Powerpoint processing method and device, computer equipment, readable storage medium and program product

Abstract

The invention relates to a presentation processing method and device, computer equipment, a readable storage medium and a program product. The method comprises the following steps: acquiring an original presentation page, and inputting the original presentation page into a pre-trained large language model to obtain a presentation structured text of the original presentation page; identifying statistical data contained in the structured text of the PowerPoint, and generating a target statistical chart matched with the statistical data; inputting the PowerPoint element set into a pre-trained PowerPoint layout classification model to obtain a PowerPoint typesetting type matched with the PowerPoint element set; and according to the presentation file typesetting type, embedding the presentation file element set into a candidate presentation file page matched with the presentation file typesetting type to obtain an optimized presentation file page. By adopting the method, the generation effect of generating the presentation manuscript by artificial intelligence can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computers and artificial intelligence, and in particular to a presentation processing method and device, computer equipment, a computer readable storage medium, and a computer program product. BACKGROUND

[0002] A presentation is a document content in a visual form for presenting and explaining information.

[0003] With the continuous development of artificial intelligence technology, people have begun to use artificial intelligence technology in the production of presentations, thereby improving the work efficiency of people. However, in the traditional technology, the artificial intelligence technology only adds the text content input by the user to the selected presentation template in the process of producing the presentation, and the generated presentation often cannot meet the increasingly high presentation production requirements of the user, and there is a problem of poor generation effect of the presentation.

[0004] However, the current artificial intelligence generated presentation has the problem of poor generation effect. SUMMARY

[0005] Therefore, it is necessary to provide a presentation processing method and device, computer equipment, a computer readable storage medium, and a computer program product capable of improving the generation effect of the artificial intelligence generated presentation to solve the above technical problems.

[0006] In a first aspect, the present application provides a presentation processing method, which comprises:

[0007] obtaining an original presentation page, inputting the original presentation page into a pre-trained large language model to obtain a presentation structured text of the original presentation page;

[0008] identifying statistical data contained in the presentation structured text, generating a target statistical chart matched with the statistical data;

[0009] inputting a presentation element set into a pre-trained presentation layout classification model to obtain a presentation layout type matched with the presentation element set; the presentation element set at least includes the original presentation page, the presentation structured text, and the target statistical chart;

[0010] embedding the presentation element set into a candidate presentation page matched with the presentation layout type according to the presentation layout type to obtain an optimized presentation page.

[0011] In one embodiment, the original presentation page is a presentation page including at least a text box, and the obtaining of the original presentation page comprises:

[0012] In response to a presentation material uploading operation, obtaining a presentation material uploaded by the presentation material uploading operation as the original presentation page.

[0013] inputting the presentation material into a pre-trained large language model, performing text expansion on the presentation material by the pre-trained large language model, and obtaining a text expansion result;

[0014] generating the original presentation page according to the text expansion result; a text box of the original presentation page displays the text expansion result;

[0015] Alternatively,

[0016] In response to a presentation page uploading operation, obtaining a presentation page uploaded by the presentation page uploading operation as the original presentation page.

[0017] In one of the embodiments, the inputting the original presentation page into a pre-trained large language model to obtain a presentation structured text of the original presentation page comprises:

[0018] According to the original presentation page, constructing a text structured instruction for the pre-trained large language model; the text structured instruction is used to instruct the pre-trained large language model to extract text content in the original presentation page and convert the text content into a text in a preset text structured format;

[0019] inputting the text structured instruction into the pre-trained large language model to obtain a response content of the pre-trained large language model to the text structured instruction, the response content at least including the presentation structured text.

[0020] In one of the embodiments, the identifying the statistical data contained in the presentation structured text and generating a target statistical chart matched with the statistical data comprises:

[0021] According to the presentation structured text, constructing a statistical data processing instruction for the pre-trained large language model; the statistical data processing instruction is used to instruct the pre-trained large language model to extract statistical data from the presentation structured text and output structured statistical data;

[0022] inputting the statistical data processing instruction into the pre-trained large language model to obtain a response content of the pre-trained large language model to the statistical data processing instruction, the response content at least including the structured statistical data;

[0023] inputting the statistical data into a fine-tuned statistical chart type analysis model to obtain a target chart type matched with the statistical data, and mapping the statistical data to the target statistical chart by invoking a chart generation function matched with the target chart type.

[0024] In one of the embodiments, in the case where the material requirement corresponding to the presentation types includes presentation pictures, embedding the set of presentation elements into the candidate presentation page matched with the presentation types according to the presentation types to obtain the optimized presentation page includes:

[0025] generating a presentation picture matched with the content of the set of presentation elements by a pre-trained visual large model, and adding the presentation picture to the set of presentation elements to obtain an updated set of presentation elements;

[0026] embedding the updated set of presentation elements into the candidate presentation page according to the presentation types to obtain the optimized presentation page.

[0027] In one of the embodiments, the method further includes:

[0028] obtaining a local adjustment instruction input to the optimized presentation page, wherein the local adjustment instruction includes a natural language for adjusting at least part of the optimized presentation page;

[0029] inputting the local adjustment instruction into a pre-trained large language model to obtain an adjustment function for the optimized presentation page and an input parameter corresponding to the adjustment function;

[0030] performing a local adjustment operation on the optimized presentation page according to the adjustment function and the input parameter corresponding to the adjustment function.

[0031] In a second aspect, the present application further provides a presentation processing device, which includes:

[0032] a content obtaining module configured to obtain an original presentation page, and input the original presentation page into a pre-trained large language model to obtain a presentation structured text of the original presentation page;

[0033] a data identifying module configured to identify statistical data contained in the presentation structured text, and generate a target statistical chart matched with the statistical data;

[0034] The layout classification module is used to input the presentation element set into a pre-trained presentation layout classification model to obtain the presentation layout type that matches the presentation element set; the presentation element set includes at least the original presentation page, the presentation structured text, and the target statistical chart;

[0035] The element embedding module is used to embed the presentation element set into a candidate presentation page that matches the presentation layout type, according to the presentation layout type, to obtain an optimized presentation page.

[0036] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the above-described method.

[0037] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method.

[0038] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the above-described method.

[0039] The aforementioned presentation processing method, apparatus, computer equipment, computer-readable storage medium, and computer program product acquire the original presentation page and input it into a pre-trained large language model to obtain the structured text of the original presentation page. They then identify the statistical data contained in the structured text and generate a target statistical chart matching the statistical data. By inputting a set of presentation elements, including the original presentation page, the structured text, and the target statistical chart, into a pre-trained presentation layout classification model, they obtain a presentation layout type matching the set of presentation elements. Finally, they embed the set of presentation elements into a format matching the presentation layout type. The system selects candidate presentation pages that match the document layout type and then optimizes them to produce the final presentation page. This effectively leverages the understanding, data analysis, and writing capabilities of large language models based on artificial intelligence to optimize the presentation format of the original presentation pages. It also combines text with charts and graphics for secondary creation, utilizing the data analysis and processing capabilities of AI to perform more refined optimization of presentation pages containing statistical data. This avoids simply adding user-inputted text to the selected presentation template, resulting in a presentation that better meets the user's presentation creation needs and improving the overall quality of AI-generated presentations. Attached Figure Description

[0040] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0041] Figure 1 This is a diagram illustrating the application environment of a presentation processing method in one embodiment.

[0042] Figure 2 This is a flowchart illustrating a presentation processing method in one embodiment;

[0043] Figure 3 This is a flowchart illustrating a presentation processing method in another embodiment;

[0044] Figure 4 This is a logic diagram of a presentation processing method in one embodiment;

[0045] Figure 5 This is a structural block diagram of a presentation processing device in one embodiment;

[0046] Figure 6 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0047] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0048] In practice, the collection and processing of data in this application should strictly comply with the requirements of relevant national laws and regulations, obtain the informed consent or separate consent of the data subject, and carry out subsequent data use and processing within the scope of laws and regulations and the authorization of the data subject.

[0049] The presentation document processing method provided in this application embodiment can be applied to, for example, Figure 1 In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be integrated onto server 104, or it can be located in the cloud or on another network server.

[0050] Terminal 102 can acquire the original presentation page and input it into a pre-trained large language model to obtain the structured text of the original presentation page; wherein, the pre-trained large language model can be deployed on server 104; terminal 102 can identify the statistical data contained in the structured text of the presentation and generate a target statistical chart that matches the statistical data; terminal 102 can input the set of presentation elements into a pre-trained presentation layout classification model to obtain a presentation layout type that matches the set of presentation elements; the set of presentation elements includes at least the original presentation page, the structured text of the presentation, and the target statistical chart; terminal 102 can embed the set of presentation elements into candidate presentation pages that match the presentation layout type according to the presentation layout type to obtain an optimized presentation page.

[0051] The terminal 102 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, smart in-vehicle systems, and projection devices. Portable wearable devices can include smartwatches, smart bracelets, and head-mounted displays. Head-mounted displays can be virtual reality (VR) devices, augmented reality (AR) devices, and smart glasses. The server 104 can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0052] In one exemplary embodiment, such as Figure 2 As shown, a presentation processing method is provided, which can be applied to... Figure 1 Taking terminal 102 as an example, the explanation includes the following steps S202 to S208. Wherein:

[0053] Step S202: Obtain the original presentation page and input it into the pre-trained large language model to obtain the structured text of the original presentation page.

[0054] In this context, an original presentation page refers to a presentation page that has not undergone any enhancements or modifications. In practical applications, a presentation page can also be referred to as a slide. For example, an original presentation page could be slide D, which only contains text boxes. The text boxes in slide D display text content.

[0055] Here, structured text in a presentation refers to the structured text corresponding to the text content contained in the original presentation page. The structured format can be Markdown format.

[0056] Markdown format is a symbolic text format.

[0057] In practice, the terminal can receive the original presentation page uploaded by the user and input the original presentation page into a pre-trained large language model. The pre-trained large language model will then perform structuring processing on the text in the original presentation page and return a pre-formatted structured presentation text t.

[0058] In practical applications, this pre-trained large language model can also polish the text in the original presentation page and return a pre-formatted structured text t of the presentation.

[0059] Step S204: Identify the statistical data contained in the structured text of the presentation and generate a target statistical chart that matches the statistical data.

[0060] In practice, the terminal can identify the statistical data contained in the structured text of the presentation using a pre-trained large language model. Then, the terminal can call a chart generation tool to convert the statistical data contained in the structured text of the presentation into corresponding target statistical charts. The statistical data contained in the structured text of the presentation can then be visualized using the target statistical charts.

[0061] Step S206: Input the set of presentation elements into the pre-trained presentation layout classification model to obtain the presentation layout type that matches the set of presentation elements.

[0062] The presentation element set includes at least the original presentation page, the presentation structured text, and the target statistical chart.

[0063] In practice, the terminal can use the original presentation page, the structured text of the presentation, and the target statistical chart as the presentation element set D1. Then, the terminal can input the presentation element set D1 into a pre-trained presentation layout classification model to obtain the presentation layout type that matches the presentation element set. In practical applications, presentation layout types can include progressive graphic layout types, parallel graphic layout types, timeline layout types, SWOT (a four-quadrant layout method) layout types, fishbone diagram layout types, and general-to-specific structure layout types, etc.

[0064] Step S208: According to the presentation layout type, embed the presentation element set into the candidate presentation page that matches the presentation layout type to obtain the optimized presentation page.

[0065] In practical implementation, different presentation layout types have corresponding layout functions f. The terminal can input a set of presentation elements into the layout function f, which then organizes the set of presentation elements into presentation pages with that layout type. This achieves the goal of embedding the set of presentation elements into candidate presentation pages that match the presentation layout type, resulting in an optimized presentation page.

[0066] In the above presentation processing method, the original presentation page is obtained and input into a pre-trained large language model to obtain the structured text of the original presentation page. Then, by identifying the statistical data contained in the structured text, a target statistical chart matching the statistical data is generated. Next, a set of presentation elements, including the original presentation page, the structured text, and the target statistical chart, is input into a pre-trained presentation layout classification model to obtain a presentation layout type matching the set of presentation elements. Finally, the set of presentation elements is embedded into a candidate presentation layout that matches the layout type. The original presentation page is then optimized. This effectively leverages the understanding, data analysis, and writing capabilities of a large language model based on artificial intelligence to optimize the presentation format of the original presentation page. Text is combined with charts and graphics for secondary creation, utilizing the data analysis and processing capabilities of AI technology to perform more refined optimization of presentation pages containing statistical data. This avoids directly adding user-inputted text content to the selected presentation template, resulting in a presentation that better meets the user's presentation creation needs and improving the overall quality of AI-generated presentations.

[0067] In an exemplary embodiment, obtaining the original presentation page includes: in response to a presentation material upload operation, obtaining the presentation material uploaded by the presentation material upload operation; inputting the presentation material into a pre-trained large language model, performing text expansion on the presentation material through the pre-trained large language model to obtain the text expansion result; and generating the original presentation page based on the text expansion result.

[0068] The original presentation page is a presentation page that includes at least text boxes. Optionally, the original presentation page can be a slide containing only text boxes.

[0069] The text boxes on the original presentation page display expanded text results.

[0070] Presentation materials can refer to slide titles or slide outlines.

[0071] In practice, during the process of acquiring the original presentation slides, the user can input slide titles or outlines to the terminal, thus enabling the upload of presentation materials. The terminal can respond to the presentation material upload operation and retrieve the uploaded presentation materials.

[0072] Then, the terminal can input the slide title or slide outline into the generalized large language model, and the generalized large language model will expand the text based on the slide title or slide outline to obtain the slide text; then, the original presentation page D including the slide text will be output.

[0073] Optionally, obtaining the original presentation page includes: in response to the presentation page upload operation, obtaining the presentation page uploaded by the presentation page upload operation as the original presentation page.

[0074] In practice, the original presentation page can be obtained on the terminal, or the user can directly upload the original presentation page to the terminal so that the terminal can obtain the original presentation page. Specifically, the user can input a presentation page upload operation into the terminal, and the terminal can respond to the presentation page upload operation and obtain the presentation page uploaded by the presentation page upload operation as the original presentation page.

[0075] The technical solution of this embodiment, in response to the presentation material upload operation, obtains the presentation material uploaded by the operation, and inputs the presentation material into a pre-trained large language model. The pre-trained large language model performs text expansion on the presentation material to obtain the text expansion result. Based on the text expansion result, the original presentation page is generated. This allows users to optimize their slides by simply inputting the slide title or slide outline without preparing the original presentation, thus improving the efficiency of users in creating slide files.

[0076] In one exemplary embodiment, inputting the original presentation page into a pre-trained large language model to obtain the presentation structured text of the original presentation page includes: constructing text structured instructions for the pre-trained large language model based on the original presentation page; inputting the text structured instructions into the pre-trained large language model to obtain the response content of the pre-trained large language model to the text structured instructions, wherein the response content includes at least the presentation structured text.

[0077] Among them, text structuring instructions are used to instruct pre-trained large language models to extract the text content from the original presentation pages and convert the text content into text in a preset text structuring format. In practical applications, text structuring instructions can be referred to as text structuring prompts.

[0078] In practice, the original presentation page is input into a pre-trained large language model to obtain the structured text of the original presentation page. The terminal can then construct text structured instructions for the pre-trained large language model based on the original presentation page.

[0079] For example, structured text prompts can look like this:

[0080] Prompt1=

[0081] “

[0082] I will give you a text, and your task is to convert the text into a structured Markdown format.

[0083] The text is as follows:

[0084] {input_text}

[0085] You must adhere to the following guidelines:

[0086] - Do not add or modify the text content.

[0087] - Your output must always be in Markdown format. Do not interpret yourself or output anything else.

[0088] "

[0089] Of course, the terminal can also control the large language model to polish the text content of the slides. The structured prompts with polishing function are:

[0090] Prompt2=

[0091] “

[0092] I will give you a text, and your task is to polish the text and convert it into a structured Markdown format.

[0093] The text is as follows:

[0094] {input_text}

[0095] You must adhere to the following guidelines:

[0096] - The polished text should not exceed {n} characters.

[0097] - The writing must be fluent, and the revised text must convey the same meaning as the original text.

[0098] - Do not modify the data.

[0099] - Your output must always be in Markdown format. Do not interpret yourself or output anything else.

[0100] "

[0101] Then, the terminal can input the text structured instructions into the pre-trained large language model to obtain the response content of the pre-trained large language model to the text structured instructions. The response content includes at least the structured text of the presentation.

[0102] The technical solution of this embodiment constructs text-structured instructions for a pre-trained large language model based on the original presentation page; inputs the text-structured instructions into the pre-trained large language model to obtain the response content of the pre-trained large language model to the text-structured instructions. The response content includes at least the structured text of the presentation, which enables the terminal to effectively and accurately control the pre-trained large language model to extract the structured text of the presentation from the original presentation page.

[0103] In an exemplary embodiment, identifying statistical data contained in the structured text of a presentation and generating a target statistical chart that matches the statistical data includes: constructing statistical data processing instructions for a pre-trained large language model based on the structured text of the presentation; inputting the statistical data processing instructions into the pre-trained large language model to obtain the response content of the pre-trained large language model to the statistical data processing instructions, wherein the response content includes at least the structured statistical data; inputting the statistical data into a fine-tuned statistical chart type analysis model to obtain a target chart type that matches the statistical data; and mapping the statistical data to the target statistical chart by calling a chart generation function that matches the target chart type.

[0104] The statistical data processing instructions are used to instruct the pre-trained large language model to extract statistical data from the structured text of the presentation and output structured statistical data. In practical applications, statistical data processing instructions can refer to statistical data processing prompts.

[0105] In practice, when the terminal identifies the statistical data contained in the structured text of the presentation and generates a target statistical chart that matches the statistical data, the terminal can construct statistical data processing instructions for the pre-trained large language model based on the structured text of the presentation.

[0106] The statistical data processing instructions are as follows:

[0107] Prompt=

[0108] “

[0109] I will give you a text in Markdown format. Your task is to determine whether the text contains statistical data. Statistical data means that the text contains a series of numerical values ​​that have statistical significance.

[0110] The text is as follows:

[0111] {𝑡}

[0112] If the text contains statistics, then create a table of the statistics in the text and return it in Markdown table format. If the text does not contain statistics, then return "No statistics contained". Do not explain yourself or output anything else.

[0113] "

[0114] Then, the terminal inputs the statistical data processing instructions into the pre-trained large language model, and obtains the response content of the pre-trained large language model to the statistical data processing instructions. The response content includes at least structured statistical data, such as a statistical data chart in Markdown table format.

[0115] Then, the terminal can input statistical data into the fine-tuned statistical chart type analysis model to obtain the target chart type that matches the statistical data. By calling the chart generation function that matches the target chart type, the statistical data is mapped to the target statistical chart.

[0116] Specifically, the terminal can invoke a fine-tuned large language model G to analyze the statistical data and determine the appropriate chart type. The input to the fine-tuned large language model G is a statistical data chart in Markdown table format, and the output is a statistical chart type C suitable for the table content. chart Then, the terminal can call the chart generation function to convert the Markdown table-formatted statistical data chart into a C# chart. chart Target statistical charts.

[0117] The technical solution of this embodiment constructs statistical data processing instructions for a pre-trained large language model based on the structured text of the presentation, and inputs the statistical data processing instructions into the pre-trained large language model to obtain the response content of the pre-trained large language model to the statistical data processing instructions. The response content includes at least structured statistical data. Then, the statistical data is input into a fine-tuned statistical chart type analysis model to obtain the target chart type that matches the statistical data. By calling the chart generation function that matches the target chart type, the statistical data is mapped to the target statistical chart. This can effectively map the statistical data in the structured text of the presentation into a visual statistical chart, effectively improving the readability of the optimized presentation page.

[0118] In an exemplary embodiment, when the material requirements corresponding to the presentation layout type include presentation images, the optimized presentation page is obtained by embedding the presentation element set into candidate presentation pages that match the presentation layout type, according to the presentation layout type. This includes: generating presentation images that match the content of the presentation element set using a pre-trained visual model; adding the presentation images to the presentation element set to obtain an updated presentation element set; and embedding the updated presentation element set into candidate presentation pages according to the presentation layout type to obtain the optimized presentation page.

[0119] In practice, when the material requirements corresponding to the presentation layout type include presentation images, the terminal embeds the presentation element set into the candidate presentation page that matches the presentation layout type according to the presentation layout type. In the process of obtaining the optimized presentation page, the terminal can also generate presentation images that match the content of the presentation element set through a pre-trained visual model.

[0120] Specifically, when the accompanying image is a required parameter of the layout function f and the current presentation element set does not include an image, the terminal can call an open-source visual model to generate a presentation image that matches the content of the presentation element set, and add the presentation image to the presentation element set to obtain an updated presentation element set. This updated presentation element set can be represented as D2.

[0121] Then, the terminal can input the updated set of presentation elements into the layout function f. The layout function f organizes the set of presentation elements into a presentation page with the presentation layout type. This achieves the embedding of the set of presentation elements into candidate presentation pages that match the presentation layout type, resulting in an optimized presentation page.

[0122] The technical solution of this embodiment generates presentation images that match the content of the presentation element set through a pre-trained visual large model. The presentation images are then added to the presentation element set to obtain an updated presentation element set. According to the presentation layout type, the updated presentation element set is embedded into the candidate presentation page to obtain an optimized presentation page. This achieves the goal of using the large model's image generation capabilities to add images that match the text content and layout style to the user-indicated optimized slides, effectively improving the content richness of the optimized presentation page and enhancing the generation effect of AI-generated presentations.

[0123] In an exemplary embodiment, the method further includes: obtaining local adjustment instructions input to the optimized presentation page; inputting the local adjustment instructions into a pre-trained large language model to obtain an adjustment function for the optimized presentation page and the input parameters corresponding to the adjustment function; and performing a local adjustment operation on the optimized presentation page according to the adjustment function and the input parameters corresponding to the adjustment function.

[0124] The partial adjustment instructions include natural language for adjusting at least a portion of the optimized presentation page.

[0125] In practice, the terminal can also obtain the user's local adjustment instructions for the optimized presentation page, and input the local adjustment instructions into the pre-trained large language model to obtain the adjustment function for the optimized presentation page and the corresponding input parameters of the adjustment function; according to the adjustment function and the corresponding input parameters of the adjustment function, the local adjustment operation is performed on the optimized presentation page.

[0126] Specifically, a large language model H can be trained using function calling techniques. This large language model H can understand user input commands, such as local adjustment commands, by combining existing slides, and output the corresponding operation function sequence [h0, h1, ..., h]. n ] and the corresponding input parameters [i0,i1,...,i n ];

[0127] Then, the terminal can perform local adjustments to the optimized presentation page based on the sequence of operation functions output by the large language model H and the input parameters.

[0128] Here is an example:

[0129] When a user inputs "Increase the font size of the title," the large language model H outputs a font size adjustment function and its corresponding input parameters. The input parameters include the object to be adjusted and the font size. The object is the title, and the font size is slightly larger than the existing slide title. The font size adjustment function is then executed. If the large language model H cannot determine the corresponding operation function and input parameters, it can remove follow-up questions about the user's intent to clarify the specific target of the slide adjustment. If the user's input command cannot be implemented, the large language model H can output a rejection statement.

[0130] The technical solution of this embodiment obtains the local adjustment instructions input to the optimized presentation page, inputs the local adjustment instructions into a pre-trained large language model, obtains the adjustment function and the corresponding input parameters for the optimized presentation page, and then performs local adjustment operations on the optimized presentation page according to the adjustment function and the corresponding input parameters. This allows users to input natural language instructions to adjust at least a part of the optimized presentation page, and use these natural language instructions to control the terminal to perform local adjustment operations on the optimized presentation page. Users do not need to have professional presentation production knowledge to perform local adjustments to the presentation page, effectively improving the efficiency of presentation file production.

[0131] In another embodiment, such as Figure 3 As shown, a presentation processing method is provided, which can be applied to... Figure 1 Taking the terminal in the example, the explanation includes the following steps:

[0132] Step S302: Obtain the original presentation page. Based on the original presentation page, construct text structuring instructions for the pre-trained large language model. The text structuring instructions are used to instruct the pre-trained large language model to extract the text content from the original presentation page and convert the text content into text in a preset text structuring format.

[0133] Step S304: Input the text structure instructions into the pre-trained large language model to obtain the response content of the pre-trained large language model to the text structure instructions. The response content includes at least the structured text of the presentation.

[0134] Step S306: Based on the structured text of the presentation, construct statistical data processing instructions for the pre-trained large language model; the statistical data processing instructions are used to instruct the pre-trained large language model to extract statistical data from the structured text of the presentation and output structured statistical data.

[0135] Step S308: Input the statistical data processing instruction into the pre-trained large language model to obtain the response content of the pre-trained large language model to the statistical data processing instruction. The response content includes at least structured statistical data.

[0136] Step S310: Input the statistical data into the fine-tuned statistical chart type analysis model to obtain the target chart type that matches the statistical data. By calling the chart generation function that matches the target chart type, the statistical data is mapped to the target statistical chart.

[0137] Step S312: Input the presentation element set into the pre-trained presentation layout classification model to obtain the presentation layout type that matches the presentation element set; the presentation element set includes at least the original presentation page, the presentation structured text, and the target statistical chart.

[0138] Step S314: According to the presentation layout type, embed the presentation element set into the candidate presentation page that matches the presentation layout type to obtain the optimized presentation page.

[0139] It should be noted that the specific limitations of the above steps can be found in the specific limitations of a presentation processing method described above.

[0140] For the convenience of those skilled in the art, Figure 4 An example logic diagram of a presentation processing method is provided. A user can input a slide title or slide outline into the terminal, enabling the upload of presentation materials. The terminal responds to this upload operation by retrieving the uploaded presentation materials. Then, the terminal inputs the slide title or slide outline into a general-purpose large language model, which performs text expansion based on the slide title or outline to obtain the slide text. Finally, the terminal outputs the original presentation page D, including the slide text. Alternatively, a user can input a presentation page upload operation into the terminal. The terminal responds to this operation by retrieving the uploaded presentation page as the original presentation page. The original presentation page is then input into a pre-trained large language model to obtain the structured presentation text of the original presentation page.

[0141] Then, based on the data analysis capabilities of the large model, statistical data contained in the structured text of the presentation can be identified, and target statistical charts matching the statistical data can be generated. The terminal can then use the original presentation page, the structured text of the presentation, and the target statistical charts as a set of presentation elements, D1. The terminal can then input the set of presentation elements, D1, into a pre-trained presentation layout classification model to obtain the presentation layout type that matches the set of presentation elements. The terminal can then input the set of presentation elements into a layout function f, which organizes the set of presentation elements into presentation pages with that layout type. This achieves the goal of embedding the set of presentation elements into candidate presentation pages that match the presentation layout type, resulting in an optimized presentation page.

[0142] In addition, the terminal can also obtain the user's local adjustment instructions for the optimized presentation page, and input the local adjustment instructions into the pre-trained large language model to obtain the adjustment function for the optimized presentation page and the corresponding input parameters of the adjustment function; according to the adjustment function and the corresponding input parameters of the adjustment function, the local adjustment operation is performed on the optimized presentation page.

[0143] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0144] Based on the same inventive concept, this application also provides a presentation processing apparatus for implementing the presentation processing method described above. The solution provided by this apparatus is similar to the implementation scheme described in the above method; therefore, the specific limitations in one or more presentation processing apparatus embodiments provided below can be found in the limitations of the presentation processing method described above, and will not be repeated here.

[0145] In one exemplary embodiment, such as Figure 5 As shown, a presentation processing device is provided, comprising:

[0146] The content acquisition module 510 is used to acquire the original presentation page, input the original presentation page into a pre-trained large language model, and obtain the presentation structured text of the original presentation page.

[0147] The data recognition module 520 is used to identify the statistical data contained in the structured text of the presentation and generate a target statistical chart that matches the statistical data.

[0148] The layout classification module 530 is used to input the presentation element set into a pre-trained presentation layout classification model to obtain the presentation layout type that matches the presentation element set; the presentation element set includes at least the original presentation page, the presentation structured text, and the target statistical chart;

[0149] The element embedding module 540 is used to embed the presentation element set into a candidate presentation page that matches the presentation layout type according to the presentation layout type, so as to obtain an optimized presentation page.

[0150] In one embodiment, the original presentation page is a presentation page that includes at least text boxes. The content acquisition module 510 is configured to, in response to a presentation material upload operation, acquire the presentation material uploaded by the presentation material upload operation; input the presentation material into a pre-trained large language model, perform text expansion on the presentation material through the pre-trained large language model, and obtain a text expansion result; generate the original presentation page based on the text expansion result; the text boxes of the original presentation page display the text expansion result; or, in response to a presentation page upload operation, acquire the presentation page uploaded by the presentation page upload operation as the original presentation page.

[0151] In one embodiment, the content acquisition module 510 is further configured to construct a text structuring instruction for the pre-trained large language model based on the original presentation page; the text structuring instruction is used to instruct the pre-trained large language model to extract the text content in the original presentation page and convert the text content into text in a preset text structuring format; the text structuring instruction is input to the pre-trained large language model to obtain the response content of the pre-trained large language model to the text structuring instruction, the response content including at least the presentation structuring text.

[0152] In one embodiment, the data recognition module 520 is further configured to construct statistical data processing instructions for the pre-trained large language model based on the structured text of the presentation; the statistical data processing instructions are used to instruct the pre-trained large language model to extract statistical data from the structured text of the presentation and output structured statistical data; input the statistical data processing instructions to the pre-trained large language model to obtain the response content of the pre-trained large language model to the statistical data processing instructions, the response content including at least the structured statistical data; input the statistical data to a fine-tuned statistical chart type analysis model to obtain the target chart type matching the statistical data, and map the statistical data to the target statistical chart by calling the chart generation function matching the target chart type.

[0153] In one embodiment, when the material requirements corresponding to the presentation layout type include presentation images, the element embedding module 540 is used to generate presentation images that match the content of the presentation element set using a pre-trained visual large model, add the presentation images to the presentation element set to obtain an updated presentation element set; and embed the updated presentation element set into the candidate presentation page according to the presentation layout type to obtain the optimized presentation page.

[0154] In one embodiment, the apparatus is further configured to acquire a local adjustment instruction input to the optimized presentation page; the local adjustment instruction includes natural language for adjusting at least a portion of the optimized presentation page; input the local adjustment instruction to a pre-trained large language model to obtain an adjustment function for the optimized presentation page and input parameters corresponding to the adjustment function; and perform a local adjustment operation on the optimized presentation page according to the adjustment function and the input parameters corresponding to the adjustment function.

[0155] Each module in the aforementioned presentation processing device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0156] In one exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 6As shown, the computer device includes a processor, memory, input / output interfaces, a communication interface, a display unit, and an input device. The processor, memory, and input / output interfaces are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interfaces. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The input / output interfaces are used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, Near Field Communication (NFC), or other technologies. When the computer program is executed by the processor, it implements a presentation processing method. The display unit is used to form a visually visible image and can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0157] Those skilled in the art will understand that Figure 6 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0158] In one embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, which, when executed by the processor, causes the processor to perform the steps of the presentation processing method described above. The steps of the presentation processing method described here may be steps from one of the presentation processing methods in the various embodiments described above.

[0159] In one embodiment, a computer-readable storage medium is provided, storing a computer program that, when executed by a processor, causes the processor to perform the steps of the presentation processing method described above. The steps of the presentation processing method described here may be steps from one of the presentation processing methods in the various embodiments described above.

[0160] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, causes the processor to perform the steps of the presentation processing method described above. The steps of the presentation processing method described here may be steps from one of the presentation processing methods in the various embodiments described above.

[0161] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0162] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic resistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0163] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0164] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for processing presentation slides, characterized in that, The method includes: Obtain the original presentation page, input the original presentation page into a pre-trained large language model, and obtain the presentation structured text of the original presentation page; Identify statistical data contained in the structured text of the presentation and generate a target statistical chart that matches the statistical data; The presentation element set is input into a pre-trained presentation layout classification model to obtain a presentation layout type that matches the presentation element set; the presentation element set includes at least the original presentation page, the presentation structured text, and the target statistical chart; According to the presentation layout type, the set of presentation elements is embedded into a candidate presentation page that matches the presentation layout type to obtain an optimized presentation page.

2. The method according to claim 1, characterized in that, The original presentation page is a presentation page that includes at least text boxes. Obtaining the original presentation page includes: In response to the presentation material upload operation, obtain the presentation material uploaded in the presentation material upload operation; The presentation material is input into a pre-trained large language model, and the pre-trained large language model is used to perform text expansion on the presentation material to obtain the text expansion result. Based on the text expansion results, the original presentation page is generated; the text boxes on the original presentation page display the text expansion results. or, In response to the presentation page upload operation, the presentation page uploaded by the presentation page upload operation is obtained as the original presentation page.

3. The method according to claim 2, characterized in that, The step of inputting the original presentation page into a pre-trained large language model to obtain the structured presentation text of the original presentation page includes: Based on the original presentation page, construct text structuring instructions for the pre-trained large language model; the text structuring instructions are used to instruct the pre-trained large language model to extract the text content in the original presentation page and convert the text content into text in a preset text structuring format; The text structuring instruction is input into the pre-trained large language model to obtain the response content of the pre-trained large language model to the text structuring instruction, and the response content includes at least the structured text of the presentation.

4. The method according to claim 1, characterized in that, The step of identifying statistical data contained in the structured text of the presentation and generating a target statistical chart matching the statistical data includes: Based on the structured text of the presentation, a statistical data processing instruction is constructed for the pre-trained large language model; the statistical data processing instruction is used to instruct the pre-trained large language model to extract statistical data from the structured text of the presentation and output structured statistical data. The statistical data processing instruction is input into the pre-trained large language model to obtain the response content of the pre-trained large language model to the statistical data processing instruction, and the response content includes at least the structured statistical data. The statistical data is input into a fine-tuned statistical chart type analysis model to obtain the target chart type that matches the statistical data. By calling the chart generation function that matches the target chart type, the statistical data is mapped to the target statistical chart.

5. The method according to claim 1, characterized in that, When the material requirements corresponding to the presentation layout type include presentation images, the step of embedding the presentation element set into a candidate presentation page matching the presentation layout type to obtain an optimized presentation page includes: Using a pre-trained large visual model, a presentation image matching the content of the presentation element set is generated, and the presentation image is added to the presentation element set to obtain an updated presentation element set. According to the presentation layout type, the updated set of presentation elements is embedded into the candidate presentation page to obtain the optimized presentation page.

6. The method according to claim 1, characterized in that, The method further includes: Obtain partial adjustment instructions input on the optimized presentation page; the partial adjustment instructions include natural language for adjusting at least a portion of the optimized presentation page; The local adjustment instructions are input into a pre-trained large language model to obtain the adjustment function for the optimized presentation page and the input parameters corresponding to the adjustment function; Based on the adjustment function and the corresponding input parameters, perform local adjustment operations on the optimized presentation page.

7. A presentation processing device, characterized in that, The device includes: The content acquisition module is used to acquire the original presentation page, input the original presentation page into a pre-trained large language model, and obtain the presentation structured text of the original presentation page; The data recognition module is used to identify statistical data contained in the structured text of the presentation and generate target statistical charts that match the statistical data. The layout classification module is used to input the presentation element set into a pre-trained presentation layout classification model to obtain the presentation layout type that matches the presentation element set; the presentation element set includes at least the original presentation page, the presentation structured text, and the target statistical chart; The element embedding module is used to embed the presentation element set into a candidate presentation page that matches the presentation layout type, according to the presentation layout type, to obtain an optimized presentation page.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images