Manuscript generation method and device based on artificial intelligence, computer equipment and medium

Through artificial intelligence technology, the key data in the PPT file is automatically extracted and the speech template is filled, which solves the problem of low efficiency in the generation of speeches for financial business promotion and achieves rapid and efficient speech generation.

CN118964649APending Publication Date: 2024-11-15PING AN BANK CO LTD

Patent Information

Application Number
CN202411026075.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-29
Publication Date
2024-11-15

AI Technical Summary

Technical Problem

The existing method of generating speeches for financial business promotion requires financial business personnel to spend a lot of time to review and integrate the content in the PPT file, resulting in low generation efficiency.

Method used

Using an artificial intelligence-based method, PPT files are automatically processed through title recognition model, paragraph extraction model and processing library, title data, paragraph text data and chart data are extracted, and they are filled into preset speech templates to generate Target speech.

Benefits of technology

It realizes automatic and rapid generation of target speeches that match PPT files, significantly improving the efficiency and intelligence of speech generation, and reducing the time overhead of manual integration.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118964649A_ABST
    Figure CN118964649A_ABST
Patent Text Reader

Abstract

The invention belongs to the field of artificial intelligence and the field of financial science and technology, and relates to an artificial intelligence-based manuscript generation method, which comprises the following steps of: obtaining a PPT file; performing title recognition on the PPT file based on a title recognition model to obtain title data; performing paragraph extraction on the PPT file based on a paragraph extraction model to obtain paragraph text data Performing graph data extraction on the PPT file based on the processing library to obtain picture data and graph data; performing format conversion processing on the picture data and the chart data to obtain a first image file and a second image file; determining placeholders matched with the to-be-filled data from the speech template; and filling the to-be-filled data into a placeholder matched with the filled data in the speech draft template to obtain a target speech draft. The invention further provides a manuscript generation device based on artificial intelligence, computer equipment and a storage medium. In addition, the target speech may be stored in the blockchain. According to the invention, the generation efficiency and the generation intelligence of the target speech are effectively improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence development technology and the field of financial technology, and in particular to artificial intelligence-based document generation methods, devices, computer equipment and storage media. Background Art

[0002] In the field of financial business, in order to enhance customers' understanding of relevant financial business and popularize relevant financial business knowledge to customers, financial enterprises usually use financial business personnel to promote financial business to customers through speeches, so as to facilitate customers' understanding of relevant financial business.

[0003] At present, the method of generating a speech for financial business promotion is to first send a PPT file related to financial business promotion to the financial business personnel, and then the financial business personnel refer to the content of the PPT file and manually integrate the content information in the PPT file to generate the required speech for financial business promotion. However, this method of generating a speech requires the financial business personnel to spend a lot of time to refer to and integrate the content in the PPT file, resulting in low efficiency in generating the speech. Summary of the invention

[0004] The purpose of the embodiments of the present application is to propose a manuscript generation method, device, computer equipment and storage medium based on artificial intelligence to solve the technical problem that the existing method of generating speeches for financial business promotion requires financial business personnel to spend a lot of time to review and integrate the contents of PPT files, resulting in low efficiency in generating speeches.

[0005] In order to solve the above technical problems, the present application embodiment provides a method for generating a document based on artificial intelligence, which adopts the following technical solution:

[0006] Obtain a PPT file; wherein the PPT file includes one or more slides;

[0007] Performing title recognition processing on the PPT file based on a preset title recognition model to obtain corresponding title data;

[0008] Perform paragraph extraction processing on the PPT file based on a preset paragraph extraction model to obtain corresponding paragraph text data;

[0009] Performing image data extraction processing on the PPT file based on a preset processing library to obtain corresponding image data and chart data;

[0010] Performing format conversion processing on the picture data to obtain a first image file, and performing format conversion processing on the chart data to obtain a second image file;

[0011] Determine a placeholder matching the data to be filled in from a preset speech template; wherein the data to be filled in includes the title data, the paragraph text data, the first image file and the second image file;

[0012] The data to be filled is filled into the placeholder in the speech template that matches the filled data to obtain the corresponding target speech.

[0013] Furthermore, the steps of converting the image data into a first image file and converting the chart data into a second image file specifically include:

[0014] Saving the image data as a temporary file;

[0015] Copying the temporary file to a preset new folder to obtain a corresponding designated file, and using the designated file as the first image file;

[0016] Performing screenshot processing on the chart data based on a preset screenshot tool to obtain corresponding screenshot data;

[0017] The screenshot data is saved in a file format to obtain a corresponding screenshot file, and the screenshot file is used as the second image file.

[0018] Furthermore, the step of determining a placeholder matching the data to be filled in from the preset speech template specifically includes:

[0019] Calling the speech template;

[0020] Determine the title placeholder, text placeholder, picture placeholder and chart placeholder in the speech template based on preset recognition rules;

[0021] using the title placeholder as a first placeholder that matches the title data;

[0022] Using the text placeholder as a second placeholder that matches the paragraph text data;

[0023] Using the picture placeholder as a third placeholder matching the first image file;

[0024] The chart placeholder is used as a fourth placeholder matched with the second image file.

[0025] Furthermore, the step of filling the data to be filled into the placeholder in the speech template that matches the filled data to obtain the corresponding target speech specifically includes:

[0026] Filling the title data into the first placeholder in the speech template;

[0027] Filling the paragraph text data into the second placeholder;

[0028] Filling the first image file into the third placeholder;

[0029] Filling the second image file into the fourth placeholder to obtain a filled speech template;

[0030] The filled speech template is used as the target speech.

[0031] Furthermore, the step of filling the third placeholder with the first image file specifically includes:

[0032] Obtaining area size data of the third placeholder;

[0033] Performing a size adjustment process on the first image file based on the region size data to obtain a corresponding third image file;

[0034] Obtaining style setting information of the speech template;

[0035] Performing format adjustment processing on the third image file based on the style setting information to obtain a corresponding fourth image file;

[0036] Performing alignment processing on the fourth image file to obtain a corresponding fifth image file;

[0037] The third placeholder is filled with the fifth image file.

[0038] Furthermore, before the step of performing title recognition processing on the PPT file based on a preset title recognition model to obtain corresponding title data, the method further includes:

[0039] Acquire pre-collected designated text data; wherein the designated text data is data that has been titled;

[0040] Dividing the designated sample data into a training set and a test set;

[0041] Get the preset classification model;

[0042] Based on the training set, the classification model is trained using a supervised learning method to obtain a trained specified classification model;

[0043] Performing model evaluation on the specified classification model based on the test set;

[0044] If the designated classification model passes the model evaluation, the designated classification model is used as the title recognition model.

[0045] Furthermore, after the step of filling the data to be filled into the placeholder in the speech template that matches the filled data to obtain the corresponding target speech, the method further includes:

[0046] Presenting the target speech;

[0047] Determining whether a modification request for the target speech script is received, triggered by a user;

[0048] If yes, receiving an editing operation input by the user for the target speech;

[0049] Performing corresponding modification processing on the target speech draft based on the editing operation to obtain a modified target speech draft;

[0050] The modified target speech is stored.

[0051] In order to solve the above technical problems, the embodiment of the present application further provides a document generation device based on artificial intelligence, which adopts the following technical solution:

[0052] A first acquisition module is used to acquire a PPT file; wherein the PPT file includes one or more slides;

[0053] A recognition module, used to perform title recognition processing on the PPT file based on a preset title recognition model to obtain corresponding title data;

[0054] A first extraction module is used to perform paragraph extraction processing on the PPT file based on a preset paragraph extraction model to obtain corresponding paragraph text data;

[0055] A second extraction module is used to perform image data extraction processing on the PPT file based on a preset processing library to obtain corresponding image data and chart data;

[0056] a conversion module, configured to perform format conversion processing on the image data to obtain a first image file, and to perform format conversion processing on the chart data to obtain a second image file;

[0057] A first determining module is used to determine a placeholder matching the data to be filled in from a preset speech template; wherein the data to be filled in includes the title data, the paragraph text data, the first image file and the second image file;

[0058] The processing module is used to fill the data to be filled into the placeholder in the speech template that matches the filled data to obtain the corresponding target speech.

[0059] In order to solve the above technical problems, the embodiment of the present application further provides a computer device, which adopts the following technical solution:

[0060] Obtain a PPT file; wherein the PPT file includes one or more slides;

[0061] Performing title recognition processing on the PPT file based on a preset title recognition model to obtain corresponding title data;

[0062] Perform paragraph extraction processing on the PPT file based on a preset paragraph extraction model to obtain corresponding paragraph text data;

[0063] Performing image data extraction processing on the PPT file based on a preset processing library to obtain corresponding image data and chart data;

[0064] Performing format conversion processing on the picture data to obtain a first image file, and performing format conversion processing on the chart data to obtain a second image file;

[0065] Determine a placeholder matching the data to be filled in from a preset speech template; wherein the data to be filled in includes the title data, the paragraph text data, the first image file and the second image file;

[0066] The data to be filled is filled into the placeholder in the speech template that matches the filled data to obtain the corresponding target speech.

[0067] In order to solve the above technical problems, the embodiment of the present application further provides a computer-readable storage medium, which adopts the following technical solution:

[0068] Obtain a PPT file; wherein the PPT file includes one or more slides;

[0069] Performing title recognition processing on the PPT file based on a preset title recognition model to obtain corresponding title data;

[0070] Perform paragraph extraction processing on the PPT file based on a preset paragraph extraction model to obtain corresponding paragraph text data;

[0071] Performing image data extraction processing on the PPT file based on a preset processing library to obtain corresponding image data and chart data;

[0072] Performing format conversion processing on the picture data to obtain a first image file, and performing format conversion processing on the chart data to obtain a second image file;

[0073] Determine a placeholder matching the data to be filled in from a preset speech template; wherein the data to be filled in includes the title data, the paragraph text data, the first image file and the second image file;

[0074] The data to be filled is filled into the placeholder in the speech template that matches the filled data to obtain the corresponding target speech.

[0075] Compared with the prior art, the embodiments of the present application have the following beneficial effects:

[0076] The present application first obtains a PPT file; wherein the PPT file includes one or more slides; then, based on a preset title recognition model, the PPT file is subjected to title recognition processing to obtain corresponding title data; and based on a preset paragraph extraction model, the PPT file is subjected to paragraph extraction processing to obtain corresponding paragraph text data; and based on a preset processing library, the PPT file is subjected to image data extraction processing to obtain corresponding picture data and chart data; then, the picture data is subjected to format conversion processing to obtain a first image file, and the chart data is subjected to format conversion processing to obtain a second image file; subsequently, a placeholder matching the data to be filled is determined from a preset speech template; wherein the data to be filled includes the title data, the paragraph text data, the first image file and the second image file; finally, the data to be filled is filled into the placeholder matching the fill data in the speech template to obtain the corresponding target speech. After acquiring the PPT file, the present application will automatically and quickly extract the title data, paragraph text data, image data and chart data in the PPT file based on the use of the title recognition model, paragraph extraction model and processing library, and then quickly and accurately fill the data to be filled into the placeholders in the speech template that match the fill data according to the placeholders that match the data to be filled in determined from the preset speech template to obtain the corresponding target speech. Different from the existing method of manually integrating the content information in the PPT file to generate the required speech for financial business promotion, the present application can automatically and quickly generate a target speech that matches the PPT file, effectively improving the generation efficiency and intelligence of the target speech. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] In order to more clearly illustrate the scheme in the present application, a brief introduction is given below to the drawings required for use in the description of the embodiments of the present application. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0078] Figure 1 is an exemplary system architecture diagram to which the present application may be applied;

[0079] Figure 2 A flowchart of an embodiment of a method for generating a document based on artificial intelligence according to the present application;

[0080] Figure 3 is a structural schematic diagram of an embodiment of an artificial intelligence-based document generation device according to the present application;

[0081] Figure 4 It is a structural diagram of an embodiment of a computer device according to the present application. DETAILED DESCRIPTION

[0082] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by technicians in the technical field of the present application; the terms used in the specification of the application herein are only for the purpose of describing specific embodiments and are not intended to limit the present application; the terms "including" and "having" and any variations thereof in the specification and claims of the present application and the above-mentioned drawings are intended to cover non-exclusive inclusions. The terms "first", "second", etc. in the specification and claims of the present application or the above-mentioned drawings are used to distinguish different objects, not to describe a specific order.

[0083] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0084] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings.

[0085] like Figure 1As shown, the system architecture 100 may include terminal devices 101, 102, 103, a network 104 and a server 105. The network 104 is used to provide a medium for communication links between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0086] Users can use terminal devices 101, 102, 103 to interact with server 105 through network 104 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 101, 102, 103, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social platform software, etc.

[0087] The terminal devices 101, 102, and 103 can be various electronic devices with display screens and supporting web browsing, including but not limited to smart phones, tablet computers, e-book readers, MP3 players (Moving Picture Experts Group Audio Layer III), MP4 (Moving Picture Experts Group Audio Layer IV) players, laptop computers, desktop computers, and the like.

[0088] The server 105 may be a server that provides various services, such as a background server that provides support for web pages displayed on the terminal devices 101 , 102 , and 103 .

[0089] It should be noted that the artificial intelligence-based document generation method provided in the embodiment of the present application is generally executed by a server / terminal device, and accordingly, the artificial intelligence-based document generation device is generally set in the server / terminal device.

[0090] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0091] Continue to refer Figure 2, showing a flowchart of an embodiment of the method for generating a document based on artificial intelligence according to the present application. According to different requirements, the order of the steps in the flowchart can be changed, and some steps can be omitted. The method for generating a document based on artificial intelligence provided in the embodiment of the present application can be applied to any scenario where automatic document generation is required, and the method for generating a document based on artificial intelligence can be applied to products in these scenarios, for example, document generation for the promotion of insurance business knowledge in the field of finance and insurance. The method for generating a document based on artificial intelligence comprises the following steps:

[0092] Step S201, obtaining a PPT file; wherein the PPT file includes one or more slides.

[0093] In this embodiment, the electronic device (eg, Figure 1 The server / terminal device shown in the figure) can obtain the PPT file through a wired connection or a wireless connection. It should be noted that the above-mentioned wireless connection method may include but is not limited to 3G / 4G / 5G connection, Wi-Fi connection, Bluetooth connection, WiMAX connection, Zigbee connection, UWB (Ultra Wi-Fi Band) connection, and other wireless connection methods currently known or to be developed in the future. The above-mentioned PPT (Power Point, presentation software) file may be a PPT file used for the promotion of insurance business knowledge in the financial insurance field. The PPT file may be a PPT file input by the user according to the processing requirements for generating a speech related to the PPT file. Among them, the slides of the above-mentioned PPT file include at least data such as titles, paragraph texts, images, and charts.

[0094] Step S202: performing title recognition processing on the PPT file based on a preset title recognition model to obtain corresponding title data.

[0095] In this embodiment, the PPT file can be input into the above-mentioned title recognition model so that the title recognition model can perform title recognition processing on the PPT file and output corresponding title data. The training and generation process of the above-mentioned title recognition model will be further described in detail in the subsequent specific embodiments of the present application, and will not be elaborated on here.

[0096] Step S203: perform paragraph extraction processing on the PPT file based on a preset paragraph extraction model to obtain corresponding paragraph text data.

[0097] In this embodiment, the PPT file can be input into the above paragraph extraction model to perform paragraph extraction processing on the PPT file through the paragraph extraction model, and the corresponding paragraph text data will be output. Specifically, the training generation process of the above paragraph extraction model includes: obtaining pre-collected text sample data; wherein the text sample data is text data that has been processed with paragraph boundary annotation; dividing the text sample data into a training data set and a test data set; obtaining a preset sequence-based model; based on the training data set, using a supervised learning method to train the classification model, and optimizing the model parameters by minimizing the error between the predicted boundary and the actual boundary to obtain a trained target model; based on the test data set, the target model is evaluated; if the target model passes the model evaluation, the target model is used as the paragraph extraction model. Among them, the paragraph extraction model obtained by training is a model for processing paragraph segmentation tasks, and identifies the paragraph boundaries in the text to be identified, and divides the text into multiple paragraph texts. Paragraph segmentation is the process of dividing continuous text content into multiple paragraph texts according to semantic or structural coherence. In addition, since paragraph division needs to consider the contextual information and coherence in the text, the sequence-based model can capture this long-distance dependency well. The above-mentioned sequence-based model can specifically adopt an LSTM (Long Short-Term Memory) model or a GRU (Gated Recur rent Unit) model. This application uses a sequence-based model as a basic model to train and generate a paragraph extraction model for processing paragraph division tasks, which can effectively ensure the processing accuracy of paragraph extraction processing for PPT files.

[0098] Step S204: extracting the image data from the PPT file based on a preset processing library to obtain corresponding image data and chart data.

[0099] In this embodiment, the above processing library can specifically adopt the python-pptx library. The python-pptx library is a library for specifically reading and parsing PPT files. The shape data in each slide in the PPT file can be traversed by using the python-pptx library, and the picture object and the chart object can be found from the shape data, and the image object can be used as the above image data, and the chart object can be used as the above chart data.

[0100] Step S205 , performing format conversion processing on the picture data to obtain a first image file, and performing format conversion processing on the chart data to obtain a second image file.

[0101] In this embodiment, the specific implementation process of the above-mentioned format conversion processing of the picture data to obtain the first image file, and the format conversion processing of the chart data to obtain the second image file will be further described in detail in subsequent specific embodiments of this application, and will not be elaborated in detail here.

[0102] Step S206, determining a placeholder matching the data to be filled in from a preset speech template; wherein the data to be filled in includes the title data, the paragraph text data, the first image file and the second image file.

[0103] In this embodiment, the specific implementation process of determining the placeholders matching the data to be filled in from the preset speech template will be described in further detail in subsequent specific embodiments of the present application, and will not be elaborated on here.

[0104] Step S207, filling the data to be filled into the placeholder in the speech template that matches the filled data, to obtain the corresponding target speech.

[0105] In this embodiment, the specific implementation process of filling the data to be filled into the placeholder in the speech template that matches the filled data to obtain the corresponding target speech will be described in further detail in subsequent specific embodiments of the present application and will not be elaborated on here.

[0106] After acquiring the PPT file, the present application will automatically and quickly extract the title data, paragraph text data, image data and chart data in the PPT file based on the use of the title recognition model, paragraph extraction model and processing library, and then quickly and accurately fill the data to be filled into the placeholders in the speech template that match the fill data according to the placeholders that match the data to be filled in determined from the preset speech template to obtain the corresponding target speech. Different from the existing method of manually integrating the content information in the PPT file to generate the required speech for financial business promotion, the present application can automatically and quickly generate a target speech that matches the PPT file, effectively improving the generation efficiency and intelligence of the target speech.

[0107] In some optional implementations, step S205 includes the following steps:

[0108] The image data is saved as a temporary file.

[0109] In this embodiment, the processing library does not support directly saving the image data as a file, so the image data is first saved as a temporary file, and then the temporary file is copied to a blank folder to complete the conversion of the image data into an image file.

[0110] The temporary file is copied to a preset newly created folder to obtain a corresponding designated file, and the designated file is used as the first image file.

[0111] In this embodiment, the preset newly created folder is a blank folder.

[0112] The chart data is processed by screenshot based on a preset screenshot tool to obtain corresponding screenshot data.

[0113] In this embodiment, the screenshot tool may be a general screenshot tool.

[0114] The screenshot data is saved in a file format to obtain a corresponding screenshot file, and the screenshot file is used as the second image file.

[0115] In this embodiment, the processing library does not support directly saving the chart data as a file. Therefore, the image data is firstly captured as corresponding screenshot data using a screenshot tool, and then the screenshot data is saved as a file format to complete the conversion of the chart data into an image file.

[0116] The present application saves the image data as a temporary file; then copies the temporary file to a preset new folder to obtain a corresponding designated file, and uses the designated file as the first image file; then performs screenshot processing on the chart data based on a preset screenshot tool to obtain corresponding screenshot data; subsequently saves the screenshot data as a file format to obtain a corresponding screenshot file, and uses the screenshot file as the second image file. The present application saves the image data as a temporary file, and then copies the temporary file to a preset new folder to obtain a first image file, thereby realizing fast and intelligent completion of the format conversion processing of the image data, that is, realizing the need to extract the image file from the PPT file, and effectively solving the problem that the above-mentioned processing library does not support directly saving the above-mentioned image data as a file. In addition, by using a screenshot tool to perform screenshot processing on the chart data to obtain screenshot data, and saving the screenshot data as a file format to obtain a second image file, it is possible to realize fast and intelligent completion of the format conversion processing of the chart data, that is, realizing the need to extract the chart file from the PPT file, and effectively solving the problem that the above-mentioned processing library does not support directly saving the above-mentioned chart data as a file.

[0117] In some optional implementations of this embodiment, step S206 includes the following steps:

[0118] Call the speech template.

[0119] In this embodiment, the above-mentioned speech template is a template file compiled according to the actual speech construction requirements. The speech template contains multiple placeholders, and the placeholders are used to indicate the location where the content needs to be filled. A placeholder is a preset container or area for placing corresponding content on the above-mentioned speech template, such as a title, paragraph text, picture, chart, etc. The placeholder can help users organize the content to be filled in the speech according to the style and layout requirements of the speech template, and ensure that the presentation of the content to be filled is professional and consistent. Specifically, each placeholder included in the speech template will contain corresponding text prompt information. For the placeholder that needs to be filled with a title, the corresponding text prompt information is "A title needs to be added here"; for the placeholder that needs to be filled with a paragraph text, the corresponding text prompt information is "A paragraph text needs to be added here"; for the placeholder that needs to be filled with a picture, the corresponding text prompt information is "A picture needs to be added here"; for the placeholder that needs to be filled with a chart, the corresponding text prompt information is "A chart needs to be added here".

[0120] The title placeholder, text placeholder, picture placeholder and chart placeholder in the speech template are determined based on preset recognition rules.

[0121] In this embodiment, the text prompt information of each placeholder included in the speech template can be obtained, and for each placeholder, the text prompt information of the placeholder is added with type keywords for identification, and then the category information of the placeholder is determined according to the obtained recognition result. Specifically, the process of determining the title placeholder in the speech template may include: if the text prompt information of the placeholder is "A title needs to be added here", the recognition result "title" can be obtained by adding type keywords to the text prompt information for identification, and then the recognition result "title" is used as the category information of the placeholder, and the placeholder is used as the title placeholder in the speech template. Similarly, the process of determining the text placeholder, picture placeholder and chart placeholder in the speech template can refer to the above-mentioned process of determining the title placeholder, and no further elaboration is made here.

[0122] The title placeholder is used as the first placeholder matched with the title data.

[0123] In this embodiment, after determining the title placeholder in the speech template, the title placeholder is automatically determined as the first placeholder matching the title data, that is, the title placeholder is the area in the speech template for placing the title data.

[0124] The text placeholder is used as a second placeholder that matches the paragraph text data.

[0125] In this embodiment, after determining the text placeholder in the speech template, the text placeholder is automatically determined as the second placeholder matching the paragraph text data, that is, the text placeholder is the area in the speech template for placing the paragraph text data.

[0126] The picture placeholder is used as a third placeholder matching the first image file.

[0127] In this embodiment, after determining the image placeholder in the speech template, the image placeholder is automatically determined as the third placeholder matching the first image file, that is, the image placeholder is the area in the speech template for placing the first image file.

[0128] The chart placeholder is used as a fourth placeholder matched with the second image file.

[0129] In this embodiment, after determining the chart placeholder in the speech template, the chart placeholder is automatically determined as the fourth placeholder matching the second image file, that is, the chart placeholder is the area in the speech template for placing the second image file.

[0130] The present application calls the speech template; then determines the title placeholder, text placeholder, picture placeholder and chart placeholder in the speech template based on preset recognition rules; then uses the title placeholder as the first placeholder that matches the title data; uses the text placeholder as the second placeholder that matches the paragraph text data; uses the picture placeholder as the third placeholder that matches the first image file; and uses the chart placeholder as the fourth placeholder that matches the second image file. The present application can quickly determine the title placeholder, text placeholder, picture placeholder and chart placeholder in the speech template based on the use of recognition rules, and then can accurately determine the placeholder that matches the data to be filled based on the obtained title placeholder, text placeholder, picture placeholder and chart placeholder, effectively improving the processing efficiency and accuracy of determining the placeholder that matches the data to be filled.

[0131] In some optional implementations, step S207 includes the following steps:

[0132] The title data is filled into the first placeholder in the speech template.

[0133] In this embodiment, the first position information of the first placeholder in the speech template can be determined, and then the title data can be filled into the first placeholder in the speech template based on the first position information.

[0134] The paragraph text data is filled into the second placeholder.

[0135] In this embodiment, the second position information of the second placeholder in the speech template can be determined, and then the paragraph text data can be filled into the second placeholder based on the second position information.

[0136] Fill the third placeholder with the first image file.

[0137] In this embodiment, the specific implementation process of filling the first image file into the third placeholder will be further described in detail in subsequent specific embodiments of the present application, and will not be elaborated on here.

[0138] Fill the fourth placeholder with the second image file to obtain a filled speech template.

[0139] In this embodiment, the fourth position information of the fourth placeholder in the speech template can be determined, and then the second image file can be filled into the fourth placeholder based on the fourth position information.

[0140] The filled speech template is used as the target speech.

[0141] The present application fills the title data into the first placeholder in the speech template; fills the paragraph text data into the second placeholder; fills the first image file into the third placeholder; and fills the second image file into the fourth placeholder, so as to quickly and accurately fill various data to be filled into the matching placeholders in the speech template, and uses the obtained filled speech template as the target speech, thereby realizing automatic generation of the target speech, improving the generation efficiency of the target speech, and ensuring the data accuracy of the generated target speech.

[0142] In some optional implementations, filling the third placeholder with the first image file comprises the following steps:

[0143] Get the area size data of the third placeholder.

[0144] In this embodiment, after the speech template is written, the area sizes of all placeholders in the speech are counted and stored in the corresponding placeholder information. Therefore, the placeholder information of the speech template can be queried later to extract the area size data matching the picture placeholder, i.e., the third placeholder, from the placeholder information.

[0145] The first image file is resized based on the region size data to obtain a corresponding third image file.

[0146] In this embodiment, the size adjustment process includes: obtaining the size data of the first image file, and calculating the product of the area size data of the third placeholder and the preset ratio to obtain the specified size data, and then adjusting the size data of the first image file to be the same as the specified size data to obtain the corresponding third image file. The value range of the preset ratio is 0.6-0.8. The specific value of the preset ratio can be determined according to actual usage requirements. By resizing the first image file based on the area size data and the preset ratio, it can be ensured that the image file subsequently filled into the third placeholder can adapt to the area corresponding to the third placeholder, rather than the problem of the image file being too large or too small, thereby improving the intelligence and accuracy of the filling of the third image file.

[0147] Get the style setting information of the speech template.

[0148] In this embodiment, after the speech template is written, the style setting information corresponding to the speech will also be stored. The style setting information can be set according to the actual construction requirements of the speech.

[0149] The format of the third image file is adjusted based on the style setting information to obtain a corresponding fourth image file.

[0150] In this embodiment, the initial border, initial shadow, and initial color of the third image file can be adjusted to be the same as the specific border, specific shadow, and specific color according to the style information such as the specific border, specific shadow, and specific color contained in the style setting information to obtain a fourth image file, thereby ensuring that the format of the fourth image file is consistent with the overall style of the above-mentioned speech template, thereby improving the adaptability of the filling of the fourth image file.

[0151] Perform alignment processing on the fourth image file to obtain a corresponding fifth image file.

[0152] In this embodiment, the fourth image file may be aligned by calling a general alignment tool to ensure the neatness of the fourth image file within the third placeholder.

[0153] The third placeholder is filled with the fifth image file.

[0154] In this embodiment, the third position information of the third placeholder in the speech template can be determined, and then the fifth image file can be filled into the third placeholder based on the third position information.

[0155] The present application obtains the area size data of the third placeholder; then resizes the first image file based on the area size data to obtain the corresponding third image file; then obtains the style setting information of the speech template; and performs format adjustment on the third image file based on the style setting information to obtain the corresponding fourth image file; subsequently aligns the fourth image file to obtain the corresponding fifth image file; and finally fills the fifth image file into the third placeholder. In the process of filling the first image file into the third placeholder, the present application also intelligently resizes, formats and processes the first image file based on the area size data of the third placeholder and the style setting information of the speech template to obtain the fifth image file, and then fills the fifth image file into the third placeholder, effectively ensuring that the fifth image file filled into the third placeholder can adapt to the area corresponding to the third placeholder, and the format of the fifth image file is consistent with the overall style of the above-mentioned speech template, and has neatness in the third placeholder, thereby improving the intelligence, accuracy, adaptability and neatness of filling the third image file.

[0156] In some optional implementations of this embodiment, before step S202, the electronic device may further perform the following steps:

[0157] Acquire pre-collected designated text data; wherein the designated text data is data that has been titled.

[0158] In this embodiment, the above-mentioned designated text data is text data containing titles of various types and styles. Among them, for the title tagging processing of the designated text data, manual tagging or automatic machine tagging can be adopted. The automatic machine tagging method is preferably adopted to improve the processing efficiency of the label tagging processing.

[0159] The specified sample data is divided into a training set and a test set.

[0160] In this embodiment, the designated sample data may be divided into a training set and a test set according to a preset ratio value. The selection of the ratio value is not specifically limited and may be determined according to actual division requirements, for example, may be set to 8:2.

[0161] Get the preset classification model.

[0162] In this embodiment, the above classification model can specifically adopt any one of naive Bayes, support vector machine, convolutional neural network, recurrent neural network, etc. For small data sets or scenarios that require fast training, naive Bayes is a good choice. It assumes that the features are independent of each other, thereby simplifying the calculation process. Support vector machine is a powerful classifier that can process high-dimensional data and find the optimal classification boundary. It usually has good performance for text classification tasks. Convolutional neural networks and recurrent neural networks can automatically learn the feature representation of text and perform well on large-scale data sets. Among them, by using a classification model with a simple structure and good performance as a basic model to train and generate a title recognition model for the classification task of title recognition, the workload of model training can be effectively reduced while ensuring the accuracy of the model output, and the cost of building the model is also low.

[0163] Based on the training set, the classification model is trained using a supervised learning method to obtain a trained specified classification model.

[0164] In this embodiment, the classification model can be trained by using the training set including input features and corresponding target outputs (title labels) so that the classification model can learn the mapping relationship between the input features and the target outputs and obtain a trained specified classification model.

[0165] The specified classification model is evaluated based on the test set.

[0166] In this embodiment, the test set can be input into the specified classification model to perform title recognition processing on the test set using the specified classification model, and the specified classification model can be evaluated according to the classification result output by the specified classification model, and an evaluation index value of the model evaluation can be generated. The evaluation index value is the calculated accuracy value and recall value used to evaluate the performance of the model.

[0167] If the designated classification model passes the model evaluation, the designated classification model is used as the title recognition model.

[0168] In this embodiment, it can be determined whether the above accuracy value is greater than a preset accuracy threshold, and whether the above recall value is greater than a preset recall threshold. If the above accuracy value is greater than the accuracy threshold, and the above recall value is greater than the recall threshold, it is determined that the specified classification model passes the model evaluation, and the specified classification model is used as the title recognition model.

[0169] The present application obtains pre-collected designated text data; then divides the designated sample data into a training set and a test set; then obtains a preset classification model; and based on the training set, trains the classification model using a supervised learning method to obtain a trained designated classification model; then performs a model evaluation on the designated classification model based on the test set; if the designated classification model passes the model evaluation, the designated classification model is used as the title recognition model. The present application divides the pre-collected designated text data into a training set and a test set, and then trains and evaluates the preset classification model based on the training set and the test set, thereby achieving rapid and accurate construction of the required title recognition model. The present application uses a classification model with a simple structure and good performance as a basic model to train and generate a title recognition model for the classification task of title recognition, which can effectively reduce the workload of model training while ensuring the accuracy of the model output, and the cost of building the model is also low.

[0170] In some optional implementations of this embodiment, after step S207, the electronic device may further perform the following steps:

[0171] The target speech is presented.

[0172] In this embodiment, after the target speech is automatically generated, the target speech can be sent to a presentation interface of a communication terminal of a speaker corresponding to the target speech for presentation.

[0173] Determine whether a modification request for the target speech script triggered by a user is received.

[0174] In this embodiment, the user is the speaker corresponding to the target speech. The user can click the modification button in the display interface to trigger a modification request for the target speech displayed in the display interface.

[0175] If so, receive the editing operation of the target speech input by the user.

[0176] In this embodiment, the user can perform corresponding editing operations on the target speech in the above display interface according to the actual construction requirements of the target speech, wherein the editing operations may include operations such as adding content, deleting content, reducing content, and replacing content.

[0177] Based on the editing operation, the target speech is modified accordingly to obtain a modified target speech.

[0178] In this embodiment, the corresponding modification process of the target speech draft can be completed by responding to the editing operation of the target speech draft input by the user.

[0179] The modified target speech is stored.

[0180] In this embodiment, there is no specific limitation on the storage method of the modified target speech. For example, blockchain storage, local database storage, cloud server storage, etc. can be used.

[0181] The present application displays the target speech; then determines whether a modification request for the target speech is received triggered by a user; if so, receives the editing operation for the target speech input by the user; then performs corresponding modification processing on the target speech based on the editing operation to obtain a modified target speech; and subsequently stores the modified target speech. After completing the automatic generation of the target speech, the present application will also display the target speech for the corresponding user to review, and will intelligently perform corresponding modification processing on the target speech according to the editing operation for the target speech input by the user, so as to obtain a speech that meets the user's personal style and needs, thereby ensuring the accuracy of the modified target speech and effectively improving the user experience. In addition, the generated modified target speech will also be stored to ensure the data security of the modified target speech.

[0182] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0183] It should be emphasized that in order to further ensure the privacy and security of the above-mentioned target speech, the above-mentioned target speech can also be stored in a node of a blockchain.

[0184] The blockchain referred to in this application is a new application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanism, encryption algorithm, etc. Blockchain is essentially a decentralized database, a string of data blocks generated by cryptographic methods. Each data block contains a batch of network transaction information, which is used to verify the validity of its information (anti-counterfeiting) and generate the next block. Blockchain can include the underlying blockchain platform, platform product service layer, and application service layer.

[0185] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through computer-readable instructions, and the computer-readable instructions can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the aforementioned storage medium can be a non-volatile storage medium such as a disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0186] It should be understood that, although the steps in the flowchart of the accompanying drawings are displayed in sequence as indicated by the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.

[0187] Further references Figure 3 , as a response to the above Figure 2 The present application provides an embodiment of a document generation device based on artificial intelligence, and the device embodiment is similar to Figure 2 Corresponding to the method embodiment shown, the device can be specifically applied to various electronic devices.

[0188] like Figure 3 As shown, the artificial intelligence-based document generation device 300 described in this embodiment includes: a first acquisition module 301, a recognition module 302, a first extraction module 303, a second extraction module 304, a conversion module 305, a first determination module 306 and a processing module 307.

[0189] in:

[0190] The first acquisition module 301 is used to acquire a PPT file; wherein the PPT file includes one or more slides;

[0191] The recognition module 302 is used to perform title recognition processing on the PPT file based on a preset title recognition model to obtain corresponding title data;

[0192] The first extraction module 303 is used to perform paragraph extraction processing on the PPT file based on a preset paragraph extraction model to obtain corresponding paragraph text data;

[0193] The second extraction module 304 is used to perform image data extraction processing on the PPT file based on a preset processing library to obtain corresponding image data and chart data;

[0194] The conversion module 305 is used to perform format conversion processing on the picture data to obtain a first image file, and to perform format conversion processing on the chart data to obtain a second image file;

[0195] A first determination module 306 is used to determine a placeholder matching the data to be filled in from a preset speech template; wherein the data to be filled in includes the title data, the paragraph text data, the first image file and the second image file;

[0196] The processing module 307 is used to fill the data to be filled into the placeholder in the speech template that matches the filled data to obtain the corresponding target speech.

[0197] In some optional implementations of this embodiment, the conversion module 305 includes:

[0198] A saving submodule, used for saving the image data as a temporary file;

[0199] A first processing submodule, used for copying the temporary file to a preset new folder to obtain a corresponding designated file, and using the designated file as the first image file;

[0200] A screenshot submodule, used to perform screenshot processing on the chart data based on a preset screenshot tool to obtain corresponding screenshot data;

[0201] The second processing submodule is used to save the screenshot data into a file format to obtain a corresponding screenshot file, and use the screenshot file as the second image file.

[0202] In some optional implementations of this embodiment, the first determining module 306 includes:

[0203] A calling submodule, used for calling the speech template;

[0204] A first determination submodule, used to determine the title placeholder, text placeholder, picture placeholder and chart placeholder in the speech template based on a preset recognition rule;

[0205] A second determining submodule, configured to use the title placeholder as a first placeholder matching the title data;

[0206] A third determination submodule, configured to use the text placeholder as a second placeholder matching the paragraph text data;

[0207] a fourth determining submodule, configured to use the picture placeholder as a third placeholder matching the first image file;

[0208] The fifth determining submodule is configured to use the chart placeholder as a fourth placeholder matching the second image file.

[0209] In some optional implementations of this embodiment, the processing module 307 includes:

[0210] A first filling submodule, used to fill the title data into the first placeholder in the speech template;

[0211] A second filling submodule, used for filling the paragraph text data into the second placeholder;

[0212] A third filling submodule, used for filling the first image file into the third placeholder;

[0213] A fourth filling submodule, used for filling the second image file into the fourth placeholder to obtain a filled speech template;

[0214] The sixth determination submodule is used to use the filled speech template as the target speech.

[0215] In some optional implementations of this embodiment, the third filling submodule includes:

[0216] A first acquiring unit, configured to acquire area size data of the third placeholder;

[0217] A first adjustment unit, configured to perform a size adjustment process on the first image file based on the region size data to obtain a corresponding third image file;

[0218] A second acquisition unit, used to acquire style setting information of the speech template;

[0219] A second adjustment unit, configured to perform format adjustment processing on the third image file based on the style setting information to obtain a corresponding fourth image file;

[0220] an alignment unit, configured to perform alignment processing on the fourth image file to obtain a corresponding fifth image file;

[0221] A filling unit is used to fill the fifth image file into the third placeholder.

[0222] In some optional implementations of this embodiment, the document generation device based on artificial intelligence further includes:

[0223] A second acquisition module is used to acquire pre-collected designated text data; wherein the designated text data is data that has been titled;

[0224] A partitioning module, used for partitioning the specified sample data into a training set and a test set;

[0225] The third acquisition module is used to acquire a preset classification model;

[0226] A training module, used to train the classification model based on the training set using a supervised learning method to obtain a trained specified classification model;

[0227] An evaluation module, used for performing model evaluation on the specified classification model based on the test set;

[0228] The second determination module is used to use the specified classification model as the title recognition model if the specified classification model passes the model evaluation.

[0229] In some optional implementations of this embodiment, the document generation device based on artificial intelligence further includes:

[0230] A display module, used for displaying the target speech;

[0231] A judgment module, used to judge whether a modification request for the target speech script triggered by a user is received;

[0232] A receiving module, configured to receive an editing operation input by the user for the target speech;

[0233] A modification module, used for performing corresponding modification processing on the target speech draft based on the editing operation to obtain a modified target speech draft;

[0234] A storage module is used to store the modified target speech manuscript.

[0235] To solve the above technical problems, the present application also provides a computer device. Figure 4 , Figure 4 This is a basic structural block diagram of the computer device in this embodiment.

[0236] The computer device 4 includes a memory 41, a processor 42, and a network interface 43 that are interconnected through a system bus. It should be noted that the figure only shows a computer device 4 with components 41-43, but it should be understood that it is not required to implement all the components shown, and more or fewer components can be implemented instead. Among them, those skilled in the art can understand that the computer device here is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to microprocessors, application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital processors (DSPs), embedded devices, etc.

[0237] The computer device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The computer device may interact with a user through a keyboard, a mouse, a remote controller, a touch pad, or a voice control device.

[0238] The memory 41 includes at least one type of readable storage medium, and the readable storage medium includes flash memory, hard disk, multimedia card, card-type memory (for example, SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 41 can be an internal storage unit of the computer device 4, such as a hard disk or memory of the computer device 4. In other embodiments, the memory 41 can also be an external storage device of the computer device 4, such as a plug-in hard disk equipped on the computer device 4, a smart memory card (Smart Med ia Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. Of course, the memory 41 can also include both the internal storage unit of the computer device 4 and its external storage device. In this embodiment, the memory 41 is generally used to store the operating system and various application software installed on the computer device 4, such as computer-readable instructions of the manuscript generation method based on artificial intelligence, etc. In addition, the memory 41 can also be used to temporarily store various types of data that have been output or are to be output.

[0239] The processor 42 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chips in some embodiments. The processor 42 is generally used to control the overall operation of the computer device 4. In this embodiment, the processor 42 is used to run computer-readable instructions or process data stored in the memory 41, such as computer-readable instructions for running the artificial intelligence-based manuscript generation method.

[0240] The network interface 43 may include a wireless network interface or a wired network interface. The network interface 43 is generally used to establish a communication connection between the computer device 4 and other electronic devices.

[0241] The present application also provides another embodiment, namely, providing a computer-readable storage medium, wherein the computer-readable storage medium stores computer-readable instructions, and the computer-readable instructions can be executed by at least one processor to enable the at least one processor to perform the steps of the above-mentioned artificial intelligence-based document generation method.

[0242] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0243] Obviously, the embodiments described above are only some embodiments of the present application, rather than all embodiments. The preferred embodiments of the present application are given in the accompanying drawings, but they do not limit the patent scope of the present application. The present application can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present application more thorough and comprehensive. Although the present application is described in detail with reference to the aforementioned embodiments, for those skilled in the art, it is still possible to modify the technical solutions recorded in the aforementioned specific implementation methods, or to perform equivalent replacement of some of the technical features therein. Any equivalent structure made using the contents of the specification and drawings of this application, directly or indirectly used in other related technical fields, is similarly within the scope of patent protection of this application.

Claims

1. A manuscript generation method based on artificial intelligence, characterized in that: The steps include: Obtain a PPT file; wherein the PPT file includes one or more slides; Performing title recognition processing on the PPT file based on a preset title recognition model to obtain corresponding title data; Perform paragraph extraction processing on the PPT file based on a preset paragraph extraction model to obtain corresponding paragraph text data; Performing image data extraction processing on the PPT file based on a preset processing library to obtain corresponding image data and chart data; Performing format conversion processing on the picture data to obtain a first image file, and performing format conversion processing on the chart data to obtain a second image file; Determine a placeholder matching the data to be filled in from a preset speech template; wherein the data to be filled in includes the title data, the paragraph text data, the first image file and the second image file; The data to be filled is filled into the placeholder in the speech template that matches the filled data to obtain the corresponding target speech.

2. The method for generating a manuscript based on artificial intelligence according to claim 1, characterized in that: The steps of converting the image data into a first image file and converting the chart data into a second image file specifically include: Saving the image data as a temporary file; Copying the temporary file to a preset new folder to obtain a corresponding designated file, and using the designated file as the first image file; Performing screenshot processing on the chart data based on a preset screenshot tool to obtain corresponding screenshot data; The screenshot data is saved in a file format to obtain a corresponding screenshot file, and the screenshot file is used as the second image file.

3. The method for generating a manuscript based on artificial intelligence according to claim 1, characterized in that: The step of determining a placeholder matching the data to be filled from the preset speech template specifically includes: Calling the speech template; Determine the title placeholder, text placeholder, picture placeholder and chart placeholder in the speech template based on preset recognition rules; using the title placeholder as a first placeholder that matches the title data; Using the text placeholder as a second placeholder that matches the paragraph text data; Using the picture placeholder as a third placeholder matching the first image file; The chart placeholder is used as a fourth placeholder matched with the second image file.

4. The method for generating a manuscript based on artificial intelligence according to claim 3, characterized in that: The step of filling the data to be filled into the placeholder in the speech template that matches the filled data to obtain the corresponding target speech specifically includes: Filling the title data into the first placeholder in the speech template; Filling the paragraph text data into the second placeholder; Filling the first image file into the third placeholder; Filling the second image file into the fourth placeholder to obtain a filled speech template; The filled speech template is used as the target speech.

5. The method for generating a manuscript based on artificial intelligence according to claim 4, characterized in that: The step of filling the first image file into the third placeholder specifically includes: Obtaining area size data of the third placeholder; Performing a size adjustment process on the first image file based on the region size data to obtain a corresponding third image file; Obtaining style setting information of the speech template; Performing format adjustment processing on the third image file based on the style setting information to obtain a corresponding fourth image file; Performing alignment processing on the fourth image file to obtain a corresponding fifth image file; The third placeholder is filled with the fifth image file.

6. The method for generating a manuscript based on artificial intelligence according to claim 1, characterized in that: Before the step of performing title recognition processing on the PPT file based on a preset title recognition model to obtain corresponding title data, the method further includes: Acquire pre-collected designated text data; wherein the designated text data is data that has been titled; Dividing the designated sample data into a training set and a test set; Get the preset classification model; Based on the training set, the classification model is trained using a supervised learning method to obtain a trained specified classification model; Performing model evaluation on the specified classification model based on the test set; If the designated classification model passes the model evaluation, the designated classification model is used as the title recognition model.

7. The method for generating a manuscript based on artificial intelligence according to claim 1, characterized in that: After the step of filling the data to be filled into the placeholder in the speech template that matches the filled data to obtain the corresponding target speech, the method further includes: Presenting the target speech; Determining whether a modification request for the target speech script is received, triggered by a user; If yes, receiving an editing operation input by the user for the target speech; Performing corresponding modification processing on the target speech draft based on the editing operation to obtain a modified target speech draft; The modified target speech is stored.

8. A document generation device based on artificial intelligence, characterized in that: include: A first acquisition module is used to acquire a PPT file; wherein the PPT file includes one or more slides; A recognition module, used to perform title recognition processing on the PPT file based on a preset title recognition model to obtain corresponding title data; A first extraction module is used to perform paragraph extraction processing on the PPT file based on a preset paragraph extraction model to obtain corresponding paragraph text data; A second extraction module is used to perform image data extraction processing on the PPT file based on a preset processing library to obtain corresponding image data and chart data; a conversion module, configured to perform format conversion processing on the image data to obtain a first image file, and to perform format conversion processing on the chart data to obtain a second image file; A first determining module is used to determine a placeholder matching the data to be filled in from a preset speech template; wherein the data to be filled in includes the title data, the paragraph text data, the first image file and the second image file; The processing module is used to fill the data to be filled into the placeholder in the speech template that matches the filled data to obtain the corresponding target speech.

9. A computer device comprising a memory and a processor, wherein the memory stores computer-readable instructions, and when the processor executes the computer-readable instructions, the steps of the document generation method based on artificial intelligence as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-readable instructions, which, when executed by a processor, implement the steps of the artificial intelligence-based document generation method as described in any one of claims 1 to 7.

Citation Information

Patent Citations

  • An intelligent PPT analysis report generation method

    CN109634575A

  • PPT automatic generation method based on AI

    CN116975329A

  • Data processing method and device, electronic equipment and computer readable storage medium

    CN117725262A

  • System and method of testing multiple lidars

    KR1020250155260A

Cited By

  • Product nameplate drawing generation method and device

    CN119647442A

  • Video copywriting generation method and device, equipment and medium

    CN119670716A

  • PPT and speech draft generation method based on artificial intelligence

    CN119988653A

  • Presentation manuscript intelligent dubbing method and device, computer equipment and storage medium

    CN121191492A