Conference summary generation method and device, electronic equipment and storage medium

Abstract

The invention provides a conference summary generation method and device, electronic equipment and a storage medium, and belongs to the technical field of data processing, and the method comprises the steps: converting multi-modal input data into standardized text data; the text data is input into a pre-trained large language model, conference information output by the large language model is obtained, and the conference information comprises conference issues, decision content, task allocation information and conference participant responsibility division; filling a preset conference summary template with the conference information to obtain a conference summary framework; and inputting the conference summary framework into the generative language model to obtain a summary text output by the generative language model. According to the scheme, the input data of multiple modes are uniformly processed, effective integration of different input sources is ensured and accurately reflected in the generated conference summary, key information in the conference is accurately extracted from the multi-mode input, errors and omissions of manual recording are avoided, the automation level of conference management is improved, and the working efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a method, device, electronic device and storage medium for generating meeting minutes. Background Art

[0002] In the operation process of modern enterprises and organizations, the generation and management of meeting minutes plays a vital role. It is a key link to ensure the smooth progress of projects and improve the efficiency of team collaboration. Meeting minutes not only record the decision-making matters, discussion points and assigned tasks in the meeting, but also provide members who failed to attend the meeting with a way to understand the content of the meeting, ensuring the transparency and consistency of information. By recording meeting minutes in detail and accurately, misunderstandings and conflicts caused by information asymmetry can be effectively avoided, and communication and cooperation between team members can be promoted.

[0003] Traditional meeting minutes usually rely on manual recording and organization, which is not only time-consuming and labor-intensive, but also prone to omissions and misunderstandings.

[0004] Therefore, there is an urgent need for a meeting minutes generation method to solve the problems of time-consuming and labor-intensive traditional manual recording of meeting minutes, as well as the proneness to omissions and misunderstandings. Summary of the invention

[0005] The present invention provides a method, device, electronic device and storage medium for generating meeting minutes, which are used to solve the defects of the prior art in which manual recording of meeting minutes is time-consuming and labor-intensive, and prone to omissions and misunderstandings.

[0006] The present invention provides a method for generating meeting minutes, comprising the following steps: Convert multimodal input data into standardized text data; Inputting the text data into a pre-trained large language model to obtain meeting information output by the large language model, wherein the meeting information includes meeting topics, decision content, task allocation information, and division of responsibilities of participants; Fill the meeting information into a preset meeting minutes template to obtain a meeting minutes framework; The meeting minutes framework is input into a generative language model to obtain a minutes text output by the generative language model.

[0007] According to a method for generating meeting minutes provided by the present invention, the step of inputting the meeting minutes framework into a generative language model to obtain a minutes text output by the generative language model comprises: Receiving the focus point information input by the user, wherein the focus point information is the focus point specified by the user when generating minutes; The meeting minutes framework and the focus information are input into a generative language model to obtain a minutes text output by the generative language model.

[0008] According to a method for generating meeting minutes provided by the present invention, after inputting the meeting minutes framework into a generative language model to obtain the minutes text output by the generative language model, the method further includes: Using a preset proofreading algorithm, verifying the content of the minutes text to obtain a verified minutes text; After receiving the review feedback information input by the user, the verification minutes text is corrected based on the review feedback information to obtain a meeting minutes document.

[0009] According to a method for generating meeting minutes provided by the present invention, after the verification minutes text is corrected based on the review feedback information to obtain the meeting minutes document, the method further includes: Identify the meeting minutes document and extract multiple tasks; The priority of each task is determined based on the dependency relationship of each task, the importance of each task, the urgency of each task and the resource requirement of each task.

[0010] According to a method for generating meeting minutes provided by the present invention, after identifying the meeting minutes document and extracting multiple tasks, the method further includes: Acquire personnel availability information and personnel skill information, wherein the personnel availability information indicates whether the personnel is available, and the personnel skill information includes the skills possessed by the personnel; Based on the personnel availability information and the personnel skill information, an executor is matched for each of the tasks.

[0011] According to a method for generating meeting minutes provided by the present invention, after identifying the meeting minutes document and extracting multiple tasks, the method further includes: Based on the task requirements of each of the tasks, using a preset resource scheduling algorithm to determine resource allocation information for each of the tasks; A task schedule is generated based on the priority of each task and the resource allocation information of each task.

[0012] The present invention also provides a conference minutes generating device, comprising the following modules: A data conversion module, used to: convert multimodal input data into standardized text data; An information extraction module is used to: input the text data into a pre-trained large language model to obtain meeting information output by the large language model, wherein the meeting information includes meeting topics, decision content, task allocation information, and division of responsibilities of participants; A data generation module is used to: fill the meeting information into a preset meeting minutes template to obtain a meeting minutes framework; The content organization module is used to: input the meeting minutes framework into the generative language model to obtain the minutes text output by the generative language model.

[0013] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method for generating meeting minutes as described above is implemented.

[0014] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the method for generating meeting minutes as described above is implemented.

[0015] The present invention also provides a computer program product, including a computer program, wherein when the computer program is executed by a processor, the computer program implements any of the above-mentioned methods for generating meeting minutes.

[0016] The meeting minutes generation method, device, electronic device and storage medium provided by the present invention convert multimodal input data into standardized text data; input the text data into a pre-trained large language model to obtain meeting information output by the large language model, the meeting information including meeting topics, decision content, task allocation information and division of responsibilities of participants; fill the meeting information into a preset meeting minutes template to obtain a meeting minutes framework; input the meeting minutes framework into a generative language model to obtain the minutes text output by the generative language model. This solution uniformly processes input data of multiple modalities to ensure the effective integration of different input sources and accurately reflect them in the generated meeting minutes, accurately extract key information in the meeting from multimodal input, avoid errors and omissions in manual recording, significantly improve the level of automation of meeting management, reduce manual intervention, improve work efficiency, and ensure the accuracy and consistency of information processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the present invention or the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 It is one of the flow charts of the method for generating meeting minutes provided by the present invention; Figure 2This is the second flow chart of the method for generating meeting minutes provided by the present invention; Figure 3 It is a schematic diagram of the use process of the meeting minutes generation system provided by the present invention; Figure 4 It is a structural schematic diagram of the device for generating meeting minutes provided by the present invention; Figure 5 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0020] It should be noted that in the description of the embodiments of the present invention, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "include one..." do not exclude the existence of other identical elements in the process, method, article or device including the elements. The orientation or position relationship indicated by the terms "upper", "lower" and the like is based on the orientation or position relationship shown in the drawings, which is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. Unless otherwise clearly specified and limited, the terms "installed", "connected" and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected, or indirectly connected through an intermediate medium, or it can be a connection between two elements. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0021] The terms "first", "second", etc. in the present invention are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments of the present invention can be implemented in an order other than those illustrated or described herein, and the objects distinguished by "first", "second", etc. are generally of the same type, and the number of objects is not limited. For example, the first object can be one or more. In addition, "and / or" means at least one of the connected objects, and the character " / " generally indicates that the objects associated with each other are in an "or" relationship.

[0022] Combine the following Figure 1-Figure 5 The present invention describes a method, device, electronic device and storage medium for generating meeting minutes provided in an embodiment of the present invention.

[0023] Figure 1 It is one of the flow charts of the method for generating meeting minutes provided by the present invention; Figure 2 This is the second flow chart of the method for generating meeting minutes provided by the present invention; like Figure 1-Figure 2 As shown, the method includes the following: S110, converting multimodal input data into standardized text data; S120, inputting the text data into a pre-trained large language model to obtain meeting information output by the large language model, wherein the meeting information includes meeting topics, decision content, task allocation information, and division of responsibilities of participants; S130, filling the meeting information into a preset meeting minutes template to obtain a meeting minutes framework; S140, inputting the meeting minutes framework into a generative language model to obtain a minutes text output by the generative language model.

[0024] It should be noted that the execution subject of the meeting minutes generation method provided in the embodiment of the present invention may be a meeting minutes generation device, which may include but is not limited to: a server, a computer device, such as a mobile phone, a tablet computer, a laptop computer, a PDA, a wearable device, an ultra-mobile personal computer (ultra-mobile personal computer, UMPC), a netbook or a personal digital assistant (personal digital assistant, PDA), etc. The execution subject of the meeting minutes generation method may also be a meeting minutes generation system, which belongs to the meeting minutes generation device. For ease of understanding, the following embodiments take the meeting minutes generation system as the execution subject to illustrate the meeting minutes generation method provided by the present invention.

[0025] In S110, the meeting minutes generation system can receive input data in various forms, including voice recordings, text documents, and PPT annotation information. Voice data is converted into text through the voice recognition module, and PPT annotations extract text content through image processing technology. Specifically, automatic speech recognition (ASR) technology is used to convert meeting recordings or voice inputs into text, and commonly used tools include Google Speech-to-Text, Azure Speech Services, etc. For text documents, large language models are directly used for semantic analysis; PPT annotations extract annotation content through image recognition (such as OCR technology), and tools such as Tesseract OCR. After obtaining the text document obtained by converting multimodal input data, it is converted into a unified text format for subsequent processing.

[0026] Optionally, after receiving the multimodal input data, the text document is corrected and the audio data is subjected to noise reduction processing.

[0027] In S120, a pre-trained large language model (such as BERT, etc.) is used to perform deep semantic analysis on text data in a unified format to identify key meeting topics, action items, decision points, people involved, and time nodes in the text. The large language model classifies the parsed information and extracts the main discussion points, decision content, task allocation information, and division of responsibilities of the participants in the meeting. Named entity recognition (NER) technology is used to identify entity information such as people, time, and location.

[0028] In S130, the extracted key information is filled into a corresponding structured framework according to a preset meeting minutes template. The template may include meeting topics, summaries, major decisions, task assignments, time nodes, and other parts.

[0029] In S140, the large language model organizes the generated minutes content according to the logical relationship of the meeting content to ensure that the minutes are clear and focused. Natural language generation (NLG) technology is used to generate coherent minutes text.

[0030] Furthermore, the system generates adaptive content based on contextual information. For example, if the meeting involves multiple projects or topics, the system will organize the minutes content according to the relevance and priority of the projects to ensure that each topic is fully recorded.

[0031] The meeting minutes generation method provided by the embodiment of the present invention converts multimodal input data into standardized text data; inputs the text data into a pre-trained large language model to obtain meeting information output by the large language model, wherein the meeting information includes meeting topics, decision content, task allocation information, and division of responsibilities of participants; fills the meeting information into a preset meeting minutes template to obtain a meeting minutes framework; inputs the meeting minutes framework into a generative language model to obtain the minutes text output by the generative language model. This solution uniformly processes input data of multiple modalities to ensure the effective integration of different input sources and accurately reflects them in the generated meeting minutes, accurately extracts key information in the meeting from multimodal input, avoids errors and omissions in manual recording, significantly improves the automation level of meeting management and task planning, reduces manual intervention, improves work efficiency, and ensures the accuracy and consistency of information processing.

[0032] In an optional embodiment, the step of inputting the meeting minutes framework into a generative language model to obtain the minutes text output by the generative language model includes: Receiving the focus point information input by the user, wherein the focus point information is the focus point specified by the user when generating minutes; The meeting minutes framework and the focus information are input into a generative language model to obtain a minutes text output by the generative language model.

[0033] During the specific implementation process, users can specify the focus when generating minutes through the system interface (such as paying special attention to the progress of certain tasks), and the system will adjust the content generation strategy according to user needs.

[0034] The meeting minutes generation method provided in the embodiment of the present invention adjusts the meeting minutes according to the focus points input by the user, so that the generated meeting minutes conform to the actual situation and meet the actual use needs of the user.

[0035] In an optional embodiment, after inputting the meeting minutes framework into the generative language model to obtain the minutes text output by the generative language model, the method further includes: Using a preset proofreading algorithm, verifying the content of the minutes text to obtain a verified minutes text; After receiving the review feedback information input by the user, the verification minutes text is corrected based on the review feedback information to obtain a meeting minutes document.

[0036] Specifically, after the minutes are generated, the system automatically verifies the content to check for omissions or logical errors. Proofreading algorithms (such as text similarity analysis) are used to ensure the integrity of the minutes. Users can review the generated minutes and provide feedback through the system interface. The system automatically corrects the minutes content based on user feedback to form the final meeting minutes document.

[0037] The meeting minutes generation method provided in the embodiment of the present invention verifies the minutes text after generating the minutes text to ensure the accuracy and completeness of the minutes; modifies the minutes content according to the user's feedback, and finally generates a meeting minutes document that meets the user's needs.

[0038] In an optional embodiment, after the verification minutes text is modified based on the review feedback information to obtain the meeting minutes document, the method further includes: Identify the meeting minutes document and extract multiple tasks; The priority of each task is determined based on the dependency relationship of each task, the importance of each task, the urgency of each task and the resource requirement of each task.

[0039] Specifically, the system automatically identifies and extracts tasks based on the task descriptions in the meeting minutes document. It uses large language models and rule engines (such as Drools) to parse and classify tasks. Intelligent agents automatically set priorities for tasks based on their importance, urgency, and resource requirements. The priority setting algorithm takes into account task dependencies, time urgency, and resource availability to ensure that critical tasks are processed first.

[0040] The meeting minutes generation method provided in the embodiment of the present invention automatically identifies and extracts various tasks, and sets priorities for various tasks according to the dependencies, importance, urgency, and resource requirements of various tasks, so as to facilitate the arrangement and planning of subsequent tasks.

[0041] In an optional embodiment, after identifying the meeting minutes document and extracting multiple tasks, the method further includes: Acquire personnel availability information and personnel skill information, wherein the personnel availability information indicates whether the personnel is available, and the personnel skill information includes the skills possessed by the personnel; Based on the personnel availability information and the personnel skill information, an executor is matched for each of the tasks.

[0042] Specifically, the system automatically matches the right person to perform the corresponding task based on the personnel responsibilities and skill requirements mentioned in the meeting minutes. Use the human resource management module (such as the API based on the HR system) to obtain the personnel's availability and skill information.

[0043] The meeting minutes generation method provided by the embodiment of the present invention automatically matches roles for various tasks according to personnel conditions, thereby ensuring optimization of human resource allocation.

[0044] In an optional embodiment, after identifying the meeting minutes document and extracting multiple tasks, the method further includes: Based on the task requirements of each of the tasks, using a preset resource scheduling algorithm to determine resource allocation information for each of the tasks; A task schedule is generated based on the priority of each task and the resource allocation information of each task.

[0045] Specifically, according to the task requirements, the intelligent agent automatically allocates the corresponding resources (such as equipment, budget, etc.). The resource scheduling algorithm takes into account the availability of resources and the priority of tasks to ensure the optimization of resource allocation. The system automatically generates a time schedule for tasks based on the priority of tasks and resource allocation. Use Gantt chart generation tools (such as Matplotlib, Plotly, etc.) to generate task schedules.

[0046] It is understandable that, in the specific implementation process, the priority of each task can be determined first, and then the resource allocation information of each task can be determined; the resource allocation information of each task can be determined first, and then the priority of each task can be determined; the resource allocation information and priority of each task can also be determined at the same time. After determining the priority and resource allocation information of each task, the task schedule can be further generated.

[0047] Furthermore, the intelligent agent identifies the dependencies between tasks and ensures that the tasks are executed in a reasonable order. The system automatically adjusts the task schedule to avoid task conflicts and resource competition.

[0048] Optionally, the system automatically generates task flowcharts based on the task schedule and dependencies. Use flowchart drawing tools (such as Graphviz, D3.js) to generate flowcharts to show the execution path and dependencies of tasks. The system optimizes the display of flowcharts through layout algorithms to ensure that the flowchart structure is clear and easy to understand. Users can adjust the flowchart through the system interface, add comments or adjust node positions.

[0049] In some embodiments, the system dynamically updates the task schedule. When the task conditions or time nodes change, the intelligent agent automatically re-plans the task and updates the schedule and flow chart in real time.

[0050] In some embodiments, the system interface provides an interactive visual display, and the user can zoom in, zoom out, or drag the flowchart to view detailed information. Use Web front-end technology (such as React, Vue.js) to build a user-friendly interactive interface.

[0051] In some embodiments, the system can export the generated work schedule and flowchart into PDF, PNG, SVG and other formats for sharing and use on other platforms.

[0052] In some embodiments, the system supports direct sharing of work arrangements and flowcharts through a team collaboration platform to facilitate collaboration and communication among team members.

[0053] In summary, the meeting minutes generation method provided by the present invention solves the problems of cumbersome manual operations, inaccurate information extraction, lack of flexibility in task scheduling, and lack of automated generation tools in the existing meeting minutes and task planning processes. Users can use a variety of input forms such as voice, text, and PPT annotations to automatically generate meeting minutes that cover information such as meeting key content, background analysis, personnel needs, and time node arrangements, and generate work arrangements including task time scheduling, personnel allocation, work content, and priority division. Continuous optimization of task scheduling and minutes generation algorithms is achieved through user feedback and external data; task planning and allocation strategies are adjusted according to feedback results, and the optimized strategies are applied in subsequent tasks to improve the overall performance of the system and the quality of generated content.

[0054] Figure 3 This is a schematic diagram of the use of the meeting minutes generation system provided by the present invention. Figure 3 , describing in detail the complete usage process from user inputting data to generating final meeting minutes, work arrangements and flow charts.

[0055] Data input and processing: Data preparation: Users can input meeting data from multiple sources, including: Voice input: directly upload meeting recordings or real-time voice input, and the system converts voice into text through voice recognition technology; Text input: upload meeting documents (such as Word, PDF) or directly paste text content into the system; PPT annotation input: upload annotated PPT files, and the system extracts the annotation content through OCR technology; Data upload: Users upload or paste data through the system's interactive interface. The system supports batch uploading of multiple files and automatically identifies file types. After uploading data, users can select pre-processing options, such as whether to perform text correction or audio noise reduction, to improve the accuracy of subsequent analysis. Preliminary analysis: The system automatically performs preliminary analysis on the uploaded multimodal data, converts speech into text, and formats all input data uniformly; users can view a preview of the analysis results and make necessary manual adjustments (such as correcting text with recognition errors or adding missing information).

[0056] Meeting Minutes Generation: Generation parameter settings: Users can choose the template and style of meeting minutes, such as formal minutes, brief summary or project tracking report; users can also specify specific focus points, such as recording the progress of certain tasks, highlighting certain topics, etc. Minutes generation: The system uses a large language model to perform in-depth semantic analysis on input data, extract key tasks, decision points, background analysis, and other content, and automatically generate meeting minutes; users can preview the minutes content in real time and make fine adjustments through options on the interface, such as changing the order of paragraphs, adjusting wording, or adding comments; Minutes verification and confirmation: The system will automatically perform content verification to check whether the generated minutes have omissions, logical errors or format problems. Users can view the verification report through the interface; users can make final confirmation of the minutes or make corrections based on the verification suggestions. The confirmed minutes will be saved by the system and can be exported to Word, PDF or other formats.

[0057] Work schedule generation and visualization: Task analysis and planning: The system automatically identifies task items based on the generated meeting minutes, and analyzes the time nodes, responsible personnel, resource requirements and other information of each task; users can manually adjust the priority or allocation of tasks, or choose to automatically optimize task allocation and resource scheduling; Timetable generation: The system generates a work schedule based on task priorities, dependencies, and staff availability. Users can choose different display formats such as Gantt charts or timelines. Users can make fine adjustments after the schedule is generated, such as dragging to adjust task times, adding task notes, or modifying task dependencies. Flowchart generation: The system automatically generates task flowcharts based on task dependencies and time schedules to show the order and dependencies between tasks. Users can edit flowcharts through the interface, such as adjusting node positions, modifying connection line styles, or adding explanatory text. The layout of the flowchart can be automatically optimized to ensure clarity and readability of the graphics. Dynamic adjustment and real-time update: The system supports dynamic adjustment of task schedules. When users update tasks, adjust personnel, or modify time nodes during the project, the system will update the work schedule and flowchart in real time. Users can use the system's automatic update function to ensure that all task schedules and flowcharts are always up to date. Visual display and export: Users can view the final work schedule and flowchart through the system's visual interface. The system supports zooming in, zooming out, and dragging the view to view task information at different levels. Users can export work schedules and flowcharts to PDF, PNG, SVG and other formats, or directly embed them into reports, emails or collaboration platforms (such as Slack) for sharing.

[0058] Feedback and Optimization: User feedback collection: Users can provide feedback at any time through the system interface during use, such as making suggestions or pointing out problems with the generated meeting minutes or work arrangements; the system automatically collects feedback information and uses it to improve task identification, content generation, and process optimization algorithms; Automatic system optimization: The system automatically optimizes the meeting minutes generation and task planning algorithms based on the collected user feedback and historical data, gradually improving the accuracy of the generated content and user satisfaction. The optimization after each generation will be applied to the next task, ensuring that the system continues to improve its performance during continuous use; Continuous Improvement and Updates: The system development team can release updates based on collective user feedback and actual usage to enhance system functionality, fix potential issues, and introduce new optimization solutions; users can obtain the latest version and functional improvements through the system's automatic update function to ensure that they always have the best user experience.

[0059] Here is an example of a system-generated meeting minutes: Meeting Name: Project A Progress Update Meeting

[0060] Meeting time: August 15, 2024, 10:00 AM

[0061] Participants: Director Wang, Manager Li, Engineer Zhang, Assistant Chen

[0062] Meeting topic 1: Current progress report of Project A

[0063] Background: Project A has entered the third month of the development phase, and the current focus is on the functional implementation of module B and the testing and verification of module C.

[0064] Main content: Engineer Zhang reported on the development progress of module B. Currently, 70% of the functional development has been completed and the remaining development work is expected to be completed next Monday.

[0065] The preliminary test results of Module C are good. Assistant Chen is compiling the test report and plans to submit the first draft this Friday.

[0066] Decision: Director Wang requires that the development of Module B be completed and comprehensive testing begin before next Friday, and also specifies the time node for submitting the test report for Module C.

[0067] Meeting Topic 2: Resource Allocation and Personnel Arrangement

[0068] Background: Due to the tight project schedule, it is necessary to concentrate resources to complete key tasks within the next two weeks.

[0069] Main content: Manager Li proposed to allocate more resources to support the development of Module B, including adding two developers and speeding up the development progress.

[0070] Director Wang approved the resource allocation plan and asked Manager Li to be responsible for coordinating personnel allocation matters.

[0071] Decision: Two more developers will be added to Module B. Mr. Wang, who was originally responsible for Module D, will temporarily join the development team of Module B to ensure that the task is completed on time.

[0072] Meeting Topic 3: Risk Assessment and Countermeasures

[0073] Background: As projects enter a critical phase, it is critical to identify and manage potential risks.

[0074] Main content: Engineer Zhang pointed out that the integration test of module C may encounter interface compatibility issues and recommended a detailed technical review before testing.

[0075] Assistant Chen pointed out the resource conflicts that may arise during the testing process and suggested preparing a backup testing environment in advance.

[0076] Decision: A technical review meeting will be held this Thursday, with Engineer Zhang in charge of organizing it to ensure that the interface issues of module C are resolved before the integration test. At the same time, Assistant Chen is responsible for preparing a backup test environment to deal with possible resource conflicts.

[0077] Meeting Topic 4: Next Steps of Work

[0078] Background: To ensure that the project is delivered on time, the time nodes and responsible persons for each task need to be clearly defined.

[0079] Main content: Manager Li is responsible for coordinating the resource allocation of Module B and reporting on the progress before next Wednesday.

[0080] Engineer Zhang is required to complete the development of module B and start the integration test of module C by next Friday.

[0081] Assistant Chen is responsible for submitting the test report for Module C before this Friday and preparing a backup test environment.

[0082] Decision-making: Each person responsible for a task must complete the task strictly according to the time node and report progress regularly.

[0083] Meeting summary: Project A is currently progressing well, but faces certain technical and resource challenges. By properly allocating resources and conducting risk assessment and response in advance, it is expected that the development and testing work can be completed on time.

[0084] The next meeting is scheduled for next Monday at 10 am, focusing on discussing the integration test results of module C and the final development progress of module B.

[0085] To sum up, the meeting minutes generation method provided by the present invention can generate high-quality meeting minutes and detailed work arrangements in a short time by automatically processing meeting data, significantly reducing the time for manual recording and sorting. Users no longer need to manually extract key information in the meeting. The system will automatically complete the entire process from data input to minutes generation, greatly improving the efficiency of meeting management and task scheduling; it can process data from various input forms such as voice, text, and PPT annotations, and realize the unified processing and fusion of multimodal information. Whether it is a meeting recording, text record or annotation in a PPT, it can accurately parse and extract key information, making the generation of meeting minutes and work arrangements more comprehensive and accurate; the system's built-in intelligent agent can automatically identify tasks, set priorities, assign personnel and schedule resources based on the task descriptions in the meeting minutes. The intelligent agent has adaptive scheduling capabilities and can dynamically adjust the execution order and allocation plan of tasks according to real-time changing task conditions to ensure the rationality and executability of work arrangements; through the semantic parsing and context understanding capabilities of a large language model, it generates logically clear and coherent meeting minutes. The minutes cover the key topics of the meeting, background analysis, task allocation and time node arrangements to ensure that all important information is accurately recorded. Users can customize the content and style of the minutes according to actual needs to meet the application needs of different scenarios; it supports real-time updating and dynamic adjustment of task arrangements. When task conditions or priorities change, the system can automatically re-plan tasks and implement them. Timetables and flowcharts can be updated in real time to ensure that work arrangements are always in line with current actual conditions. Users can view the latest task arrangements and progress at any time through the system to facilitate timely adjustment of work plans; work schedules and task flowcharts are automatically generated, and the execution path and dependencies of tasks are intuitively displayed through visualization tools. Users can edit and adjust flowcharts through the interactive interface, and export the final work arrangements and flowcharts into multiple formats for easy sharing and collaboration. Visual display helps users better understand and manage complex task arrangements, and improves the transparency and controllability of task execution; through automated information processing and intelligent scheduling, it can reduce errors that may occur in manual recording and scheduling of tasks. The system's built-in verification and feedback mechanism ensures that the generated meeting minutes and work arrangements are accurate, reducing the risks caused by human errors; based on user feedback, the task planning and minutes generation algorithms are continuously optimized to continuously improve overall performance and effectively improve the accuracy of meeting minutes and task planning generation.

[0086] The following describes a conference minutes generating device provided by an embodiment of the present invention. The conference minutes generating device described below and the conference minutes generating method described above can refer to each other.

[0087] Figure 4 is a schematic diagram of the structure of the conference minutes generating device provided by the present invention, such as Figure 4 As shown, the meeting minutes generating device may include but is not limited to; The data conversion module 410 is used to: convert the multimodal input data into standardized text data; The information extraction module 420 is used to: input the text data into a pre-trained large language model to obtain the meeting information output by the large language model, wherein the meeting information includes the meeting agenda, decision content, task allocation information, and division of responsibilities of participants; The data generation module 430 is used to: fill the meeting information into a preset meeting minutes template to obtain a meeting minutes framework; The content organization module 440 is used to: input the meeting minutes framework into the generative language model to obtain the minutes text output by the generative language model.

[0088] It should be noted that the conference minutes generating device provided in the embodiment of the present invention can execute the conference minutes generating method described in any of the above embodiments when it is specifically run, which will not be elaborated in this embodiment.

[0089] Figure 5 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 5 As shown, the electronic device may include: a processor 510, a communications interface 520, a memory 530 and a communication bus 540, wherein the processor 510, the communications interface 520 and the memory 530 communicate with each other via the communication bus 540. The processor 510 may call the logic instructions in the memory 530 to execute the meeting minutes generation method, which includes: converting multimodal input data into standardized text data; Inputting the text data into a pre-trained large language model to obtain meeting information output by the large language model, wherein the meeting information includes meeting topics, decision content, task allocation information, and division of responsibilities of participants; Fill the meeting information into a preset meeting minutes template to obtain a meeting minutes framework; The meeting minutes framework is input into a generative language model to obtain a minutes text output by the generative language model.

[0090] In addition, the logic instructions in the above-mentioned memory 530 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0091] On the other hand, the present invention further provides a computer program product, the computer program product includes a computer program, the computer program can be stored in a non-transitory computer-readable storage medium, when the computer program is executed by a processor, the computer can execute the meeting minutes generation method provided by the above methods, the method includes: converting multimodal input data into standardized text data; Inputting the text data into a pre-trained large language model to obtain meeting information output by the large language model, wherein the meeting information includes meeting topics, decision content, task allocation information, and division of responsibilities of participants; Fill the meeting information into a preset meeting minutes template to obtain a meeting minutes framework; The meeting minutes framework is input into a generative language model to obtain a minutes text output by the generative language model.

[0092] In another aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which is implemented when the computer program is executed by a processor to execute the meeting minutes generation method provided by the above methods, the method comprising: converting multimodal input data into standardized text data; Inputting the text data into a pre-trained large language model to obtain meeting information output by the large language model, wherein the meeting information includes meeting topics, decision content, task allocation information, and division of responsibilities of participants; Fill the meeting information into a preset meeting minutes template to obtain a meeting minutes framework; The meeting minutes framework is input into a generative language model to obtain a minutes text output by the generative language model.

[0093] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0094] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0095] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for generating meeting minutes, characterized in that: include: Convert multimodal input data into standardized text data; Inputting the text data into a pre-trained large language model to obtain meeting information output by the large language model, wherein the meeting information includes meeting topics, decision content, task allocation information, and division of responsibilities of participants; Fill the meeting information into a preset meeting minutes template to obtain a meeting minutes framework; The meeting minutes framework is input into a generative language model to obtain a minutes text output by the generative language model.

2. The method for generating meeting minutes according to claim 1, characterized in that: The step of inputting the meeting minutes framework into the generative language model to obtain the minutes text output by the generative language model includes: Receiving the focus point information input by the user, wherein the focus point information is the focus point specified by the user when generating minutes; The meeting minutes framework and the focus information are input into a generative language model to obtain a minutes text output by the generative language model.

3. The method for generating meeting minutes according to claim 1, characterized in that: After inputting the meeting minutes framework into the generative language model to obtain the minutes text output by the generative language model, the method further includes: Using a preset proofreading algorithm, verifying the content of the minutes text to obtain a verified minutes text; After receiving the review feedback information input by the user, the verification minutes text is corrected based on the review feedback information to obtain a meeting minutes document.

4. The method for generating meeting minutes according to claim 3, characterized in that: After the verification minutes text is modified based on the review feedback information to obtain the meeting minutes document, the method further includes: Identify the meeting minutes document and extract multiple tasks; The priority of each task is determined based on the dependency relationship of each task, the importance of each task, the urgency of each task and the resource requirement of each task.

5. The method for generating meeting minutes according to claim 4, characterized in that: After the meeting minutes document is identified and multiple tasks are extracted, the following steps are also included: Acquire personnel availability information and personnel skill information, wherein the personnel availability information indicates whether the personnel is available, and the personnel skill information includes the skills possessed by the personnel; Based on the personnel availability information and the personnel skill information, an executor is matched for each of the tasks.

6. The method for generating meeting minutes according to claim 4, characterized in that: After the meeting minutes document is identified and multiple tasks are extracted, the following steps are also included: Based on the task requirements of each of the tasks, using a preset resource scheduling algorithm to determine resource allocation information for each of the tasks; A task schedule is generated based on the priority of each task and the resource allocation information of each task.

7. A device for generating meeting minutes, characterized in that: include: A data conversion module, used to: convert multimodal input data into standardized text data; An information extraction module is used to: input the text data into a pre-trained large language model to obtain meeting information output by the large language model, wherein the meeting information includes meeting topics, decision content, task allocation information, and division of responsibilities of participants; A data generation module is used to: fill the meeting information into a preset meeting minutes template to obtain a meeting minutes framework; The content organization module is used to: input the meeting minutes framework into the generative language model to obtain the minutes text output by the generative language model.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the method for generating meeting minutes as described in any one of claims 1 to 6 is implemented.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for generating meeting minutes as described in any one of claims 1 to 6 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method for generating meeting minutes as described in any one of claims 1 to 6 is implemented.

Images