Text content interaction method and device, equipment and storage medium

Abstract

The invention provides a text content interaction method and device, equipment and a storage medium, and relates to the field of artificial intelligence, in particular to the technical fields of image processing, data processing, large models, computer communication and the like. The text content interaction method comprises the steps of obtaining a target chapter to which currently displayed text content belongs; determining a target paragraph of the target chapter; when the target paragraph is displayed, an illustration is intelligently generated at the target paragraph based on related information of the target paragraph, and the related information at least comprises a target plot in the target paragraph, feature description of a leading agent in the target paragraph, atmosphere and emotion description in the target paragraph and scene description in the target paragraph. According to the method, the illustration conforming to the drama of the target paragraph can be generated, the illustration and the text content of the target paragraph supplement each other, and the reading experience of a user can be effectively improved.

Description

Technical Field

[0001] The present disclosure relates to the technical fields of image processing, data processing, large models and computer communications in the field of artificial intelligence, and in particular to a text content interaction method, device, equipment and storage medium. Background Art

[0002] With the development of technology, electronic devices are becoming more and more popular, and more and more users are accustomed to reading e-books through electronic devices. E-books, with their convenient reading method, have greatly increased users' reading frequency and reading time.

[0003] Currently, e-book content, such as novels, can provide users with a visual reading method, but the user reading experience is relatively simple. Therefore, a new text content interaction solution is urgently needed. Summary of the invention

[0004] The present disclosure provides a text content interaction method, apparatus, device and storage medium for improving user reading experience.

[0005] According to a first aspect of the present disclosure, a text content interaction method is provided, comprising:

[0006] Get the target chapter to which the currently displayed text content belongs;

[0007] Identify target paragraphs for target chapters;

[0008] When displaying a target paragraph, an illustration is intelligently generated at the target paragraph based on relevant information of the target paragraph, and the relevant information includes at least a target plot in the target paragraph, a characteristic description of a protagonist intelligent entity in the target paragraph, an atmosphere and emotional description in the target paragraph, and a scene description in the target paragraph.

[0009] According to a second aspect of the present disclosure, a text content interaction device is provided, comprising:

[0010] An acquisition unit, used to acquire the target chapter to which the currently displayed text content belongs;

[0011] Determination unit, used to determine the target paragraph of the target chapter;

[0012] A generating unit is used to intelligently generate illustrations at the target paragraph based on relevant information of the target paragraph when displaying the target paragraph, wherein the relevant information at least includes a target plot in the target paragraph, a feature description of a protagonist intelligent entity in the target paragraph, an atmosphere and emotion description in the target paragraph, and a scene description in the target paragraph.

[0013] According to a third aspect of the present disclosure, an electronic device is provided, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the text content interaction method described in the first aspect.

[0014] According to a fourth aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable the computer to execute the text content interaction method described in the first aspect.

[0015] According to a fifth aspect of the present disclosure, a computer program product is provided, comprising: a computer program, wherein the computer program is stored in a readable storage medium, and at least one processor of an electronic device can read the computer program from the readable storage medium, and the at least one processor executes the computer program so that the electronic device executes the text content interaction method described in the first aspect.

[0016] The technology disclosed in the present invention solves the problem that the user has a relatively single reading experience when reading the text content of an e-book. The present invention determines the target paragraph of the target chapter to which the currently displayed text content belongs; when displaying the target paragraph, an illustration is intelligently generated at the target paragraph based on the relevant information of the target paragraph, wherein the relevant information includes at least the target plot in the target paragraph, the characteristic description of the protagonist intelligent body in the target paragraph, the atmosphere and emotional description in the target paragraph, and the scene description in the target paragraph, which can ensure that the generated illustration conforms to the plot of the target paragraph and accurately reflects the core plot and atmosphere of the target paragraph; and the illustration is directly embedded in the target paragraph, which can make the illustration and the text content of the target paragraph complement each other, enhance the user's reading immersion, and better feel the plot of the target paragraph, thereby effectively improving the user's reading experience.

[0017] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure.

[0019] Figure 1 It is a schematic diagram of an application scenario to which the text content interaction method disclosed in the present invention is applicable;

[0020] Figure 2 is a schematic diagram according to a first embodiment of the present disclosure;

[0021] Figure 3 is a schematic diagram according to a second embodiment of the present disclosure;

[0022] Figure 4 is a schematic diagram according to a third embodiment of the present disclosure;

[0023] Figure 5 is a schematic diagram according to a fourth embodiment of the present disclosure;

[0024] Figure 6 is a schematic diagram according to a fifth embodiment of the present disclosure;

[0025] Figure 7 is a schematic diagram according to a sixth embodiment of the present disclosure;

[0026] Figure 8 FIG. 8 is a schematic block diagram of an example electronic device 800 that may be used to implement embodiments of the present disclosure. DETAILED DESCRIPTION

[0027] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0028] The present invention discloses a text content interaction method, device, equipment and storage medium, which are applied to technical fields such as action recognition, visual position positioning and image processing in the field of computer vision to achieve the purpose of more accurate text content interaction.

[0029] In the technical solution of the present disclosure, the collection, storage, use, processing, transmission, provision and disclosure of user personal information involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0030] Currently, e-book content, such as novels, can provide users with a visual reading method, but the user reading experience is relatively simple. Therefore, a new text content interaction solution is urgently needed.

[0031] In order to solve the above problems, the present disclosure provides a text content interaction method. For a target paragraph of a target chapter to which the currently displayed text content belongs, the target paragraph contains key elements such as strong emotional fluctuations, key event turning points, or conflicts and reconciliations between main characters. When the target paragraph is displayed, an illustration is intelligently generated at the target paragraph based on relevant information of the target paragraph. The content of the illustration is consistent with the plot of the target paragraph and can accurately reflect the core plot and atmosphere of the target paragraph. The illustration and the text content of the target paragraph complement each other, which can effectively enhance the user's reading experience.

[0032] In addition, the present disclosure also provides users with the ability to have conversations with main characters in the text content, to have group chats with main characters from different texts, and to upload custom voice packages to play novel content, thereby constructing a new form of multimodal artificial intelligence (AI) interactive novels, enhancing user reading immersion and improving user reading experience.

[0033] It can be seen that the text content interaction method provided by the present disclosure can be applied to e-books, specifically novels.

[0034] Figure 1 It is a schematic diagram of an application scenario applicable to the text content interaction method of the present disclosure. The application scenario may include: a server cluster 11 and a terminal 12. Among them, the server cluster 11 includes multiple servers 111 and a memory 112, and the terminal 12 can be a mobile phone, a tablet computer, a laptop computer, a desktop computer or a smart home appliance. Taking the user reading a novel through the terminal 12 as an example, when the user reads the target paragraph of a chapter of the novel, the server 111 intelligently generates an illustration for the target paragraph, and displays the illustration at the target paragraph through the terminal 12. Among them, the server 111 and the terminal 12 communicate through a wireless network or a wired network, and the memory 112 is used to store the data generated by the server 111 in the process of generating illustrations.

[0035] It should be noted that Figure 1 This is only a schematic diagram of an application scenario provided by the embodiment of the present disclosure. Figure 1 does not limit the equipment included in Figure 1 The positional relationship between the devices is limited.

[0036] The following specific embodiments are used to describe in detail the technical solution of the present invention and how the technical solution of the present invention solves the above-mentioned technical problems. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present invention will be described below in conjunction with the accompanying drawings.

[0037] Figure 2Schematic diagram of the first embodiment of the present disclosure. Figure 2 As shown, the text content interaction method provided by the first embodiment of the present disclosure includes:

[0038] S201, obtaining the target chapter to which the currently displayed text content belongs.

[0039] In the disclosed embodiment, the currently displayed text content belongs to the target text, and the target text is, for example, a novel. For example, taking the target text as a novel, the novel contains multiple chapters. When the user reads the text content of the novel, the target chapter to which the currently displayed text content belongs can be obtained according to the currently displayed text content.

[0040] S202. Determine the target paragraph of the target chapter.

[0041] In this step, for each paragraph included in the target chapter, the target paragraph of the target chapter can be determined based on whether the paragraph contains strong emotional fluctuations, key event turning points, or conflicts and reconciliations between main characters and other key elements. The target paragraph can also be understood as the climax paragraph of the target chapter. Exemplarily, the target paragraph of the target chapter can be automatically detected and determined by a large language model. For specific information on how to determine the target paragraph of the target chapter, please refer to the subsequent embodiments.

[0042] S203. When displaying the target paragraph, intelligently generate an illustration at the target paragraph based on relevant information of the target paragraph, wherein the relevant information at least includes a target plot in the target paragraph, a feature description of the protagonist intelligent entity in the target paragraph, an atmosphere and emotion description in the target paragraph, and a scene description in the target paragraph.

[0043] In this step, after determining the target paragraph of the target chapter, an illustration that matches the plot can be intelligently generated at the target paragraph based on the relevant information of the target paragraph when the target paragraph is displayed. The illustration is directly embedded in the target paragraph, such as embedding the illustration below the target paragraph. The illustration and the text content complement each other and can effectively improve the user's reading experience. Among them, for the relevant information of the target paragraph, the target plot in the target paragraph is the core basis for the creation of the illustration, including the development, turning point, conflict, etc. of the event. The target plot can also be understood as the core plot in the target paragraph; the behavior, expression, interaction, etc. of the protagonist intelligent body and its related characters are important elements in the illustration, which can enhance the sense of substitution and story of the illustration; the atmosphere and emotional color created by the target paragraph are also important references for the creation of illustrations, such as the expression of emotions such as tension, sadness, and joy; the description of the scene in the target paragraph, such as location, environment, time, etc., is also a factor that needs to be considered when creating illustrations to ensure the consistency of the illustration with the original content. In addition, generating illustrations for the target paragraph instead of generating illustrations for one or more other paragraphs of the target chapter can reduce the amount of data calculation while improving the user's reading experience.

[0044] For example, the large language model can be used to intelligently generate illustrations based on the relevant information of the target paragraph, wherein the overall consistency of the characters and painting style of the raw image is controlled by prompts and seeds. For details on how to intelligently generate illustrations based on the relevant information of the target paragraph using the large language model, please refer to the subsequent embodiments.

[0045] Optionally, the protagonist agent is generated in the following manner: a target text to which the text content belongs is input into a large language model, and a protagonist agent is generated based on the main characters in the target text.

[0046] For example, if the target text is a novel, the novel content can be input into the large language model, the novel content can be summarized and analyzed, the main characters in the novel can be automatically identified and extracted, and the corresponding protagonist agents can be generated. For example, a novel can generate at most n protagonist agents, for example, if n is 3, then a novel can generate at most 3 protagonist agents.

[0047] In the disclosed embodiment, a target paragraph of the target chapter is determined for the target chapter to which the currently displayed text content belongs; when the target paragraph is displayed, an illustration is intelligently generated at the target paragraph based on relevant information of the target paragraph, wherein the relevant information includes at least the target plot in the target paragraph, the characteristic description of the protagonist intelligent body in the target paragraph, the atmosphere and emotional description in the target paragraph, and the scene description in the target paragraph, so as to ensure that the generated illustration conforms to the plot of the target paragraph and accurately reflects the core plot and atmosphere of the target paragraph; and the illustration is directly embedded in the target paragraph, so that the illustration and the text content of the target paragraph complement each other, enhance the user's reading immersion, and better feel the plot of the target paragraph, thereby effectively improving the user's reading experience.

[0048] Figure 3 FIG. 2 is a schematic diagram of a second embodiment of the present disclosure. Based on the above embodiment, the present disclosure further describes the text content interaction method. Figure 3 As shown, the text content interaction method provided by the second embodiment of the present disclosure may include:

[0049] S301, obtaining the target chapter to which the currently displayed text content belongs.

[0050] The implementation principle and technical effects of S301 may refer to the aforementioned embodiments and will not be described in detail.

[0051] S302: Determine the target paragraph of the target chapter through the large language model.

[0052] It can be understood that the large language model has the capabilities of natural language processing, computer vision, recommendation, knowledge graph, automated machine learning and multimodal learning, and different functional modules can be developed in the form of plug-ins, and each functional module can be connected to different application systems. The large language model in the embodiment of the present disclosure can be trained based on the current large language model of long text content. In this step, the target paragraph of the target chapter can be determined by the large language model.

[0053] Further, optionally, the target paragraph is determined by: inputting the target text and the second prompt word into the large language model to obtain the target paragraph of the target chapter output by the large language model, and the second prompt word is used to indicate the target element of the target paragraph.

[0054] Exemplarily, the content of the second prompt word is, for example, "Please identify and extract a climax paragraph from each chapter; the climax paragraph should contain strong emotional fluctuations, key event turning points, or conflicts and reconciliations between main characters, etc.", wherein strong emotional fluctuations, key event turning points, or conflicts and reconciliations between main characters, etc. are the target elements of the climax paragraph, and the target elements can also be understood as the key elements of the climax paragraph. The second prompt word is used to guide the large language model to accurately identify and locate the target paragraph in each chapter (including the target chapter). The target text and the second prompt word can be input into the large language model to obtain the target paragraph of the target chapter output by the large language model.

[0055] S303. When displaying the target paragraph, at the target paragraph, an illustration is intelligently generated based on the relevant information of the target paragraph through the large language model. The relevant information of the target paragraph includes at least the target plot in the target paragraph, the characteristic description of the protagonist intelligent body in the target paragraph, the atmosphere and emotion description in the target paragraph, and the scene description in the target paragraph; the content of the illustration includes at least the scene corresponding to the target paragraph, the state of the protagonist intelligent body, the target plot elements in the target paragraph, and the atmosphere and emotion contained in the target paragraph.

[0056] In this step, illustrations can be intelligently generated based on the relevant information of the target paragraph through the large language model. The content of the illustrations includes at least the following information: depicting the specific scene in which the target paragraph takes place, including background information such as location, environment, and time, to create an atmosphere consistent with the original text; the protagonist intelligent agent and its related characters are the core elements of the illustration, and the illustrations include their images, actions, expressions, etc., showing their status and interactions in the target paragraph; the target plot elements in the target paragraph are integrated into the illustration, such as props, actions, dialogues, etc., to intuitively convey the core plot of the original text, and the target plot elements can also be understood as core plot elements; through color, light and shadow, composition and other techniques, the illustrations convey the emotional color and atmosphere contained in the target paragraph, such as tension, sadness, joy, etc., thereby enhancing the reader's reading experience.

[0057] Further, optionally, intelligently generating illustrations based on relevant information of the target paragraph through a large language model may include: inputting the target text, a first prompt word and a seed value to which the text content belongs into the large language model, and generating illustrations based on the relevant information of the target paragraph; wherein the first prompt word and the seed value are used to control the overall consistency of the characters and the painting style of the raw picture; the first prompt word includes at least a feature description of the protagonist intelligent entity, a description of the scene and atmosphere, and requirements for the painting style; the seed value is a first fixed seed value corresponding to the target text and / or a second fixed seed value corresponding to the protagonist intelligent entity in the target text.

[0058] Exemplarily, when the large language model is used to intelligently generate illustrations based on the relevant information of the target paragraph, the first prompt word is the key to guiding the large language model to generate illustrations. In order to control the overall consistency of the characters and painting style of the raw picture, the first prompt word includes at least the following key information: the description of the characteristics of the protagonist intelligent body, such as the gender, age, hairstyle, clothing, expression, etc. of the protagonist intelligent body, needs to be described in sufficient detail so that the large language model can accurately capture and generate a character image that meets the description; the description of the scene and atmosphere, such as "in front of the ancient castle, a princess in a gorgeous long dress smiles and looks into the distance", such a description helps the large language model to generate a background and atmosphere that matches the plot of the target text (such as a novel); for the style requirements, the required style needs to be clearly specified, such as "watercolor style", "oil painting style" or "anime style", etc., which helps to ensure that the generated illustrations are consistent in style. Further, according to the content and style requirements of the target text, a set of standard prompt templates are constructed, which include detailed character descriptions, scene and atmosphere descriptions, and style requirements.

[0059] When intelligently generating illustrations based on relevant information of the target paragraph through a large language model, the seed value is an important parameter that determines the random starting point of the generated illustrations. Taking the target text as a novel as an example, in order to control the overall consistency of the characters and style of the raw pictures, the following strategy can be adopted to set the seed value: assign a corresponding first fixed seed value to each novel and / or assign a corresponding second fixed seed value to each protagonist agent in the novel, and use the above fixed seed value when generating all illustrations of the novel or the protagonist agent. In this way, no matter how many times it is generated, as long as the first prompt word and the seed value remain unchanged, the generated illustrations will be consistent in character image and style. Furthermore, a seed value management module can be established in the system to record and manage the seed value of each novel or each protagonist agent. When the user needs to generate a new illustration, the system can automatically retrieve and apply the corresponding seed value.

[0060] In this embodiment, after the first prompt word and the seed value are determined, the target text to which the text content belongs, the first prompt word and the seed value may be input into a large language model, thereby generating an illustration based on relevant information of the target paragraph.

[0061] In the disclosed embodiment, a target paragraph of a target chapter to which the currently displayed text content belongs is determined through a large language model. When the target paragraph is displayed, an illustration is intelligently generated at the target paragraph based on relevant information of the target paragraph through the large language model. On the basis of improving the efficiency of illustration generation and reducing costs, the generated illustration can better fit the plot of the target paragraph, so that the illustration and the text content of the target paragraph complement each other, enhance the user's reading immersion, and thus effectively improve the user's reading experience.

[0062] On the basis of the above embodiment, optionally, the prompt words of the large language model also include a third prompt word for performing risk control processing on the illustration. The text content interaction method provided by the embodiment of the present disclosure may also include: when generating the illustration through the large language model, performing risk control processing on the content of the illustration according to the third prompt word; and / or, after generating the illustration through the large language model, performing risk control processing on the content of the illustration according to the third prompt word; if the processing result of the risk control processing is that sensitive content exists, triggering the large language model to regenerate the illustration.

[0063] Exemplarily, in order to ensure that the generated illustration content complies with laws, regulations and ethical standards, the large language model can be used to perform pre- and post-risk control on the illustration. Among them, the third prompt word is a prompt word used to perform risk control on the illustration. The content of the third prompt word is, for example, "Please ensure the content of the illustration: healthy and positive, strictly prohibit vulgar, indecent or illegal image elements; respect the original plot, and do not distort or mislead readers." In this embodiment, specifically, when generating illustrations, the large language model can be used to perform risk control on the content of the illustration according to the third prompt word to achieve pre-risk control. After generating the illustration, the large language model can be used to perform risk control on the content of the illustration according to the third prompt word to achieve post-risk control. When performing risk control, check whether the above-mentioned third prompt word is hit. If the above-mentioned third prompt word is hit, that is, the result of the risk control processing is that sensitive content exists, then the large language model is triggered to automatically retry, that is, regenerate the illustration. If it still returns a failure after retrying more than, for example, 2 times, the user is prompted and the reason for the failure is recorded.

[0064] Figure 4 is a schematic diagram according to the third embodiment of the present disclosure. Figure 4 As shown, considering that the user can have a dialogue with the protagonist agent contained in the text content to further enhance the user's reading experience, the text content interaction method provided in the third embodiment of the present disclosure may include:

[0065] S401. Highlight the identifier of the protagonist intelligent agent contained in the text content, and display a virtual control corresponding to the identifier of the protagonist intelligent agent, where the virtual control is used to enter a dialogue interface between the user and the protagonist intelligent agent.

[0066] Exemplarily, the identifier of the protagonist intelligent agent is, for example, the name of the protagonist intelligent agent, and the virtual control is, for example, a bubble prompt button. In the case where the text content contains the protagonist intelligent agent, the identifier of the protagonist intelligent agent can be highlighted, and the virtual control corresponding to the identifier of the protagonist intelligent agent can be displayed, such as displaying the virtual control in the upper right corner of the identifier of the protagonist intelligent agent. Specifically, for example, after page 5 of the novel, when the name of a protagonist intelligent agent appears for the first time, the name of the protagonist intelligent agent can be highlighted in the text content, and a bubble prompt button can be displayed in the upper right corner of the name of the protagonist intelligent agent to guide the user to have a conversation. The user can click the bubble prompt button to enter the conversation interface between the user and the protagonist intelligent agent.

[0067] Optionally, the logo of the protagonist intelligent agent that has appeared can be permanently displayed in the text content, and the user can click the virtual control corresponding to the logo of the protagonist intelligent agent at any time to communicate with the protagonist intelligent agent.

[0068] S402. In response to receiving an operation instruction for a virtual control, a dialogue interface between the user and the protagonist agent is displayed.

[0069] Exemplarily, a user may click on a virtual control corresponding to the identifier of the protagonist agent, and accordingly, the electronic device executing the embodiment of the method displays a dialogue interface between the user and the protagonist agent in response to receiving an operation instruction for the virtual control.

[0070] S403. In response to the first dialogue content input by the user in the dialogue interface, generate and display the second dialogue content of the target protagonist agent, wherein the second dialogue content is generated by a large language model based on the content of the target text and the speaking style of the protagonist agent.

[0071] In this step, the second dialogue content is generated by the large language model based on the content of the target text and the speaking style of the protagonist agent, which can better communicate with the user and improve the quality of the dialogue. Exemplarily, the user inputs the first dialogue content in the dialogue interface, and accordingly, the electronic device executing the embodiment of the method generates and displays the second dialogue content of the target protagonist agent in response to the first dialogue content input by the user in the dialogue interface.

[0072] Optionally, the contextual history of the conversation can be retained for each protagonist agent to conduct a coherent conversation.

[0073] Based on the above embodiments, Figure 5 is a schematic diagram according to the fourth embodiment of the present disclosure. Figure 5As shown, taking the target text as a novel as an example, the text content of the currently displayed novel contains the name of a protagonist agent, such as "name 1"; a bubble prompt button 501 is displayed in the upper right corner of "name 1", and the left side of the bubble prompt button 501 is the virtual avatar of the protagonist agent. The user can click the bubble prompt button 501 to enter the dialogue interface 502 between the user and the protagonist agent. The dialogue interface 502 is, for example, a semi-floating panel that is pulled up or emerged while the text content is kept displayed. In the dialogue interface 502, the interface background is, for example, the virtual portrait of the protagonist agent; 503 represents a return arrow, which is used to return to the interface corresponding to the text content; area 504 is a voice playback area, which is used to play the audio generated by the protagonist agent based on the text content; area 505 is used to display the introduction of the protagonist agent; area 506 is used to display the story scene generated by the large language model; 507 is a "expand and talk" virtual control, which is used to respond to the user's click operation and continue to generate the story through the large language model; 508 is used for the user to input the first dialogue content.

[0074] In the disclosed embodiment, by highlighting the identifier of the protagonist intelligent agent contained in the text content and displaying the virtual control corresponding to the identifier of the protagonist intelligent agent, it is convenient for the user to enter the dialogue interface with the protagonist intelligent agent and then have a dialogue with the protagonist intelligent agent, wherein the second dialogue content of the protagonist intelligent agent is generated by a large language model based on the content of the target text and the speaking style of the protagonist intelligent agent, which can better conduct dialogues with users and improve the quality of dialogues, thereby enhancing user participation and stickiness.

[0075] Figure 6 is a schematic diagram according to the fifth embodiment of the present disclosure. Figure 6 As shown, considering that a user can have a group chat with multiple protagonist agents from different texts to further enhance the user's reading experience, the text content interaction method provided in the fifth embodiment of the present disclosure may include:

[0076] S601. In response to a group creation instruction for multiple protagonist agents from different texts, a group member selection area is displayed.

[0077] For example, a user can select multiple protagonist agents from different texts (such as different novels) in the agent management page to create a group for group chat. Accordingly, the electronic device executing the embodiment of the method displays a group member selection area in response to a group creation instruction for multiple protagonist agents from different texts.

[0078] S602. In response to the selection operation of multiple protagonist intelligent agents in the group member selection area, a group chat detail interface corresponding to the multiple protagonist intelligent agents is displayed, and a story scene is displayed in the conversation display area of ​​the group chat detail interface, as well as a third conversation content of the user based on the story scene and a fourth conversation content of the multiple protagonist intelligent agents. The story scene is randomly generated by the large language model.

[0079] In this step, the user can select multiple protagonist agents from different texts to join the group in the group member selection area. Accordingly, the electronic device executing the embodiment of the present method displays the group chat details interface corresponding to the multiple protagonist agents in response to the selection operation of the multiple protagonist agents in the group member selection area, and displays the story scene in the dialogue display area of ​​the group chat details interface. Among them, the story scene is randomly generated by the large language model, and the user can have a dialogue with each protagonist agent under the story scene. In the group chat details interface, the group chat name, the names and virtual avatars of different protagonist agents, and the name and virtual avatar of the user are displayed. The third dialogue content of the user and the fourth dialogue content of different protagonist agents among the multiple protagonist agents are also displayed in the group chat details interface.

[0080] Optionally, in response to a generation instruction to continue generating stories based on story scenes, the stories are continued to be generated using the large language model.

[0081] For example, in the group chat details interface, a virtual control of "expand chat" is displayed, and the user can click on the virtual control to continue generating the story through the large language model.

[0082] Optionally, determine the preset time point of the new day; display the fifth conversation content of multiple protagonist intelligent agents in the group chat details interface, and push a message to the user, the message is used to prompt the user to check the group chat information.

[0083] Exemplarily, when a preset time point of a new day is reached, multiple protagonist intelligent agents in a group chat can automatically conduct a conversation, the fifth conversation content of the multiple protagonist intelligent agents can be displayed on the group chat details interface, and a message can be pushed to the user to prompt the user to check the group chat information, so as to facilitate more interaction with the user.

[0084] Optionally, the user can exit the group chat and delete the group chat and the corresponding historical conversations.

[0085] In the disclosed embodiment, in response to a group creation instruction for multiple protagonist agents from different texts, a group member selection area is displayed; in response to a selection operation on multiple protagonist agents in the group member selection area, a group chat details interface corresponding to the multiple protagonist agents is displayed, and a story scene is displayed in the conversation display area of ​​the group chat details interface, as well as a third conversation content of the user based on the story scene and a fourth conversation content of the multiple protagonist agents. The story scene is randomly generated by a large language model, which enables users to conveniently conduct group chats with multiple protagonist agents from different texts, thereby further enhancing the user's reading experience.

[0086] On the basis of the above embodiment, optionally, the prompt word of the large language model also includes a fourth prompt word for performing risk control processing on the conversation content. The text content interaction method provided by the embodiment of the present disclosure may also include: when generating the target conversation content through the large language model, performing risk control processing on the target conversation content according to the fourth prompt word, the target conversation content includes the second conversation content and / or the fourth conversation content; and / or, after generating the target conversation content through the large language model, performing risk control processing on the generated target conversation content according to the fourth prompt word; if the processing result of the risk control processing is that sensitive content exists, triggering the large language model to regenerate the target conversation content.

[0087] Exemplarily, in order to ensure that the generated conversation content complies with laws, regulations and ethical standards, the second conversation content and the fourth conversation content can be pre- and post-risk controlled through a large language model. Among them, the content of the fourth prompt word is, for example, "When generating a conversation, please ensure that the conversation content: avoids descriptions of illegal, erroneous, and violent behavior; strictly prohibits immoral, vulgar, and indecent language; remains neutral, avoids sensitive topics or controversial content; respects multiculturalism, eliminates discrimination and prejudice; does not embed any form of advertising or spam." Specifically, when generating the target conversation content through a large language model, the target conversation content can be risk-controlled according to the fourth prompt word to achieve pre-risk control. After the target conversation content is generated by the large language model, risk control processing can be performed on the generated target conversation content according to the fourth prompt word to implement post-risk control. When performing risk control processing, it is checked whether the above-mentioned fourth prompt word is hit. If the above-mentioned fourth prompt word is hit, that is, the processing result of the risk control processing is that sensitive content exists, then the large language model is triggered to automatically retry, that is, the target conversation content is regenerated. If failure is still returned after retrying more than, for example, 2 times, the user is prompted and the reason for the failure is recorded.

[0088] Based on the above embodiment, optionally, the complete prompt words input into the large language model may include prompt words for generating the protagonist intelligent agent, a first prompt word, a second prompt word, a third prompt word and a fourth prompt word, which can ensure that the content generated by the large language model complies with laws, regulations and ethical standards.

[0089] On the basis of the above-mentioned embodiments, optionally, the text content interaction method provided by the embodiments of the present disclosure may further include: receiving a custom voice package uploaded by a user in response to a voice upload instruction for a target text to which the text content belongs; and using the audio in the custom voice package to play the text content.

[0090] For example, if the target text is a novel, the function of users uploading custom voice packages can be added to the "Listen to Novels" function of the novel. Users can select their own custom voice packages and upload them to the system as personalized voice packages for the "Listen to Novels" function. Users can select their own uploaded custom voice packages to listen to novels. For example, if the custom voice package uploaded by the user is the voice of the user's grandmother, the user can experience the feeling of the grandmother telling him or her a story, thereby bringing a more vivid and intelligent reading experience to the user.

[0091] Based on the above embodiments, the text content interaction method provided by the embodiments of the present disclosure can be applied to novels. Specifically, based on the "novel" vertical category in the novel application, through the understanding and extraction of novels by a large language model, it can provide users with the capabilities of "one-click creation of novel character intelligent entities", "character intelligent entity group chat", "intelligent illustration generation", and uploading custom voice packages to play novel content, etc., to build a new form of multimodal AI interactive novels, enhance user reading immersion, meet user needs for dialogue with fan characters, and increase the number of daily active users (Daily Active Users, DAU) and the number of times a novel product stays, creating new paid scenarios for novels.

[0092] The following are embodiments of the device disclosed herein, which can be used to execute the method embodiments disclosed herein. For details not disclosed in the device embodiments disclosed herein, please refer to the method embodiments disclosed herein.

[0093] Figure 7 is a schematic diagram according to the sixth embodiment of the present disclosure. Figure 7 As shown, the text content interaction device 700 provided by the sixth embodiment of the present disclosure includes an acquisition unit 701, a determination unit 702 and a generation unit 703. Among them:

[0094] The acquisition unit 701 is used to acquire the target chapter to which the currently displayed text content belongs.

[0095] The determination unit 702 is used to determine a target paragraph of a target chapter.

[0096] The generation unit 703 is used to intelligently generate an illustration at the target paragraph based on relevant information of the target paragraph when displaying the target paragraph, and the relevant information at least includes the target plot in the target paragraph, the characteristic description of the protagonist intelligent body in the target paragraph, the atmosphere and emotion description in the target paragraph, and the scene description in the target paragraph.

[0097] In some embodiments, the generation unit 703 includes: a first generation module (not shown in the figure), which is used to intelligently generate illustrations at the target paragraph based on relevant information of the target paragraph through a large language model, and the content of the illustrations includes at least the scene corresponding to the target paragraph, the status of the protagonist intelligent agent, the target plot elements in the target paragraph, and the atmosphere and emotions contained in the target paragraph.

[0098] In some embodiments, the prompt words of the large language model include a first prompt word, and the first generation module includes: a generation submodule (not shown in the figure), which is used to input the target text, the first prompt word and the seed value to which the text content belongs into the large language model, and generate illustrations based on the relevant information of the target paragraph; wherein the first prompt word and the seed value are used to control the overall consistency of the characters and the painting style of the raw picture; the first prompt word at least includes a feature description of the protagonist intelligent entity, a description of the scene and atmosphere, and a painting style requirement; the seed value is a first fixed seed value corresponding to the target text and / or a second fixed seed value corresponding to the protagonist intelligent entity in the target text.

[0099] In some embodiments, the prompt word of the large language model also includes a second prompt word, and the determination unit 702 includes a determination module (not shown in the figure) for determining the target paragraph in the following manner, and the determination module is used to: input the target text and the second prompt word into the large language model to obtain the target paragraph of the target chapter output by the large language model, and the second prompt word is used to indicate the target element of the target paragraph.

[0100] In some embodiments, the prompt words of the large language model also include a third prompt word for performing risk control processing on the illustration, and the text content interaction device 700 also includes a first risk control unit (not shown in the figure), which is used to: when generating the illustration through the large language model, perform risk control processing on the content of the illustration according to the third prompt word; and / or, after generating the illustration through the large language model, perform risk control processing on the content of the illustration according to the third prompt word; if the processing result of the risk control processing is that sensitive content exists, the large language model is triggered to regenerate the illustration.

[0101] In some embodiments, the generation unit 703 also includes a second generation module (not shown in the figure) for generating a protagonist intelligent agent in the following manner, and the second generation module is used to: input the target text to which the text content belongs into the large language model, and generate a protagonist intelligent agent based on the main characters in the target text.

[0102] In some embodiments, the text content includes an identifier of a protagonist intelligent agent, and the text content interaction device 700 also includes a dialogue unit (not shown in the figure), which is used to: highlight the identifier and display a virtual control corresponding to the identifier, the virtual control is used to enter a dialogue interface between the user and the protagonist intelligent agent; in response to receiving an operation instruction for the virtual control, display the dialogue interface between the user and the protagonist intelligent agent; in response to a first dialogue content input by the user in the dialogue interface, generate and display a second dialogue content of the target protagonist intelligent agent, the second dialogue content is generated by a large language model based on the content of the target text and the speaking style of the protagonist intelligent agent.

[0103] In some embodiments, the text content interaction device 700 also includes a group chat unit (not shown in the figure), which is used to: display a group member selection area in response to a group creation instruction for multiple protagonist agents from different texts; display a group chat details interface corresponding to the multiple protagonist agents in response to a selection operation on multiple protagonist agents in the group member selection area, and display a story scene in the dialogue display area of ​​the group chat details interface, as well as display a third dialogue content of the user based on the story scene and a fourth dialogue content of the multiple protagonist agents, wherein the story scene is randomly generated by the large language model.

[0104] In some embodiments, the prompt words of the large language model also include a fourth prompt word for performing risk control processing on the conversation content, and the text content interaction device 700 also includes a second risk control unit (not shown in the figure), which is used to: when generating the target conversation content through the large language model, perform risk control processing on the target conversation content according to the fourth prompt word, and the target conversation content includes the second conversation content and / or the fourth conversation content; and / or, after generating the target conversation content through the large language model, perform risk control processing on the generated target conversation content according to the fourth prompt word; if the processing result of the risk control processing is that sensitive content exists, the large language model is triggered to regenerate the target conversation content.

[0105] In some embodiments, the group chat unit is further configured to: in response to a generation instruction to continue generating a story based on a story scene, continue generating the story using the large language model.

[0106] In some embodiments, the group chat unit is also used to: determine the preset time point when a new day arrives; display the fifth conversation content of multiple protagonist intelligent agents in the group chat details interface, and push messages to users, where the messages are used to prompt users to check group chat information.

[0107] In some embodiments, the text content interaction device 700 also includes a voice processing unit (not shown in the figure), which is used to: receive a custom voice package uploaded by a user in response to a voice upload instruction for a target text to which the text content belongs; and use the audio in the custom voice package to play the text content.

[0108] Figure 7 The provided text content interaction device can execute the steps in the method embodiment corresponding to the above-mentioned text content interaction method. Its implementation principle and technical effects are similar and will not be repeated here.

[0109] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the solution provided by any of the above embodiments.

[0110] According to an embodiment of the present disclosure, the present disclosure further provides a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to enable a computer to execute a solution provided by any of the above embodiments.

[0111] According to an embodiment of the present disclosure, the present disclosure also provides a computer program product, which includes: a computer program, the computer program is stored in a readable storage medium, at least one processor of an electronic device can read the computer program from the readable storage medium, and at least one processor executes the computer program so that the electronic device executes the solution provided by any of the above embodiments.

[0112] Figure 8 8 is a schematic block diagram of an example electronic device 800 that can be used to implement an embodiment of the present disclosure. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0113] like Figure 8 As shown, the electronic device 800 includes a computing unit 801, which can store data stored in a read-only memory (ROM) ( Figure 8 A computer program in ROM 802 is loaded from storage unit 808 to random access memory (RAM) ( Figure 8The computer programs in the RAM 803 are used to perform various appropriate actions and processes. The RAM 803 may also store various programs and data required for the operation of the electronic device 800. The computing unit 801, the ROM 802, and the RAM 803 are connected to each other via a bus 804. The input / output (I / O) interface ( Figure 8 The I / O interface 805 is also connected to the bus 804 .

[0114] Multiple components in the electronic device 800 are connected to the I / O interface 805, including: an input unit 806, such as a keyboard, a mouse, etc.; an output unit 807, such as various types of displays, speakers, etc.; a storage unit 808, such as a disk, an optical disk, etc.; and a communication unit 809, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 809 allows the electronic device 800 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0115] The computing unit 801 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 801 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence computing chips, various computing units running machine learning model algorithms, digital signal processors (DSP), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 801 performs the various methods and processes described above, such as the text content interaction method. For example, in some embodiments, the text content interaction method may be implemented as a computer software program, which is tangibly included in a machine-readable medium, such as a storage unit 808. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 800 via the ROM 802 and / or the communication unit 809. When the computer program is loaded into the RAM 803 and executed by the computing unit 801, one or more steps of the text content interaction method described above may be executed. Alternatively, in other embodiments, the computing unit 801 may be configured to execute the text content interaction method in any other appropriate manner (eg, by means of firmware).

[0116] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard parts (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0117] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0118] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0119] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0120] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: Local Area Networks (LANs), Wide Area Networks (WANs), and the Internet.

[0121] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship between the client and the server is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services ("Virtual Private Server", or "VPS" for short). The server may also be a server of a distributed system, or a server combined with a blockchain.

[0122] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this document does not limit this.

[0123] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, fusions, sub-fusions and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present disclosure should be included in the protection scope of the present disclosure.

Claims

1. A text content interaction method, comprising: Get the target chapter to which the currently displayed text content belongs; Determine the target paragraph of the target chapter; When displaying the target paragraph, an illustration is intelligently generated at the target paragraph based on relevant information of the target paragraph, and the relevant information includes at least the target plot in the target paragraph, the characteristic description of the protagonist intelligent entity in the target paragraph, the atmosphere and emotion description in the target paragraph, and the scene description in the target paragraph.

2. The text content interaction method according to claim 1, wherein: The intelligently generating an illustration at the target paragraph based on relevant information of the target paragraph includes: At the target paragraph, an illustration is intelligently generated based on the relevant information of the target paragraph through a large language model, and the content of the illustration includes at least the scene corresponding to the target paragraph, the state of the protagonist intelligent agent, the target plot elements in the target paragraph, and the atmosphere and emotion contained in the target paragraph.

3. The text content interaction method according to claim 2, wherein: The prompt word of the large language model includes a first prompt word, and the intelligently generating an illustration based on the relevant information of the target paragraph by using the large language model includes: Inputting the target text to which the text content belongs, the first prompt word and the seed value into the large language model, and generating an illustration based on the relevant information of the target paragraph; Among them, the first prompt word and the seed value are used to control the overall consistency of the characters and painting style of the raw picture; the first prompt word at least includes a description of the characteristics of the protagonist intelligent entity, a description of the scene and atmosphere, and the painting style requirements; the seed value is a first fixed seed value corresponding to the target text and / or a second fixed seed value corresponding to the protagonist intelligent entity in the target text.

4. The text content interaction method according to claim 3, wherein: The prompt word of the large language model further includes a second prompt word, and the target paragraph is determined in the following manner: The target text and the second prompt word are input into the large language model to obtain a target paragraph of the target chapter output by the large language model, wherein the second prompt word is used to indicate a target element of the target paragraph.

5. The text content interaction method according to claim 3, wherein: The prompt words of the large language model further include a third prompt word for performing risk control processing on the illustration, and the method further includes: When generating the illustration through the large language model, performing risk control processing on the content of the illustration according to the third prompt word; And / or, after generating the illustration through the large language model, risk control processing is performed on the content of the illustration according to the third prompt word; if the risk control processing result is that sensitive content exists, the large language model is triggered to regenerate the illustration.

6. The text content interaction method according to any one of claims 2 to 5, wherein: The protagonist agent is generated in the following way: The target text to which the text content belongs is input into the large language model, and the protagonist agent is generated according to the main characters in the target text.

7. The text content interaction method according to claim 6, wherein: The text content includes the identification of the protagonist agent, and the method further includes: Highlighting the logo and displaying a virtual control corresponding to the logo, wherein the virtual control is used to enter a dialogue interface between the user and the protagonist agent; In response to receiving an operation instruction for the virtual control, displaying a dialogue interface between the user and the protagonist agent; In response to a first dialogue content input by a user in the dialogue interface, a second dialogue content of the target protagonist agent is generated and displayed, wherein the second dialogue content is generated by the large language model based on the content of the target text and the speaking style of the protagonist agent.

8. The text content interaction method according to claim 7, wherein: Also includes: In response to a group creation instruction for a plurality of protagonist agents from different texts, displaying a group member selection area; In response to a selection operation on the multiple protagonist intelligent agents in the group member selection area, a group chat details interface corresponding to the multiple protagonist intelligent agents is displayed, and a story scene is displayed in the conversation display area of ​​the group chat details interface, as well as a third conversation content of the user based on the story scene and a fourth conversation content of the multiple protagonist intelligent agents are displayed, wherein the story scene is randomly generated by the large language model.

9. The text content interaction method according to claim 8, wherein: The prompt words of the large language model further include a fourth prompt word for performing risk control processing on the conversation content, and the method further includes: When generating target dialogue content through the large language model, performing risk control processing on the target dialogue content according to the fourth prompt word, the target dialogue content includes the second dialogue content and / or the fourth dialogue content; And / or, after generating the target conversation content through the large language model, risk control processing is performed on the generated target conversation content according to the fourth prompt word; if the processing result of the risk control processing is that sensitive content exists, the large language model is triggered to regenerate the target conversation content.

10. The text content interaction method according to claim 8, wherein: Also includes: In response to a generation instruction to continue generating a story based on the story scene, the story is continued to be generated using the large language model.

11. The text content interaction method according to claim 8, wherein: Also includes: Determine the preset time point to arrive at a new day; The fifth conversation content of the multiple protagonist intelligent agents is displayed on the group chat details interface, and a message is pushed to the user, wherein the message is used to prompt the user to view the group chat information.

12. The text content interaction method according to any one of claims 1 to 5, wherein: Also includes: In response to a voice upload instruction for a target text to which the text content belongs, receiving a custom voice package uploaded by a user; The text content is played using the audio in the custom voice package.

13. A text content interaction device, comprising: An acquisition unit, used to acquire the target chapter to which the currently displayed text content belongs; A determination unit, used for determining a target paragraph of the target chapter; A generating unit is used to intelligently generate an illustration at the target paragraph based on relevant information of the target paragraph when displaying the target paragraph, wherein the relevant information at least includes a target plot in the target paragraph, a feature description of a protagonist intelligent entity in the target paragraph, an atmosphere and emotion description in the target paragraph, and a scene description in the target paragraph.

14. The text content interaction device according to claim 13, wherein: The generating unit comprises: The first generation module is used to intelligently generate an illustration at the target paragraph based on the relevant information of the target paragraph through a large language model, and the content of the illustration at least includes the scene corresponding to the target paragraph, the state of the protagonist intelligent agent, the target plot elements in the target paragraph, and the atmosphere and emotion contained in the target paragraph.

15. The text content interaction device according to claim 14, wherein: The prompt word of the large language model includes a first prompt word, and the first generation module includes: A generating submodule, configured to input the target text to which the text content belongs, the first prompt word and the seed value into the large language model, and generate an illustration based on the relevant information of the target paragraph; Among them, the first prompt word and the seed value are used to control the overall consistency of the characters and painting style of the raw picture; the first prompt word at least includes a description of the characteristics of the protagonist intelligent entity, a description of the scene and atmosphere, and the painting style requirements; the seed value is a first fixed seed value corresponding to the target text and / or a second fixed seed value corresponding to the protagonist intelligent entity in the target text.

16. The text content interaction device according to claim 15, wherein: The prompt word of the large language model further includes a second prompt word, and the determination unit includes a determination module for determining the target paragraph in the following manner, wherein the determination module is used to: The target text and the second prompt word are input into the large language model to obtain a target paragraph of the target chapter output by the large language model, wherein the second prompt word is used to indicate a target element of the target paragraph.

17. The text content interaction device according to claim 15, wherein: The prompt words of the large language model further include a third prompt word for performing risk control processing on the illustration, and the device further includes a first risk control unit for: When generating the illustration through the large language model, performing risk control processing on the content of the illustration according to the third prompt word; and / or, after generating the illustration through the large language model, performing risk control processing on the content of the illustration according to the third prompt word; If the risk control processing result is that sensitive content exists, the large language model is triggered to regenerate the illustration.

18. The text content interaction device according to any one of claims 14 to 17, wherein: The generating unit further comprises a second generating module for generating a protagonist agent in the following manner, wherein the second generating module is used to: The target text to which the text content belongs is input into the large language model, and the protagonist agent is generated according to the main characters in the target text.

19. The text content interaction device according to claim 18, wherein: The text content includes the identification of the protagonist intelligent agent, and the device also includes a dialogue unit for: Highlighting the logo and displaying a virtual control corresponding to the logo, wherein the virtual control is used to enter a dialogue interface between the user and the protagonist agent; In response to receiving an operation instruction for the virtual control, displaying a dialogue interface between the user and the protagonist agent; In response to a first dialogue content input by a user in the dialogue interface, a second dialogue content of the target protagonist agent is generated and displayed, wherein the second dialogue content is generated by the large language model based on the content of the target text and the speaking style of the protagonist agent.

20. The text content interaction device according to claim 19, wherein: The device also includes a group chat unit, which is used to: In response to a group creation instruction for a plurality of protagonist agents from different texts, displaying a group member selection area; In response to a selection operation on the multiple protagonist intelligent agents in the group member selection area, a group chat details interface corresponding to the multiple protagonist intelligent agents is displayed, and a story scene is displayed in the conversation display area of ​​the group chat details interface, as well as a third conversation content of the user based on the story scene and a fourth conversation content of the multiple protagonist intelligent agents are displayed, wherein the story scene is randomly generated by the large language model.

21. The text content interaction device according to claim 20, wherein: The prompt words of the large language model further include a fourth prompt word for performing risk control processing on the conversation content, and the device further includes a second risk control unit for: When generating target dialogue content through the large language model, performing risk control processing on the target dialogue content according to the fourth prompt word, the target dialogue content includes the second dialogue content and / or the fourth dialogue content; and / or, after generating the target dialogue content through the large language model, performing risk control processing on the generated target dialogue content according to the fourth prompt word; If the risk control processing result is that sensitive content exists, the large language model is triggered to regenerate the target conversation content.

22. The text content interaction device according to claim 20, wherein: The group chat unit is also used for: In response to a generation instruction to continue generating a story based on the story scene, the story is continued to be generated using the large language model.

23. The text content interaction device according to claim 20, wherein: The group chat unit is also used for: Determine the preset time point to arrive at a new day; The fifth conversation content of the multiple protagonist intelligent agents is displayed on the group chat details interface, and a message is pushed to the user, wherein the message is used to prompt the user to view the group chat information.

24. The text content interaction device according to any one of claims 13 to 17, wherein: The device also includes a speech processing unit, which is used to: In response to a voice upload instruction for a target text to which the text content belongs, receiving a custom voice package uploaded by a user; The text content is played using the audio in the custom voice package.

25. An electronic device comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the text content interaction method according to any one of claims 1 to 12.

26. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to enable the computer to execute the text content interaction method according to any one of claims 1 to 12.

27. A computer program product, comprising a computer program, which, when executed by a processor, implements the steps of the text content interaction method according to any one of claims 1 to 12.

Images