, the risk control processing process can also be executed on the video server. Specifically, the client generates a risk control processing request, wherein the risk control processing request carries a story proposal to be processed, and sends the risk control processing request to the video server. The video server uses a pre-set sensitive word library to identify whether there are sensitive words in the story proposal to be processed. If there are sensitive words, the specific sensitive words are returned to the client. If there are no sensitive words, "no sensitive words" are returned to the client. The client displays a pop-up window based on the information returned by the video server. If there are sensitive words, the pop-up window displays the existence of sensitive words: xxx, aaa, etc., please modify. If there are no sensitive words, the pop-up window displays "no sensitive words".

[0111] In an optional implementation, after the meme story text is generated, the meme story text can be displayed to the user in the text editing page. In order to avoid affecting the video release due to the presence of sensitive words in the story text, the user can perform risk control processing on the meme story text. Specifically, the meme story text is identified for sensitive words; if the meme story text contains sensitive words, a reminder message is displayed in a preset manner to remind the user to make modifications. Similarly, the risk control processing of the meme story text can be performed by the client or by the video server.

[0112] Step S102, performing content splitting processing on the target story text to obtain scene information, character information and subtitle information, wherein there is a correlation between the character information and the subtitle information.

[0113] Specifically, the target story text is subjected to content splitting processing. For example, the scene information corresponding to the story can be determined by matching the words in the target story text with the preset scene keyword library, and the preset characters in the target story text can be determined. The words in the preset characters are determined as role information, etc. For example, keywords such as the first act and the second act are used to represent the scene, and the information in "%%" is used to represent the role information. It should be noted that content splitting can also be achieved by other means. Through content splitting, scene information, role information and subtitle information are finally obtained. Among them, there is an association relationship between the role information and the subtitle information. For example, the subtitle information contains the corresponding role information, or a mapping relationship is established between the role information and the subtitle information.

[0114] In addition, the target story text may also be in a preset format, such as JSON format. Therefore, the content may be split according to the definition information of the preset format. Combining the above example, after the content splitting process, the following information is obtained, wherein the subtitle information may include narration information and dialogue information:

[0115] 1. Scene Information

[0116] Scene 1: "Inside a convenience store"

[0117] Scene 2: "Next to the Shelf"

[0118] Scene 3: "Cashier"

[0119] 2. Emotions

[0120] %Daily%、%Question%、%Busy%、%Hesitation%、%Talking to oneself%、%Decision%、%Impatient%

[0121] 3. Narration

[0122] Scene 1: %Daily% (I walk into a convenience store and prepare to buy a lunch)

[0123] Scene 2: %Hesitation% (I walked to the far right, staring at the dazzling array of lunch boxes, hesitant)

[0124] Act 3: %Decision% (finally choose one, typing on the computer and walking towards the cashier)

[0125] 4. Dialogue

[0126] Scene 1:

[0127] My question: Boss, which type of bento tastes better?

[0128] Boss %busy%: (without raising his head) The ones on the far right are quite popular, you can choose.

[0129] Act II:

[0130] Me% talking to myself%: They all look good, which one should I choose?

[0131] Act III:

[0132] I %decided%: Boss, I want this.

[0133] Boss %impatient%: (looked up and quickly scanned the code) Five dollars and fifty cents, WeChat or cash?

[0134] Step S103, determining the character stickers corresponding to the subtitle information associated with the character information according to the character information, and determining the scene material corresponding to the scene information.

[0135] Specifically, a character library can be pre-established, which contains various types of character stickers, such as character stickers of meow, meow, etc. Similarly, a scene material library can also be established, which stores scene materials of various styles.

[0136] After performing content splitting processing to obtain character information and scene information, the character library can be queried based on the character information. If there are matching character stickers in the character library, the matching character stickers in the character library are determined as the character stickers corresponding to the subtitle information that is associated with the character information. If there are multiple character stickers matching the character information in the character library, one can be randomly selected from them, or multiple character stickers can be displayed to the user for selection by the user; if not, the default character sticker is determined as the character sticker for the character information, or a default character sticker list is pre-set, and a character sticker is randomly extracted from the default character sticker list, and the randomly extracted character sticker is determined as the character sticker for the character information. Each character information corresponds to a character sticker. For example, it can be a meme cat sticker.

[0137] The scene material library is queried according to the scene information. If there is a matching scene material in the scene material library, the matching scene material in the scene material library is determined as the scene material of the scene information; if not, a default scene material is determined as the scene material of the scene information, or a default scene material list is pre-set, and a scene material is randomly extracted from the scene material list, and the randomly extracted scene material is determined as the scene material of the scene information. Combined with the above example, each scene corresponds to a scene material.

[0138] In an optional implementation, the role library may include: an action role library and / or an emotion role library, and accordingly, the role information may include emotion information and / or action information;

[0139] If the character information contains emotional information, the character stickers corresponding to the subtitle information that is associated with the character information can be determined by the following method: query the emotional character library based on the emotional information to obtain character stickers that match the emotional information; and determine the queried character stickers as the character stickers corresponding to the subtitle information that is associated with the emotional information.

[0140] The emotion character library stores character stickers of various emotions. For example, the emotion character library stores the mapping relationship between emotion information and character stickers. Therefore, the emotion character library can be queried through the emotion information obtained by content splitting to determine the character stickers that match the emotion information, and the matched character stickers can be determined as the character stickers corresponding to the subtitle information that has an associated relationship with the emotion information.

[0141] If the character information includes: action information; the character sticker corresponding to the subtitle information that is associated with the character information can be determined by the following method: query the action character library according to the action information to obtain character stickers that match the action information; determine the queried character stickers as the character stickers corresponding to the subtitle information that is associated with the action information.

[0142] The action role library stores character stickers corresponding to various actions. For example, the action role library stores the mapping relationship between action information and character stickers. Therefore, the action role library can be queried through the action information obtained by content splitting to determine the character stickers that match the action information, and the matched character stickers can be determined as the character stickers corresponding to the subtitle information that has an associated relationship with the action information.

[0143] In addition, when querying the action character library, action keywords used to represent actions can be extracted from the subtitle information, and the action character library can be queried based on the action information and the action keywords to obtain character stickers that match the action information.

[0144] It should be noted that those skilled in the art can set the query priority for querying the action role library and the emotion role library, and perform query matching of character stickers according to the query priority, or set all subtitle information to match character stickers only by querying the action role library or only by querying the emotion role library.

[0145] In an optional implementation, the subtitle information may include: narration subtitle information and dialogue subtitle information. It is pre-defined here that for narration subtitle information, the role sticker is matched from the action role library first, and if it cannot be matched, the emotional role library is used. For dialogue subtitle information, the role sticker is matched from the emotional role library. Therefore, the role sticker corresponding to the role information can be determined by the following method:

[0146] The action role library is queried based on the role information that is associated with the narration subtitle information. If there are matching role stickers in the action role library, the role stickers in the queried action role library are used as the role stickers corresponding to the narration subtitle information. If there are no matching role stickers in the action role library, the emotion role library is queried based on the role information corresponding to the narration subtitle information to obtain the corresponding role stickers. When querying the action role library, the action words in the narration subtitle information can also be combined for query, that is, the action role library is queried in combination with the emotion information and the action words.

[0147] The emotional role library is searched according to the role information associated with the dialogue subtitle information to obtain the role stickers corresponding to the dialogue subtitle information.

[0148] If neither the emotional character library nor the action character library matches a suitable character sticker, the default character sticker can be determined as the character sticker of the character information, or a default character sticker list can be pre-set, and a character sticker can be randomly selected from the default character sticker list, and the randomly selected character sticker can be determined as the character sticker of the character information. Continuing with the above example, the corresponding scene materials in the convenience store and beside the shelf can be obtained, such as Figure 3 As shown, the scene material corresponding to the cash register is as follows Figure 4 As shown, the sticker images of the character stickers matching the character information are as follows Figure 5 shown.

[0149] After completing the scene material matching, the client can be used to display the scene material to the user. If the user is not satisfied with the matched scene material, the scene material can be switched. For example, the client provides the user with a scene material switching control. The user clicks the scene material switching control and is displayed to the scene material library by default. The scene material library is grouped and displayed according to different tab categories. When the user selects a scene from the previous page and enters the material library, the scene will be selected by default and the user can change the material.

[0150] Step S104, generating a story video based on the subtitle information, character stickers, and scene materials.

[0151] Specifically, video synthesis technology can be used to integrate elements such as subtitle information, character stickers, and scene materials to generate a final story video. For example, a timeline can be created, subtitle information, character stickers, and scene materials can be laid out on the timeline, and then rendered to generate a story video.

[0152] More specifically, the timeline includes: video track, audio track, subtitle track, sticker track; character stickers include: sticker images and sticker sound effects;

[0153] Laying subtitle information, character stickers, and scene materials to the timeline further includes:

[0154] Create video track, audio track, subtitle track, and sticker track in the video generation page;

[0155] Add subtitle information to the subtitle track, add sticker sound effects to the audio track, add sticker images to the sticker track, and add scene materials to the video track. According to the association between scene materials, sticker images, sticker sound effects, and subtitle information, the scene materials, character stickers, sticker sound effects, and subtitle information can be matched to avoid the generated video not corresponding to the story content due to mismatch. The subtitle information corresponds to the sticker image one by one. The audio track is not displayed in the video generation page. The audio track is used to carry the sound effects of the character stickers, preview the sound effects, and synthesize the sound effects. The video track is called the main track. When the video is actually generated, the character information is not displayed. If the subtitle information contains dialogue subtitle information and narration subtitle information, the sticker image and dialogue subtitle information, the sticker image and narration subtitle information are packaged here and filled into the sticker track and subtitle track.

[0156] The mapping relationship between character stickers and subtitle information is recorded to facilitate subsequent batch editing or modification. The audio duration carried by the character stickers is used as a benchmark to set the display duration of the subtitle information and the display duration of the sticker image to be the same as the duration of the sticker sound effect. The start display time of the sticker image and the start playback time of the sticker sound effect are set according to the subtitle information display timestamp. The entry and exit points and time of the sticker image follow the subtitle information. The coordinate information (including horizontal and vertical coordinates) corresponding to the subtitle information and the sticker image is set. The coordinate information here refers to the position information of the subtitle information and the sticker image in the scene material. For example, the mapping relationship between the aspect ratio of different scene materials and the coordinate information of the subtitle information and the coordinate information of the sticker image can be pre-set. After the aspect ratio of the scene material is determined, the mapping relationship table is queried according to the aspect ratio of the scene material to obtain the coordinate information corresponding to the subtitle information and the sticker image. In addition, it also supports users to upload scene materials with different aspect ratios. Therefore, the horizontal and vertical coordinates of the sticker image and subtitle information can be matched according to the aspect ratio of the scene material uploaded by the user. The aspect ratio can be: 16:9, 4:3, 1:1, 9:16, 3:4, etc. Set the display duration of the scene material to be the same as the total display duration of all subtitle information contained in the scene. Continuing with the above example, Figure 6 As shown in FIG. 1 , it schematically lists some subtitle information, sticker images, and scene materials added to the corresponding track. Figure 7 As shown, it gives a schematic diagram of some video frames in the generated story video.

[0157] In an optional implementation, the client provides a batch editing function, allowing the user to perform batch editing or modification of character stickers. After the user selects a character sticker in the client, the client receives a sticker batch editing request triggered by the user, wherein the sticker batch editing request carries the identifier of the character sticker to be edited;

[0158] In response to a sticker batch editing request, the subtitle information and character stickers associated with the character sticker identifier to be edited are displayed on the sticker editing page to allow users to perform batch editing. Character stickers and subtitle information correspond to the start time one by one and are displayed in chronological order; if there are character stickers or subtitle information with inconsistent start times, it means that a character sticker is paired with other character stickers, then they will be displayed separately and cannot be added in batch editing, only modification is supported.

[0159] According to the video generation method provided in the embodiment of the present application, the target story text is obtained; the target story text is split and processed to obtain scene information, character information and subtitle information, wherein there is a correlation between the character information and the subtitle information; according to the character information, the character stickers corresponding to the subtitle information that is correlated with the character information are determined, and the scene material corresponding to the scene information is determined; according to the subtitle information, the character stickers and the scene material, a story video is generated. The technical solution of the embodiment of the present application lowers the threshold for video production. Users only need to provide story proposals and use AI to empower creation to assist users in generating videos with meme-like gameplay. The video uses certain specific hot-stalk materials to express the user's emotional feelings in a certain scenario through pictures, realizing the videoization of text stories. For users who lack image materials or only have picture materials, it also inspires user inspiration and helps users generate videos faster and easier, improving video production efficiency and user experience.

[0160] Embodiment 2

[0161] Figure 8 An interactive diagram of a video generation method according to Embodiment 2 of the present application is shown, such as Figure 8 As shown in the figure, this method is to complete the generation of the video through information interaction between the client, the video server, and the model server. The specific implementation process is as follows:

[0162] The client provides users with multiple follow-up entrances, for example, a large plus sign text entry and other follow-up entrances for non-plus sign scenes, wherein the large plus sign text entry is a functional entry that allows users to automatically generate video content by entering text. After the user clicks the corresponding large plus sign text entry or other follow-up entrances for non-plus sign scenes, the client can display some gameplay cards to the user, and the gameplay attributes of the gameplay cards can be set by the user as needed. Among them, the video server stores the gameplay attributes (including: timbre, gags, etc. of tts), and returns the corresponding gameplay attributes to the client to display them to the user, so that the user can set them as needed.

[0163] After the user clicks the start creation function provided by the client, the user enters the gameplay editing page, where the user enters the story script, where the story script is the story proposal mentioned in the above example, and the client can identify sensitive words in the entered story script, and if there are sensitive words, the user can be reminded to re-enter the story script. The user can select or enter the rewriting style of the story script, where the rewriting style can be a dialogue style, a Tucao style, etc. The client sends the story script and the rewriting style to the video server, which stores second text prompt information of different styles. The video server can select the appropriate second text prompt information according to the rewriting style sent by the client, call the large language model for rewriting processing, thereby generating a rewritten story script and returning it to the client, and the client applies the rewritten story script output returned by the video server to the gameplay editing page, thereby displaying it to the user.

[0164] The client provides a "one-click generation" control. When the user clicks one-click generation, the client can perform verification on the end to verify whether there is a script or non-compliance. If there is no script or non-compliance, the user can be prompted to modify it. In addition, the rewritten story script can also be identified for sensitive words. If sensitive words exist, the user is reminded to re-enter. If there is no non-compliance and no sensitive words, the large language model can be used for rewriting. The large language model rewrites according to the first text prompt information and rewrites it into a story text with a json structure. Then, the content is split into scene information, role information, narration subtitle information, and dialogue subtitle information, and the json data is returned to the client. The client can query according to the scene information. Query and match the appropriate scene material, query and match the appropriate character sticker according to the character information, enter the editor, create the material main track, sticker track, audio track, and text track, and you can sort the scene material, character sticker, narration subtitle information, and dialogue subtitle information in the order of the storyline. After the sorting is completed, lay the scene material to the material main track, lay the sticker image in the character sticker to the sticker track, and lay the sticker sound effect in the character sticker to the audio track. Among them, the audio track is not displayed on the UI, and lay the narration subtitle information and dialogue subtitle information to the text track, so that the scene material and character sticker can correspond to the narration subtitle information and dialogue subtitle information, and then call the rendering engine to render and generate the video, and publish it to the release page. In addition, users can also batch edit and modify the text and sticker relationship on the client to improve the efficiency of video generation.

[0165] Embodiment 3

[0166] Fig. 9 FIG. 4 shows a timing diagram of a video generation method according to Embodiment 3 of the present application, such as Fig. 9As shown in the figure, this method is to complete the generation of the video through information interaction between the client, the video server, and the model server. The specific implementation process is as follows:

[0167] The user fills in the story proposal to be processed in the text editing page displayed on the client;

[0168] After obtaining the pending story proposal, the client sends an information acquisition request to the video server, wherein the information acquisition request carries the pending story proposal; the video server can identify key information based on the pending story proposal, identify the role information / scene information, and return the role stickers corresponding to the role information and the scene materials corresponding to the scene information to the client;

[0169] If the recognition is successful, the client can update the character stickers and scene materials (scene UI); if the recognition fails, the client can display a prompt box to prompt the user that the recognition failed, so that the user can manually set the character stickers and scene materials, or modify the pending story proposal and recognize again;

[0170] If the user wants to preview the sound effect, he can preview the sound effect by clicking on the character sticker presented by the client. The client can consider whether to cache the sound effect as needed. By caching the sound effect, it is convenient to quickly play the corresponding sound effect later.

[0171] If the user is not satisfied with the character stickers returned by the video server, he can switch the character stickers through the client, thereby calling up the sticker panel, so that the user can select the appropriate character stickers in the sticker panel. When switching character stickers, the client also provides a sound effect preview function and supports sound effect preview;

[0172] If the user is not satisfied with the scene material returned by the video server, he can switch the scene through the client, thereby calling up the material library, and displaying the most recently collected pictures in the material library to the user for the user to select a suitable scene. Among them, the user can only select one picture at a time as the scene after switching.

[0173] The user clicks the one-click generation control to start generating the corresponding video. The client further determines whether the word count of the pending story proposal exceeds the preset story text word count threshold. If it exceeds the preset story text word count threshold, a reminder pop-up window can be displayed to guide the user to truncate the pending story proposal in a pop-up window manner, or directly use the AI ​​model to simplify the word count of the pending story proposal. If it does not exceed the preset story text word count threshold, or after the truncation or word count simplification is completed, the pending story proposal can be sent to the video server for risk control processing. The video server mainly uses the sensitive word library to identify sensitive words in the pending story proposal and returns the risk control results to the client.

[0174] If the risk control result is that the risk control is passed, the client provides the first text prompt information to the video server, and the video server calls the AI ​​model to process the pending story proposals that have passed the risk control into a target story text in a preset format according to the first text prompt information, and then splits the content into: scene information, character information, narration subtitle information, and dialogue subtitle information, and returns json data to the client. The json data here is the scene information, character information, narration subtitle information, and dialogue subtitle information. After receiving the scene information, character information, narration subtitle information, and dialogue subtitle information returned by the video server, the client matches the scene material according to the scene information, and matches the character material (i.e., character stickers) according to the character information. Then, track laying data preparation is performed. For example, the scene material, character stickers, narration subtitle information, and dialogue subtitle information are script-sorted according to the order of development of the storyline, so that the scene material, character stickers, narration subtitle information, and dialogue subtitle information can correspond to each other. Finally, enter the editor for editing.

[0175] Embodiment 4

[0176] Fig.10 FIG. 4 shows a structural block diagram of a video generating device according to a fourth embodiment of the present application. Fig.10 As shown, the device comprises:

[0177] An acquisition module 1001 is adapted to acquire a target story text;

[0178] The content splitting processing module 1002 is adapted to perform content splitting processing on the target story text to obtain scene information, character information and subtitle information, wherein there is a correlation relationship between the character information and the subtitle information;

[0179] A determination module 1003 is adapted to determine, according to the character information, a character sticker corresponding to the subtitle information that is associated with the character information, and determine a scene material corresponding to the scene information;

[0180] The generation module 1004 is suitable for generating a story video based on subtitle information, character stickers, and scene materials.

[0181] Optionally, the character information includes emotion information;

[0182] The determination module is further adapted to: query the emotional role library according to the emotional information to obtain role stickers matching the emotional information;

[0183] The retrieved character stickers are determined as character stickers corresponding to the subtitle information that is associated with the emotion information.

[0184] Optionally, the role information includes: action information;

[0185] The determination module is further adapted to: query an action role library according to the action information to obtain a role sticker matching the action information;

[0186] The retrieved character sticker is determined as the character sticker corresponding to the subtitle information that is associated with the action information.

[0187] Optionally, the acquisition module is further adapted to: determine a story proposal to be processed;

[0188] Using the AI ​​model, the story proposal to be processed is processed into a target story text in a preset format according to the preset first text prompt information.

[0189] Optionally, the acquisition module is further adapted to: using the AI ​​model to identify keywords in the story proposal to be processed, acquiring meme elements matching the keywords, and generating a meme story text according to the meme elements and the story proposal to be processed;

[0190] The meme story text is rewritten into a target story text in a preset format according to the preset first text prompt information.

[0191] Optionally, the device further comprises: a sensitive word identification module, adapted to identify sensitive words in the meme story text;

[0192] If there are sensitive words in the meme story text, a reminder message will be displayed in a preset manner to remind the user to make modifications.

[0193] Optionally, the generation module is further adapted to: lay out subtitle information, character stickers, and scene materials on a timeline, and render and generate a story video.

[0194] Optionally, the timeline includes: a video track, an audio track, a subtitle track, and a sticker track; the character stickers include: sticker images and sticker sound effects;

[0195] The generation module is further adapted to: create a video track, an audio track, a subtitle track, and a sticker track in a video generation page;

[0196] Add subtitle information to the subtitle track, add sticker sound effects to the audio track, add sticker images to the sticker track, and add scene materials to the video track;

[0197] Record the mapping relationship between character stickers and subtitle information, set the display duration of subtitle information, sticker image display duration and sticker sound effect duration to be the same, set the start display time of sticker image and the start playback time of sticker sound effect according to the subtitle information display timestamp, and set the coordinate information corresponding to the subtitle information and sticker image.

[0198] Optionally, the device further comprises: a receiving module, adapted to receive a sticker batch editing request triggered by a user, wherein the sticker batch editing request carries an identifier of a character sticker to be edited;

[0199] The display module is adapted to respond to a sticker batch editing request and display subtitle information and character stickers associated with the character sticker identifier to be edited in the sticker editing page to enable the user to perform batch editing.

[0200] The description of each module above refers to the corresponding description in the method embodiment and will not be repeated here.

[0201] According to the video generation device provided in the embodiment of the present application, the target story text is obtained; the target story text is split and processed to obtain scene information, character information and subtitle information, wherein there is a correlation between the character information and the subtitle information; according to the character information, the character stickers corresponding to the subtitle information that is correlated with the character information are determined, and the scene material corresponding to the scene information is determined; according to the subtitle information, the character stickers and the scene material, a story video is generated. The technical solution of the embodiment of the present application lowers the threshold for video production. Users only need to provide story proposals and use AI to empower creation to assist users in generating videos with meme-like gameplay. The video uses certain specific hot-stalk materials to express the user's emotional feelings in a certain scenario through pictures, realizing the videoization of text stories. For users who lack image materials or only have picture materials, it also inspires user inspiration and helps users generate videos faster and easier, improves video production efficiency, and enhances user experience.

[0202] An embodiment of the present application provides a non-volatile computer storage medium, which stores at least one executable instruction or computer program, and the executable instruction or computer program can enable a processor to perform operations corresponding to the video generation method in any of the above method embodiments.

[0203] An embodiment of the present application provides a computer program product, which includes at least one executable instruction or computer program, and the executable instruction or computer program can enable a processor to perform operations corresponding to the video generation method in any of the above method embodiments.

[0204] Embodiment 5

[0205] Fig.11 A schematic diagram of the structure of a computing device according to Embodiment 5 of the present application is shown. The specific embodiment of the present application does not limit the specific implementation of the computing device.

[0206] like Fig.11 As shown, the computing device may include: a processor (processor) 1102 , a communication interface (Communications Interface) 1104 , a memory (memory) 1106 , and a communication bus 1108 .

[0207] The processor 1102, the communication interface 1104, and the memory 1106 communicate with each other via a communication bus 1108. The communication interface 1104 is used to communicate with other devices such as a client or other server network elements. The processor 1102 is used to execute the program 1110, which can specifically execute the relevant steps in the above-mentioned video generation method embodiment for a computing device.

[0208] Specifically, the program 1110 may include program codes, which include computer operation instructions.

[0209] The processor 1102 may be a central processing unit (CPU), or an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application. The one or more processors included in the computing device may be processors of the same type, such as one or more CPUs; or may be processors of different types, such as one or more CPUs and one or more ASICs.

[0210] The memory 1106 is used to store the program 1110. The memory 1106 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0211] The program 1110 can be specifically used to enable the processor 1102 to execute the video generation method in any of the above method embodiments. The specific implementation of each step in the program 1110 can refer to the corresponding descriptions in the corresponding steps and units in the above video generation embodiment, which will not be repeated here. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working process of the above-described devices and modules can refer to the corresponding process description in the above method embodiments, which will not be repeated here.

[0212] The algorithm and display provided herein are not inherently related to any specific computer, virtual system or other equipment. Various general purpose systems can also be used together with the teachings based on this. According to the above description, it is obvious to construct the structure required for this type of system. In addition, the present application embodiment is not directed to any specific programming language yet. It should be understood that various programming languages ​​can be utilized to realize the content of the present application embodiment described herein, and the above description of specific languages ​​is to disclose the best mode of implementation of the present application embodiment.

[0213] In the description provided herein, a large number of specific details are described. However, it is understood that the embodiments of the present application can be practiced without these specific details. In some instances, well-known methods, structures and techniques are not shown in detail so as not to obscure the understanding of this description.

[0214] Similarly, it should be understood that in order to streamline the present disclosure and help understand one or more of the various inventive aspects, in the above description of the exemplary embodiments of the present application, the various features of the embodiments of the present application are sometimes grouped together into a single embodiment, figure, or description thereof. However, the disclosed method should not be interpreted as reflecting the following intention: the claimed embodiments of the present application require more features than the features clearly stated in each claim. More specifically, as reflected in the claims below, the inventive aspects are less than all the features of the single embodiment disclosed above. Therefore, the claims following the specific embodiment are hereby explicitly incorporated into the specific embodiment, wherein each claim itself serves as a separate embodiment of the embodiment of the present application.

[0215] Those skilled in the art will appreciate that the modules in the devices in the embodiments may be adaptively changed and arranged in one or more devices different from the embodiments. The modules or units or components in the embodiments may be combined into one module or unit or component, and in addition they may be divided into a plurality of submodules or subunits or subcomponents. Except that at least some of such features and / or processes or units are mutually exclusive, all features disclosed in this specification (including the accompanying claims, abstracts and drawings) and all processes or units of any method or device disclosed in this manner may be combined in any combination. Unless otherwise expressly stated, each feature disclosed in this specification (including the accompanying claims, abstracts and drawings) may be replaced by an alternative feature providing the same, equivalent or similar purpose.

[0216] In addition, those skilled in the art will appreciate that, although some embodiments described herein include certain features included in other embodiments but not other features, the combination of features of different embodiments is meant to be within the scope of the embodiments of the present application and to form different embodiments. For example, in the claims below, any one of the claimed embodiments may be used in any combination.

[0217] The various component embodiments of the embodiments of the present application can be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. It should be understood by those skilled in the art that a microprocessor or digital signal processor (DSP) can be used in practice to implement some or all functions of some or all components according to the embodiments of the present application. The embodiments of the present application can also be implemented as a device or apparatus program (e.g., computer program and computer program product) for executing part or all of the methods described herein. Such a program implementing the embodiments of the present application can be stored on a computer-readable medium, or can have the form of one or more signals. Such a signal can be downloaded from an Internet website, or provided on a carrier signal, or provided in any other form.

[0218] It should be noted that the above embodiments illustrate rather than limit the embodiments of the present application, and that those skilled in the art may design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference symbol between brackets should not be constructed as a limitation to the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "one" or "an" preceding an element does not exclude the presence of multiple such elements. The embodiments of the present application may be implemented by means of hardware including several different elements and by means of appropriately programmed computers. In a unit claim that lists several devices, several of these devices may be embodied by the same hardware item. The use of the words first, second, and third, etc. does not indicate any order. These words may be interpreted as names.

Claims

1. A video generation method, comprising: Get the target story text; Performing content splitting processing on the target story text to obtain scene information, character information and subtitle information, wherein there is a correlation between the character information and the subtitle information; Determine, according to the character information, a character sticker corresponding to the subtitle information associated with the character information, and determine a scene material corresponding to the scene information; A story video is generated according to the subtitle information, the character stickers, and the scene materials.

2. The method according to claim 1, wherein: The character information includes emotion information; The step of determining, according to the character information, the character sticker corresponding to the subtitle information associated with the character information further comprises: Searching an emotional character library according to the emotional information to obtain character stickers matching the emotional information; The retrieved character sticker is determined as the character sticker corresponding to the subtitle information that is associated with the emotion information.

3. The method according to claim 1 or 2, wherein: The role information includes: action information; The step of determining, according to the character information, the character sticker corresponding to the subtitle information associated with the character information further comprises: Querying an action role library according to the action information to obtain a role sticker matching the action information; The retrieved character sticker is determined as the character sticker corresponding to the subtitle information associated with the action information.

4. The method according to any one of claims 1 to 3, wherein: The obtaining of the target story text further comprises: Identify pending story proposals; By using the AI ​​model, the to-be-processed story proposal is processed into a target story text in a preset format according to the preset first text prompt information.

5. The method according to claim 4, wherein: The step of processing the to-be-processed story proposal into a target story text in a preset format according to the preset first text prompt information by using the AI ​​model further includes: Using the AI ​​model, identifying keywords in the story proposal to be processed, obtaining meme elements matching the keywords, and generating a meme story text according to the meme elements and the story proposal to be processed; The meme story text is rewritten into a target story text in a preset format according to the preset first text prompt information.

6. The method according to claim 5, wherein: The method further comprises: identifying sensitive words on the meme story text; If there are sensitive words in the meme story text, a reminder message will be displayed in a preset manner to remind the user to make modifications.

7. The method according to any one of claims 1 to 6, wherein: The generating of the story video according to the subtitle information, the character stickers, and the scene material further comprises: The subtitle information, the character stickers, and the scene materials are laid out on a timeline, and rendered to generate a story video.

8. The method according to claim 7, wherein: The timeline includes: a video track, an audio track, a subtitle track, and a sticker track; the character stickers include: sticker images and sticker sound effects; The step of laying the subtitle information, the character stickers, and the scene material on the timeline further includes: Create video track, audio track, subtitle track, and sticker track in the video generation page; Add the subtitle information to the subtitle track, add the sticker sound effect to the audio track, add the sticker image to the sticker track, and add the scene material to the video track; Record the mapping relationship between character stickers and subtitle information, set the display duration of subtitle information, sticker image display duration and sticker sound effect duration to be the same, set the start display time of sticker image and the start playback time of sticker sound effect according to the subtitle information display timestamp, and set the coordinate information corresponding to the subtitle information and sticker image.

9. The method according to claim 7 or 8, wherein: The method further comprises: receiving a sticker batch editing request triggered by a user, wherein the sticker batch editing request carries an identifier of a character sticker to be edited; In response to a sticker batch editing request, subtitle information and character stickers associated with the character sticker identifier to be edited are displayed on the sticker editing page to enable the user to perform batch editing.

10. A video generating device, comprising: An acquisition module, adapted to acquire a target story text; A content splitting processing module is adapted to perform content splitting processing on the target story text to obtain scene information, character information and subtitle information, wherein there is an association relationship between the character information and the subtitle information; A determination module, adapted to determine, according to the character information, a character sticker corresponding to the subtitle information associated with the character information, and to determine a scene material corresponding to the scene information; A generation module is suitable for generating a story video based on the subtitle information, the character stickers, and the scene materials.

11. A computing device comprising: A processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus; The memory is used to store at least one executable instruction, and the executable instruction enables the processor to perform operations corresponding to the video generation method according to any one of claims 1-9.

12. A computer storage medium, wherein at least one executable instruction is stored in the storage medium, and wherein the executable instruction enables a processor to execute an operation corresponding to the video generation method according to any one of claims 1 to 9.

13. A computer program product, comprising at least one executable instruction, wherein the executable instruction enables a processor to execute operations corresponding to the video generation method according to any one of claims 1 to 9.

Images