Conference material generation method, device and storage medium

Abstract

The present application discloses a conference material generation method, device and storage medium. The present application relates to the field of data processing technology. The method generates a text model interaction instruction according to the requirements of the conference material, and inputs the text model interaction instruction into at least one large text model. Then, according to the conference material text information output by the large text model based on the text model interaction instruction, the target text information is generated. Then, the target text information is input into the large text model to generate conference material illustrations. Finally, the conference material is generated according to the target text information, the conference material illustrations and the conference template. The above method processes materials through multiple large text models, and then summarizes the processed materials according to the AI ​​large model to generate conference materials, which can effectively reduce the time users spend on preparing conference materials.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a method, device and storage medium for generating conference materials. Background Art

[0002] Meetings are one of the main platforms for synchronizing information and making decisions within an enterprise. A complete meeting process includes pre-meeting organization and material preparation, management and control during the meeting, post-meeting summary and follow-up of to-do items. Currently, for pre-meeting material preparation, it is usually the staff who organizes the materials and completes the production of the meeting PPT. In the actual process, the staff completes the pre-meeting material preparation. Due to the heavy amount of enterprise data and the cumbersome preparation of meeting materials, the relevant solutions cannot effectively reduce the time users spend on preparing meeting materials.

[0003] The above contents are only used to assist in understanding the technical solution of the present application and do not constitute an admission that the above contents are prior art. Summary of the invention

[0004] The present application provides a conference material generation method, device and storage medium, aiming to solve the problem that related solutions cannot effectively reduce the time spent by users in preparing conference materials.

[0005] To achieve the above purpose, the present application provides a method for generating conference materials, which comprises the following steps:

[0006] Generate text model interaction instructions according to the meeting material requirements, and input the text model interaction instructions into at least one text macro model;

[0007] Generate target text information according to the conference material text information output by the text big model based on the text model interaction instruction;

[0008] Input the target text information into the Wenshengtu model to generate conference material illustrations;

[0009] The conference materials are generated according to the target text information, the conference material illustrations and the conference template.

[0010] In one embodiment, after the step of generating a text model interaction instruction according to the conference material requirements and inputting the text model interaction instruction into at least one text macro model, the method further includes:

[0011] After the text model interaction instruction is input into each of the text large models, an enterprise knowledge base and a target data source are obtained;

[0012] The text big model is based on the enterprise knowledge base and the target data source, and generates the conference material text information according to the text model interaction instructions.

[0013] In one embodiment, the step of generating target text information according to the conference material text information output by the text macro model based on the text model interaction instruction includes:

[0014] Obtaining a preset weight value of each of the text macro models, wherein the preset weight value is set according to the text generation capability of the text macro model in the target domain;

[0015] According to the preset weight value, the conference material text information output by each of the text models is fused to generate the target text information.

[0016] In one embodiment, the step of generating the conference materials according to the target text information, the conference material illustrations and the conference template includes:

[0017] Get the pre-stored meeting template;

[0018] Inputting the conference template, the target text information and the conference material picture into a conference material automatic generation tool to generate a conference material file in a first format;

[0019] The format of the conference material file is converted according to a format conversion tool to generate the conference material in a second format.

[0020] In one embodiment, before the step of obtaining the conference material requirements and inputting the conference material requirements into at least one text macromodel, the step further includes:

[0021] Obtaining each of the text macro models;

[0022] Construct the AI ​​big model according to each of the text big models;

[0023] Acquire target data in the enterprise knowledge base, and fine-tune the AI ​​big model based on the target data.

[0024] In one embodiment, before the step of acquiring target data from the enterprise knowledge base and fine-tuning the AI ​​big model based on the target data, the step further includes:

[0025] Integrate with enterprise office systems, various business management systems and email systems through data synchronization tools;

[0026] The target data acquired in real time by the data synchronization tool is stored in the enterprise knowledge base.

[0027] In one embodiment, after the step of acquiring target data from the enterprise knowledge base and fine-tuning the AI ​​big model based on the target data, the step further includes:

[0028] When the conference system detects a target request of a user, obtaining a conference prompt corresponding to the user, wherein the conference prompt includes a conference topic, participants, and their backgrounds;

[0029] A conference template is determined according to the conference prompt, and the conference template is sent to a conference material generation system.

[0030] In one embodiment, after the step of determining a conference template according to the conference prompt and sending the conference template to the conference material generation application, the method further includes:

[0031] After the conference material generation system receives the conference template, it obtains data corresponding to the conference topic from various systems in real time, and passes the data, the conference prompts and the conference material template as parameters for making conference materials to the AI ​​big model;

[0032] The steps of obtaining conference material requirements, inputting the conference material requirements into at least one text macromodel, and feeding back the generated conference materials to the corresponding user terminal are performed.

[0033] In addition, to achieve the above-mentioned purpose, the present application also provides a conference material generation device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, and the computer program is configured to implement the steps of the conference material generation method as described above.

[0034] In addition, to achieve the above-mentioned purpose, the present application also provides a storage medium, which is a computer-readable storage medium, and a computer program is stored on the computer-readable storage medium. When the computer program is executed by a processor, the steps of the conference material generation method as described above are implemented.

[0035] The present application provides a conference material generation method, a conference material generation device and a storage medium. According to the requirements of the conference materials, a text model interaction instruction is generated, and the text model interaction instruction is input into at least one text big model. Then, according to the conference material text information output by the text big model based on the text model interaction instruction, target text information is generated, and the target text information is input into a text-image big model to generate conference material illustrations. Finally, according to the target text information, the conference material illustrations and the conference template, the conference materials are generated. The present application processes materials through multiple text big models, and then summarizes the processed materials according to the AI ​​big model to generate conference materials, which can effectively reduce the time spent by users on preparing conference materials. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0037] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0038] Figure 1 This is a flow chart of the first embodiment of the method for generating conference materials of the present application;

[0039] Figure 2 This is a flow chart of a second embodiment of the method for generating conference materials of the present application;

[0040] Figure 3 This is a flow chart of the third embodiment of the method for generating conference materials of the present application;

[0041] Figure 4 This is a flow chart of a fourth embodiment of the method for generating conference materials of the present application;

[0042] Figure 5 This is a schematic diagram of the architecture of the hardware operating environment of the conference material generation device involved in the embodiment of the present application.

[0043] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0044] It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0045] In order to better understand the above technical solution, exemplary embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present application and to fully convey the scope of the present application to those skilled in the art.

[0046] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0047] The main solution of this application is: generate text model interaction instructions according to the requirements of conference materials, and input the text model interaction instructions into at least one large text model; generate target text information according to the conference material text information output by the large text model based on the text model interaction instructions; input the target text information into the text and image large model to generate conference material illustrations; generate the conference materials according to the target text information, the conference material illustrations and the conference template.

[0048] Meetings are one of the main platforms for synchronizing information and making decisions within an enterprise. A complete meeting process includes pre-meeting organization and material preparation, management and control during the meeting, post-meeting summary and follow-up of to-do items. Currently, for pre-meeting material preparation, it is usually the staff who organizes the materials and completes the production of the meeting PPT. In the actual process, the staff completes the pre-meeting material preparation. Due to the heavy amount of enterprise data and the cumbersome preparation of meeting materials, the relevant solutions cannot effectively reduce the time users spend on preparing meeting materials.

[0049] By generating text model interaction instructions according to the requirements of conference materials, and inputting the text model interaction instructions into at least one text model, and then generating target text information according to the conference material text information output by the text model interaction instructions based on the text model model, and inputting the target text information into the text and image model to generate conference material illustrations, and finally generating the conference materials according to the target text information, the conference material illustrations and the conference template. This application processes materials through multiple text models, and then summarizes the processed materials according to the AI ​​model to generate conference materials, which can effectively reduce the time users spend on preparing conference materials.

[0050] Embodiment 1

[0051] Please refer to Figure 1 In the first embodiment, the conference material generation method includes the following steps:

[0052] Step S10: Generate text model interaction instructions according to the requirements of the conference materials, and input the text model interaction instructions into at least one large text model.

[0053] In this embodiment, the processing action is performed by the AI ​​big model, and the meeting material requirements can be data uploaded by the user based on the upload interface, including information such as the meeting topic, participants and their backgrounds. When the upload interface detects the meeting material requirements uploaded by the user, the front end will feed back the meeting material requirements to the AI ​​big model, so that the AI ​​big model can obtain the meeting material requirements. After obtaining the meeting material requirements, the AI ​​big model generates text model interaction instructions based on the preset interaction instruction generation method and according to the meeting material requirements. Then the text model interaction instructions are input into multiple different text big models at the same time.

[0054] Specifically, the text model can be a GPT (Generative Pre-Trained Transformer) model such as ChatGLM2-6B, Baichuan-13B and Wenxinyiyan. ChatGLM2-6B is a Chinese-English bilingual dialogue model released by Zhipu AI and Tsinghua KEG Laboratory, and Baichuan-13B is an open source large-scale language model project developed by Baichuan Intelligence, which contains 13 billion parameters. In addition, the GPT model is an Internet-based deep learning model that can be trained with data and generate text.

[0055] As an optional implementation method for generating text model interaction instructions, first determine the meeting material requirements according to the meeting material requirements, and determine the specific content and objectives of the meeting materials, such as question and answer, summary, article writing, etc. Define the task type based on the natural language processing algorithm and the meeting material requirements, such as generating an article, story or paragraph and generating a concise summary. Finally, construct specific interaction instructions based on the task type, for example, the corresponding instruction for the question and answer system is "answer the following questions about climate change." By generating text model interaction instructions, the accuracy of the model output text is improved.

[0056] Step S20: Generate target text information according to the conference material text information output by the text macro model based on the text model interaction instruction.

[0057] In this embodiment, the AI ​​big model will construct an API (Application Programming Interface) request, which carries the conference material requirements. By sending the API request to the corresponding text big model, the text model interaction instructions corresponding to the conference material requirements are input into the text big model. After the conference material requirements are input into each of the text big models, each text big model will automatically generate and output the conference material text information according to the input conference material requirements. Then collect the conference material text information obtained from each big model, and give the results output by different text models as input to the AI ​​big model. The AI ​​big model analyzes the conference material text information and selects the corresponding part from each conference material text information. The screening method can be manual review or natural language processing technology to complete the screening. Then, the output results of multiple text models are fused through the AI ​​big model, and the final conference material file information, that is, the target text information, is output. In addition, when the results are fused, different weights will be set according to the text generation capabilities of each text big model in a specific field.

[0058] Optionally, in this embodiment, after the step of generating a text model interaction instruction according to the conference material requirements and inputting the text model interaction instruction into at least one text macro model, the step further includes:

[0059] After the text model interaction instructions are input into each of the text big models, the enterprise knowledge base and the target data source are obtained; the text big model generates the conference material text information based on the enterprise knowledge base and the target data source and according to the text model interaction instructions.

[0060] Specifically, after the text model interaction instruction is input into each of the text large models, the enterprise knowledge base and the target data source are obtained, and the target data source is data obtained on the Internet through multiple interfaces. These text large models use the data on the enterprise knowledge base and the Internet to generate and output independent conference material text information. It can be understood that the enterprise knowledge base first collects internal documents, reports, project materials, etc., and then processes and summarizes the collected materials and saves them in the established enterprise knowledge base. Among them, the internal data acquisition channel can be data obtained from the enterprise OA (Office Automation) system, various business management systems and mail systems.

[0061] As an optional implementation method for the AI ​​big model to fuse the text information of conference materials, the conference material text information is directly input and fused, by constructing a long text prompt containing all the original conference material text information, and then requesting the big model to generate a comprehensive text output according to the prompt. It is also possible to fuse two paragraphs of text first, and then fuse the fused text with other texts, gradually fusing all the text information. Or the original conference material text information can be classified according to content category or theme. Each category of text is fused separately. Finally, all categories of text are merged into a whole.

[0062] Optionally, in this embodiment, the step of generating target text information according to the conference material text information output by the text big model based on the text model interaction instruction includes:

[0063] Obtaining preset weight values ​​of each of the text models, wherein the preset weight values ​​are set according to the text generation capability of the text model in the target field; the AI ​​model fuses the conference material text information output by each of the text models according to the preset weight values ​​to generate the target text information.

[0064] Specifically, before fusing the conference material text information based on the AI ​​big model, different weight values ​​are set according to the text generation capabilities of each text big model in a specific field. Then, when the AI ​​big model needs to process the conference material text information, the preset weight values ​​of each text big model are obtained, and during fusion, the conference material text information output by each text model is fused in a corresponding proportion according to the preset weight values ​​to generate and output the target text information. Among them, the preset weight values ​​are used to characterize the importance and priority of each part of the content in the fused text.

[0065] Step S30: input the target text information into the text-based image model to generate conference material illustrations.

[0066] In this embodiment, the text information of the final conference materials is input into a large model that is good at text diagrams, and illustrations are generated for the key chapters of the conference materials, and finally conference materials with text and illustrations are generated. Specifically, the generated target text information is first analyzed, the parts that need illustrations are identified, and keywords and concepts are extracted from the target text, where the keywords and concepts are used to guide the large model of text diagrams to generate images. An API request is constructed using the extracted keywords and concepts, and the request is sent to the large model of text diagrams. An API request is then sent to the large model of text diagrams, and the generated image is received. The image, i.e., the illustrations for the conference materials, is then received from the large model of text diagrams, and integrated into the conference materials.

[0067] Step S40: Generate the conference materials according to the target text information, the conference material illustrations and the conference template.

[0068] In this embodiment, a conference material automatic generation tool is provided in the conference material generation system. By inputting the conference template, conference material text information and conference material illustrations into the conference material automatic generation tool, the conference material automatic generation tool will generate conference materials in a corresponding format based on the target text information, the conference material illustrations and the conference template.

[0069] Optionally, in this embodiment, the step of generating the conference materials according to the target text information, the conference material illustrations and the conference template includes:

[0070] Obtain a pre-stored conference template; input the conference template, the target text information and the conference material image into a conference material automatic generation tool to generate a conference material file in a first format; convert the format of the conference material file according to a format conversion tool to generate the conference material in a second format.

[0071] Specifically, after the conference material pictures are generated, the conference material generation system will obtain the pre-stored templates and determine the matching conference templates according to the conference material requirements. Then the conference template, the target text information and the conference material pictures are input into the conference material automatic generation tool, which will generate a conference material file in markdown (lightweight markup language) format, and then generate the final conference material in ppt format through the built-in markdown format file conversion tool to ppt (PowerPoint, presentation software).

[0072] In the technical solution provided in this embodiment, by obtaining the requirements of conference materials, the conference materials requirements are input into at least one text big model, and then the conference materials text information output by each of the text big models is obtained, and then the conference materials text information is fused and processed based on the AI ​​big model to generate target text information, wherein the AI ​​big model is constructed by at least one text big model, and generates conference materials illustrations based on the target text information and the text-image big model, and finally generates the conference materials based on the target text information, the conference materials illustrations and the conference template. This implementation scheme processes materials through multiple text big models, and then summarizes the processed materials according to the AI ​​big model to generate conference materials, which can effectively reduce the time spent by users on preparing conference materials.

[0073] Embodiment 2

[0074] Please refer to Figure 2 In the second embodiment, before the step S10, the following steps are further included:

[0075] Step S50: Obtain each of the large text models.

[0076] Step S60: construct the AI ​​big model according to each of the text big models.

[0077] In this embodiment, the processing action is performed by the conference material generation system, in which at least one text big model API interface is pre-stored. By acquiring these API interfaces, the acquisition of at least one text big model is completed. A suitable target text big model is determined among these text models, and multiple text big models are combined for use according to actual needs, for example, one text big model is responsible for generating content, and another text big model is responsible for verifying and optimizing output quality. Alternatively, multiple big models are integrated through the API interface of the big model to obtain an AI big model.

[0078] Specifically, before building an AI big model, it is necessary to determine the functions of the AI ​​big model, such as summarizing the output content of multiple text big models, or automatically generating meeting minutes, summaries, and PPTs.

[0079] Step S70: Obtain target data from the enterprise knowledge base, and fine-tune the AI ​​big model based on the target data.

[0080] In this embodiment, the target data includes a training set, a validation set, and a test set, which are target data obtained by preprocessing the data in the enterprise knowledge base. After obtaining the target data, set the hyperparameters in the fine-tuning process, such as learning rate, batch size, training rounds, etc., and select a suitable loss function according to the task type, such as the cross entropy loss function is suitable for text generation tasks. Use the training set in the target data of the enterprise internal knowledge base to fine-tune the model, and monitor the training process loss value and validation loss during fine-tuning to ensure good model performance. And evaluate the performance of the model based on the validation set, and adjust the hyperparameters and improve the data processing flow of the large model based on the results. Finally, evaluate the generalization ability of the model based on the test set.

[0081] Optionally, the data in the enterprise knowledge base can be preprocessed by first converting all the data into a unified format, such as plain text or JSON format, and removing duplicate content to avoid introducing bias during training. Then remove meaningless information in the text, such as headers, footers, watermarks, non-text elements, etc., and continue to segment the text, splitting long sentences into short sentences and phrases. Then convert the text into a form that the model can understand, using specific markers to represent each word or fragment in the text. Finally, after completing the above processing, the processed text is determined as the target data, and the AI ​​large model is fine-tuned. In addition, the preprocessed target data should be divided into training sets, validation sets, and test sets, as well as annotated data, and the generated meeting minutes should be scored or classified to improve the efficiency of model fine-tuning.

[0082] Optionally, in this embodiment, before step S70, the following steps are further included:

[0083] Integrate with the enterprise office system, various business management systems and email system through data synchronization tools; save the target data obtained in real time by the data synchronization tools in the enterprise knowledge base.

[0084] Specifically, through data synchronization tools, it is integrated with the enterprise OA system, various business management systems, email systems and other systems. Then, data from various systems can be obtained in real time through data synchronization tools and saved in the enterprise knowledge base to ensure the effectiveness of the materials used in the generated meeting materials.

[0085] In addition, after the fine-tuned AI big model is determined, the fine-tuned AI big model is deployed in the conference material generation system, and the capabilities of the AI ​​model are encapsulated into an API interface for system calls of other user terminals. And the API interface encapsulated by the AI ​​big model is integrated with the enterprise online document editing system and the enterprise conference system to meet the conference material generation needs of multiple systems.

[0086] In the technical solution provided in this embodiment, by obtaining each of the text big models and constructing the AI ​​big model based on each of the text big models, then obtaining target data in the enterprise knowledge base, and fine-tuning the AI ​​big model based on the target data, the accuracy of the AI ​​big model in generating conference materials is improved.

[0087] Embodiment 3

[0088] Please refer to Figure 3 In the third embodiment, after step S70, the method further includes:

[0089] Step S80: When the conference system detects the target request of the user, the conference prompt corresponding to the user is obtained, wherein the conference prompt includes the conference topic, participants and their backgrounds.

[0090] Step S90: determining a conference template according to the conference prompt, and sending the conference template to a conference material generation system.

[0091] In this embodiment, before the conference initiator initiates the conference, when the user needs to prepare conference materials, the user will upload a target request to the conference system, and then the user will upload the conference reminder based on the front-end of the conference system, including the conference topic, participants, and personnel background. The conference system searches for a matching conference template in the database according to the conference reminder, and sends the conference template to the conference material generation system.

[0092] Optionally, in this embodiment, after step S90, the following further steps are included:

[0093] After the conference material generation system receives the conference template, it obtains data corresponding to the conference topic from various systems in real time, and passes the data, the conference prompts and the conference material template as parameters for making conference materials to the AI ​​big model. It executes the steps of obtaining conference material requirements, inputting the conference material requirements into at least one text big model, and feeding back the generated conference materials to the corresponding user end.

[0094] Specifically, after the conference material generation system receives the conference template, the conference material generation application obtains data related to the conference topic from various systems in real time, and passes these data, conference prompts and conference material templates as parameters for making conference materials to the AI ​​big model. Then, the conference material generation system, based on these parameters, executes the steps of obtaining conference material requirements, inputting the conference material requirements into at least one text big model, and generating conference materials. Finally, the AI ​​big model returns the generated conference materials to the conference material generation system, and the conference material generation system returns them to the user through the enterprise online conference editing system and the conference system.

[0095] Embodiment 4

[0096] Please refer to Figure 4 In the fourth embodiment, as Figure 4 As shown, in this embodiment, the AI ​​big model is constructed by multiple text big models, and after construction, the model is fine-tuned by processing the data in the internal knowledge base of the enterprise. After fine-tuning the model, the AI ​​big model is encapsulated into an API and connected to the automatic generation application of conference materials. The automatic generation application of conference materials is connected to the OA system, various business management system mail systems and other systems, and the data generated by these systems can be obtained in real time. These data can also be used as basic data for fine-tuning the AI ​​big model after preprocessing. In addition, the automatic generation application of conference materials is deployed in the conference material generation system and connected to the enterprise online document editing system and the enterprise conference system. Before and during the meeting, the user can send a conference material generation request to the automatic generation application of conference materials through the API. The automatic generation application of conference materials determines the conference material requirements according to the conference material generation request uploaded by the user, and passes the conference material requirements to the AI ​​big model through the API. The AI ​​big model executes the steps in the conference material generation method to generate conference materials. Finally, the conference material generation system feeds back the conference materials to the user.

[0097] In the technical solution provided in this embodiment, when the conference system detects the user's target request, the conference prompt corresponding to the user is obtained, wherein the conference prompt includes the meeting topic, participants and their backgrounds, a conference template is determined according to the conference prompt, and the conference template is sent to the conference material generation system, thereby reducing the time spent by the user on preparing conference materials.

[0098] Since the system introduced in the embodiment of the present application is a system used to implement the method of the embodiment of the present application, based on the method introduced in the embodiment of the present application, the person skilled in the art can understand the specific structure and deformation of the system, so it is not repeated here. All systems used in the method of the embodiment of the present application belong to the scope of protection of this application.

[0099] The present application provides a conference material generation device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the conference material generation method in the above-mentioned embodiment one.

[0100] Reference below Figure 5 , which shows a schematic diagram of the structure of a conference material generation device suitable for implementing the embodiment of the present application. The conference material generation device in the embodiment of the present application may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 5 The conference material generating device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0101] like Figure 5As shown, the conference material generating device may include a processing device 1001 (e.g., a core processor, a graphics processor, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM: Read Only Memory) 1002 or a program loaded from a storage device 1003 to a random access memory (RAM: Random Access Memory) 1004. In RAM1004, various programs and data required for the operation of the conference material generating device are also stored. The processing device 1001, ROM1002, and RAM1004 are connected to each other through a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Generally, the following systems can be connected to the I / O interface 1006: an input device 1007 including, for example, a touch screen, a touch pad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output device 1008 including, for example, a liquid crystal display (LCD: Liquid Crystal Display), a speaker, a vibrator, etc.; a storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the conference material generation device to communicate with other devices wirelessly or by wire to exchange data. Although the conference material generation device with various systems is shown in the figure, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems can be implemented or have alternatively.

[0102] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0103] The conference material generation device provided by the present application adopts the conference material generation method in the above embodiment, which can solve the technical problem that the relevant solutions cannot effectively reduce the time spent by users on preparing conference materials. Compared with the prior art, the beneficial effects of the conference material generation device provided by the present application are the same as the beneficial effects of the conference material generation method provided by the above embodiment, and the other technical features in the conference material generation device are the same as the features disclosed in the method of the previous embodiment, which will not be repeated here.

[0104] It should be understood that the various parts disclosed in this application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0105] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0106] The present application provides a computer-readable storage medium having computer-readable program instructions (ie, computer programs) stored thereon, and the computer-readable program instructions are used to execute the conference material generation method in the above-mentioned embodiment.

[0107] The computer-readable storage medium provided in the present application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared or semiconductor systems, systems or devices, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM: Random Access Memory), a read-only memory (ROM: Read Only Memory), an erasable programmable read-only memory (EPROM: Erasable Programmable Read Only Memory or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM: CD-Read Only Memory), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, system or device. The program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency: Radio Frequency), etc., or any suitable combination of the above.

[0108] The computer-readable storage medium may be included in the conference material generating device; or may exist independently without being assembled into the conference material generating device.

[0109] The above-mentioned computer-readable storage medium carries one or more programs. When the above-mentioned one or more programs are executed by the conference material generating device, the conference material generating device enables the following: to generate text model interaction instructions according to the requirements of the conference materials, and input the text model interaction instructions into at least one large text model; to generate target text information according to the conference material text information output by the large text model based on the text model interaction instructions; to input the target text information into the text-image large model to generate conference material illustrations; and to generate the conference materials according to the target text information, the conference material illustrations and the conference template.

[0110] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0111] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0112] The modules involved in the embodiments described in this application may be implemented by software or hardware, wherein the name of the module does not constitute a limitation on the unit itself in some cases.

[0113] The readable storage medium provided by the present application is a computer-readable storage medium, which stores computer-readable program instructions (i.e., computer programs) for executing the above-mentioned conference material generation method, and can solve the technical problem that the related solutions cannot effectively reduce the time spent by users on conference material preparation. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided by the present application are the same as the beneficial effects of the conference material generation method provided by the above-mentioned embodiment, and will not be repeated here.

[0114] An embodiment of the present application provides a computer program product, including a computer program, which implements the steps of the above-mentioned conference material generation method when executed by a processor.

[0115] The computer program product provided by the present application can solve the technical problem that related solutions cannot effectively reduce the time spent by users in preparing conference materials. Compared with the prior art, the beneficial effects of the computer program product provided by the embodiment of the present application are the same as the beneficial effects of the conference material generation method provided by the above embodiment, which will not be repeated here.

[0116] The above are only preferred embodiments of the present application, and are not intended to limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent processing scope of the present application.

Claims

1. A method for generating conference materials, characterized in that: The method comprises the following steps: The conference material generation system obtains each text model; The conference material generation system constructs an AI big model based on each of the text big models; The conference material generation system obtains target data from the enterprise knowledge base and fine-tunes the AI ​​big model based on the target data; Determine the meeting material requirements according to the meeting material requirements, and define the task type based on the natural language processing algorithm and according to the meeting material requirements, wherein the meeting material requirements include the meeting topic, participants and their background information; Constructing a text model interaction instruction according to the task type, and inputting the text model interaction instruction into multiple text big models at the same time, wherein the AI ​​big model constructs an API request, and by sending the API request to the corresponding text big models respectively, the text model interaction instruction corresponding to the conference material requirements is input into the text big model; After the text model interaction instruction is input into each of the text large models, an enterprise knowledge base and a target data source are obtained; The text big model is based on the enterprise knowledge base and the target data source, and generates the conference material text information according to the text model interaction instruction; Obtaining a preset weight value of each of the text macro models, wherein the preset weight value is set according to the text generation capability of the text macro model in the target domain; According to the preset weight value, the conference material text information output by each of the text models is merged to generate target text information; Input the target text information into the Wenshengtu model to generate conference material illustrations; The conference materials are generated according to the target text information, the conference material illustrations and the conference template.

2. The method according to claim 1, characterized in that The step of generating the conference materials according to the target text information, the conference material pictures and the conference template comprises: Get the pre-stored meeting template; Inputting the conference template, the target text information and the conference material illustrations into a conference material automatic generation tool to generate a conference material file in a first format; The format of the conference material file is converted according to a format conversion tool to generate the conference material in a second format.

3. The method according to claim 1, characterized in that Before the step of obtaining target data from the enterprise knowledge base and fine-tuning the AI ​​big model based on the target data, the conference material generation system further includes: Integrate with enterprise office systems, various business management systems and email systems through data synchronization tools; The target data acquired in real time by the data synchronization tool is stored in the enterprise knowledge base.

4. The method according to claim 1, characterized in that After the step of obtaining target data from the enterprise knowledge base and fine-tuning the AI ​​big model based on the target data, the conference material generation system further includes: When the conference system detects a target request of a user, obtaining a conference prompt corresponding to the user, wherein the conference prompt includes a conference topic, participants, and their backgrounds; A conference template is determined according to the conference prompt, and the conference template is sent to a conference material generation system.

5. The method according to claim 4, characterized in that After the step of determining a conference template according to the conference prompt and sending the conference template to the conference material generation application, the method further includes: After the conference material generation system receives the conference template, it obtains data corresponding to the conference topic from various systems in real time, and passes the data, the conference prompts and the conference material template as parameters for making conference materials to the AI ​​big model; The steps of obtaining conference material requirements, inputting the conference material requirements into at least one text macromodel, and feeding back the generated conference materials to the corresponding user terminal are performed.

6. A conference material generating device, characterized in that: The conference material generating device comprises: a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the conference material generating method according to any one of claims 1 to 5.

7. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the conference material generation method according to any one of claims 1 to 5 are implemented.

Images