Intelligent conference system and method based on AI

Abstract

The invention relates to the field of intelligent conferences, in particular to an intelligent conference system and method based on AI. The method comprises the following steps: setting an intelligent conference platform, obtaining identity information data and conference participation data, analyzing the identity information data and the conference participation data in combination with the intelligent conference platform, and constructing an identity matrix seat map; in combination with a conference sign-in process and sign-in completion, updating the identity matrix seating map to obtain a conference participation matrix seating map; analyzing the conference process in combination with the participation matrix seat map to obtain a conference summary path; analyzing the conference summary path in combination with the conference plan data to obtain a conference issue path; analyzing the conference issue path, and constructing a conference tracing path; and analyzing the conference tracing path in combination with the participation matrix seat map to obtain an intelligent conference summary and a seat adjustment strategy. According to the invention, the conference summary generation precision can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of intelligent meetings, in particular to an AI-based intelligent meeting system and method. BACKGROUND

[0002] With the continuous development of information technology, intelligent conference systems have gradually been applied. The goal of intelligent conference systems is to achieve automatic management of the conference process through artificial intelligence technology, thereby improving conference efficiency, reducing human errors, and improving the overall effectiveness of the conference. By using artificial intelligence technology, especially speech recognition, natural language processing, computer vision, and other technologies, intelligent conference systems can accurately capture the speech, interaction, and feedback of meeting participants, and achieve efficient management and optimization of the entire conference process.

[0003] Chinese Patent No. CN118400484A discloses an intelligent conference recording device and a conference recording modeling method, which includes a conference site module for collecting voice information and video information of the conference; a voice information collection module for detecting sound source information; a video collection module for collecting video information and analyzing facial expressions of the participants; the number and location of sound sources are determined according to the change in sound size during movement, the facial expressions of the participants at the determined sound source locations are compared to exclude errors, and the final number and location of sound sources are obtained; audio collection units are assigned one by one according to the final sound source positioning location, and audio collection of the sound sources is performed; a cloud for processing the collected voice information and video information; and a conference record generation module for generating a conference summary.

[0004] In the prior art, traditional seat allocation and identity recognition often rely on static data or manual intervention, lack the ability to dynamically adjust, and cannot be flexibly adjusted according to the actual situation of the conference. The existing system still stays on the surface in the analysis of the conference process, and cannot deeply mine the participation behavior, speech content, and task path of the meeting participants, resulting in a lack of structure and logic in the conference summary content, and a lack of effective decision support function. The existing conference summary generation method often relies on manual summarization and organization of the speech content, is easily influenced by personal subjective factors, and cannot efficiently generate accurate summaries containing key decisions and task nodes. In the conference process, the management and tracing of information such as speaking order, topic jumping, and participation degree also lack intelligent solutions, which are problems that need to be solved. SUMMARY

[0005] The present application aims to solve the problems in the background art by proposing an AI-based intelligent meeting method.

[0006] S1. Set up an intelligent meeting platform to obtain identity information data and meeting participation data. Combine the intelligent meeting platform to analyze the identity information data and meeting participation data, and construct an identity matrix seating chart.

[0007] S2. Based on the meeting check-in process and the end of check-in, update the identity matrix seating chart to obtain the participant matrix seating chart; analyze the meeting process based on the participant matrix seating chart to obtain the meeting minutes path;

[0008] S3. Analyze the meeting minutes path by combining meeting plan data to obtain the meeting agenda path; analyze the meeting agenda path to construct a meeting tracing path;

[0009] S4. Analyze the meeting tracing path by combining the meeting participant matrix and seating chart to obtain intelligent meeting minutes and seating adjustment strategies.

[0010] Preferably, the process of setting up an intelligent meeting platform, acquiring identity information data and meeting participation data, and then analyzing the identity information data and meeting participation data using the intelligent meeting platform to construct an identity matrix seating chart includes:

[0011] The intelligent meeting platform is equipped with a participant data collection terminal, a participant authentication terminal, a meeting information terminal, and a meeting recording terminal;

[0012] The participant data collection terminal is used to acquire the identity information data of pre-participants and the meeting participation data; the identity information data includes facial data, audio data, and personal information data; the meeting participation data includes meeting plan data and meeting notification documents;

[0013] The meeting information terminal is used to store all meeting distribution data, including meeting room numbers and seating charts. The meeting recording terminal is used to record the voice recordings of all participants throughout the meeting using recording equipment.

[0014] By combining AI technology with identity information data and meeting participation data obtained from the participant collection terminal, the meeting room and seats of the pre-participants are assigned, and the allocation number of the meeting room and the seating relationship of the participants are obtained; combined with the meeting distribution data and the allocation number, the meeting seat distribution map corresponding to the allocation number is obtained and recorded as the allocation seat distribution map;

[0015] By combining the personnel seating allocation relationship and the allocated seating distribution map, the identity information data of the pre-participants is mapped to the corresponding seats in the allocated seating distribution map, and the seat coordinates of the corresponding seats are obtained to construct an identity matrix seating map.

[0016] Preferably, the process of updating the identity matrix seating chart to obtain the participant seating chart, combining the meeting check-in process and the check-in completion, includes:

[0017] When the meeting is checking in, if the facial data and personal information data of the person checking in match the facial data and personal information data of the attendees at the data collection terminal, seats are assigned to the person who has passed the identity verification through the identity matrix seating chart.

[0018] If the facial data of the person checking in does not match the facial data collected by the attendee's collection terminal, and the person has a meeting notification document, the person checking in will be recorded as a temporary attendee, and a seat will be assigned to the temporary attendee through the identity matrix seating chart; if the facial data of the person checking in does not match the facial data collected by the attendee's collection terminal, and the person does not have a meeting notification document, the person checking in will not be allowed to enter the meeting.

[0019] If any assigned seats in the identity matrix seating chart become available between the end of the meeting registration period and the start of the meeting, temporary attendees will be reassigned to their corresponding seats. The changed seats will then be updated in the identity matrix seating chart to obtain the meeting attendee seating chart.

[0020] Preferably, the process of analyzing the meeting process using a participant matrix seating chart to obtain the meeting minutes path is as follows:

[0021] When the meeting begins, the speech feature extraction technology is combined with the speech recordings from the meeting recording terminal to identify the personnel related to the audio data in the seating chart of the participant matrix and record them as speakers; meeting task nodes are constructed based on the speakers in the seating chart of the participant matrix.

[0022] The system converts the continuously identified speech records of speakers into text, and uses natural language processing technology to process the converted text to obtain the speech data of the meeting participants. The speech data of the meeting participants is then stored in the meeting task node. Based on the acquired speech records, the system identifies the speaking order of the speakers and their corresponding times, and constructs the speaking time sequence link between the meeting task nodes according to the speaking order and their corresponding times.

[0023] If there is subsequent speech data of meeting participants in the already constructed meeting task node, then a meeting task sub-node is constructed within the meeting task node, and the speaking order of the participants and their corresponding times are identified. A speaking time sequence sub-link is constructed between the meeting task sub-nodes, and the corresponding speech data of the participants is stored in the meeting task sub-node. The corresponding meeting task sub-nodes and meeting task nodes are linked through the speaking time sequence sub-link and speaking time sequence link respectively to construct the meeting minutes path.

[0024] Preferably, the process of analyzing the meeting minutes path in conjunction with meeting plan data to obtain the meeting topic path includes:

[0025] By combining meeting plan data and using natural language processing technology, keywords for meeting topics are extracted from all meeting task sub-nodes and meeting participants' speech data within the meeting task nodes in the meeting minutes path. If the same keyword content exists in meeting task nodes with consecutive speaking timelines or in meeting task sub-nodes with consecutive speaking timelines, the corresponding speaking timeline and meeting task node or speaking timeline sub-node and meeting task sub-node are extracted to obtain the meeting topic path.

[0026] Preferably, the process of analyzing the meeting agenda path and constructing the meeting traceability path includes:

[0027] If the same meeting topic keyword exists within the meeting task sub-nodes or meeting task nodes across different meeting topic paths, the meeting topic path with the most identical meeting topic keywords will be designated as the main meeting topic path, and the rest will be designated as sub-meeting topic paths. The sub-meeting topic paths will be associated with the main meeting topic path based on the speaking time sequence and the time order of the speaking time sequence. The same meeting task sub-nodes or meeting task nodes associated with the main meeting topic paths will be designated as task jump nodes. The main meeting topic paths will be associated through task jump nodes to construct a meeting traceability path.

[0028] Preferably, the process of analyzing the meeting tracing path by combining the meeting participant matrix seating chart to obtain intelligent meeting minutes and seating adjustment strategies includes:

[0029] After the meeting ends, according to the speaking time sequence between each meeting task sub-node or between meeting task nodes within the meeting traceability path, the speaking data of each meeting task sub-node and between meeting task nodes are sorted out to obtain intelligent meeting minutes.

[0030] Set an active speaking frequency threshold, and count the number of meeting task sub-nodes and the total number of meeting task nodes contained within each seat coordinate in the meeting participant matrix seating chart. If the counted number is greater than or equal to the active speaking frequency threshold, organize the personal information data of the corresponding position coordinates to obtain the meeting speaking data. Use AI technology to analyze the meeting participant matrix seating chart in combination with the meeting speaking data to obtain a seating adjustment strategy.

[0031] This invention also discloses an AI-based intelligent meeting system, including a management center, which is communicatively connected to a data acquisition module, a meeting analysis module, a meeting processing module, and a meeting management module.

[0032] The data acquisition module is used to set up the intelligent meeting platform, acquire identity information data and meeting participation data, and analyze the identity information data and meeting participation data in conjunction with the intelligent meeting platform to construct an identity matrix seating chart.

[0033] The meeting analysis module is used to update the identity matrix seating chart by combining the meeting check-in process and the end of check-in, and obtain the participant matrix seating chart; it also analyzes the meeting process by combining the participation matrix seating chart to obtain the meeting minutes path.

[0034] The meeting processing module is used to analyze the meeting minutes path by combining meeting plan data to obtain the meeting agenda path; and to analyze the meeting agenda path to construct the meeting traceability path.

[0035] The meeting management module is used to analyze the meeting tracing path by combining the meeting participant matrix and seating chart, and to obtain intelligent meeting minutes and seating adjustment strategies.

[0036] Compared with the prior art, the above-mentioned technical solution of the present invention has the following beneficial technical effects: By combining the identity matrix seating chart and meeting topic path analysis, the speaking process and task association path in the meeting can be automatically identified, improving the accuracy and real-time performance of meeting minutes generation, reducing the workload of manually summarizing meeting content, and enhancing the structure and logic of meeting content, ensuring the effective traceability of task links and speaking records; By combining the dynamic adjustment mechanism of seat allocation and meeting topic path analysis, the seating arrangement of the next meeting is optimized, improving meeting participation, ensuring the spatial and temporal association between speaking and topic, and helping to improve the overall interactive effect and efficiency of the meeting. Attached Figure Description

[0037] Figure 1 This is a schematic diagram of one embodiment of the present invention. Detailed Implementation

[0038] Example 1, as Figure 1 As shown, the AI-based intelligent meeting method proposed in this invention includes the following steps:

[0039] S1. Set up an intelligent meeting platform to obtain identity information data and meeting participation data. Combine the intelligent meeting platform to analyze the identity information data and meeting participation data, and construct an identity matrix seating chart.

[0040] S2. Based on the meeting check-in process and the end of check-in, update the identity matrix seating chart to obtain the participant matrix seating chart; analyze the meeting process based on the participant matrix seating chart to obtain the meeting minutes path;

[0041] S3. Analyze the meeting minutes path by combining meeting plan data to obtain the meeting agenda path; analyze the meeting agenda path to construct a meeting tracing path;

[0042] S4. Analyze the meeting tracing path by combining the meeting participant matrix and seating chart to obtain intelligent meeting minutes and seating adjustment strategies.

[0043] It should be further explained that, in the specific implementation process, the process of setting up an intelligent meeting platform, acquiring identity information data and meeting participation data, and combining the intelligent meeting platform to analyze the identity information data and meeting participation data to construct an identity matrix seating chart includes:

[0044] The intelligent meeting platform is equipped with a participant collection terminal, a participant authentication terminal, a meeting information terminal, and a meeting recording terminal;

[0045] The participant data acquisition terminal is used to obtain the identity information data of the pre-participants and the meeting participation data; the identity information data includes facial data, audio data, and personal information data; the meeting participation data includes meeting plan data and meeting notification documents;

[0046] Specifically, the personal information data includes, but is not limited to, name, phone number, and gender; the meeting plan data refers to the information data that needs to be prepared and arranged for the planned meeting, including but not limited to the meeting date and meeting agenda; the meeting notification document refers to the meeting notification document that informs the participants of the meeting, including but not limited to the meeting check-in time, check-in end time, and meeting start time.

[0047] The participant authentication terminal is used to authenticate the identities of participants in the meeting and to track their participation.

[0048] The conference information terminal is used to store all conference distribution data; the conference distribution data includes conference room numbers and conference seating charts.

[0049] The meeting recording terminal is used to record the voice speech of all participants throughout the meeting using a recording device installed in the microphone of the meeting room.

[0050] By combining AI technology with identity information data and meeting participation data obtained from the participant collection terminal, the meeting room and seats of the pre-participants are assigned, and the allocation number of the meeting room and the seating relationship of the participants are obtained; combined with the meeting distribution data and the allocation number, the meeting seat distribution map corresponding to the allocation number is obtained and recorded as the allocation seat distribution map;

[0051] By combining the personnel seating allocation relationship and the allocated seating distribution map, the identity information data of the pre-participants is mapped to the corresponding seats in the allocated seating distribution map, and the seat coordinates of the corresponding seats are obtained to construct an identity matrix seating map.

[0052] It should be further explained that, in the specific implementation process, the identity matrix seating chart is updated based on the meeting check-in process and its completion to obtain the participant matrix seating chart; the process of analyzing the meeting process based on the participant matrix seating chart to obtain the meeting minutes path includes:

[0053] When the meeting is checking in, the identity of the person checking in is verified through the attendee authentication terminal. If the facial data and personal information data of the person checking in match the facial data and personal information data of the attendee collection terminal, then the identity verification is passed. The seat is then assigned to the person who has passed the identity verification through the identity matrix seating chart, and the seat assigned to the person in the identity matrix seating chart is in the "attendance status".

[0054] If the facial data of the person who signs in does not match the facial data of the attendees collected from the collection terminal, and the person has a meeting notification document, the person who signs in will be recorded as a temporary attendee. The facial data and personal information of the temporary attendee will be recorded, and a seat will be assigned to the temporary attendee through an identity matrix seating chart.

[0055] If the facial data of the person who signs in does not match the facial data of the attendees collected from the device, and the person does not have a meeting notification document, then the person who signs in will not be allowed to enter the meeting.

[0056] If any assigned seats in the identity matrix seating chart become available between the end of the meeting registration period and the start of the meeting, temporary attendees will be reassigned to their corresponding seats, and the corresponding pre-participants will be marked as "absentees". The changed seats will then be updated in the identity matrix seating chart to obtain the meeting attendance matrix seating chart.

[0057] When the meeting begins, the speech feature extraction technology is combined with the speech recordings from the meeting recording terminal to identify the personnel related to the audio data in the seating chart of the participant matrix and record them as speakers; meeting task nodes are constructed based on the speakers in the seating chart of the participant matrix.

[0058] The system converts the continuously identified speech records of speakers into text, and uses natural language processing technology to process the converted text to obtain the speech data of the meeting participants. The speech data of the meeting participants is then stored in the meeting task node. Based on the acquired speech records, the system identifies the speaking order of the speakers and their corresponding times, and constructs the speaking time sequence link between the meeting task nodes according to the speaking order and their corresponding times.

[0059] If there is subsequent speech data of meeting participants in the already constructed meeting task node, then a meeting task sub-node is constructed within the meeting task node, and the speaking order of the participants and their corresponding times are identified. A speaking time sequence sub-link is constructed between the meeting task sub-nodes, and the corresponding speech data of the participants is stored in the meeting task sub-node. The corresponding meeting task sub-nodes and meeting task nodes are linked through the speaking time sequence sub-link and speaking time sequence link respectively to construct the meeting minutes path.

[0060] It should be further explained that, in the specific implementation process, the meeting minutes path is analyzed in conjunction with meeting plan data to obtain the meeting agenda path; the process of analyzing the meeting agenda path and constructing the meeting traceability path includes:

[0061] By combining meeting plan data and using natural language processing technology, keywords for meeting topics are extracted from all meeting task sub-nodes and meeting participants' speech data within the meeting task nodes in the meeting minutes path. If the same keyword content exists in meeting task nodes with consecutive speech timelines or in meeting task sub-nodes with consecutive speech timelines, the corresponding speech timeline and meeting task node or speech timeline sub-node and meeting task sub-node are extracted to obtain the meeting topic path.

[0062] If the same meeting topic keyword exists within the meeting task sub-nodes or meeting task nodes across different meeting topic paths, the meeting topic path with the most identical meeting topic keywords will be designated as the main meeting topic path, and the rest will be designated as sub-meeting topic paths. The sub-meeting topic paths will be associated with the main meeting topic path based on the speaking time sequence and the time order of the speaking time sequence. The same meeting task sub-nodes or meeting task nodes associated with the main meeting topic paths will be designated as task jump nodes. The main meeting topic paths will be associated through task jump nodes to construct a meeting traceability path.

[0063] It should be further explained that, in the specific implementation process, the process of analyzing the meeting tracing path in conjunction with the meeting participant matrix seating chart to obtain intelligent meeting minutes and seating adjustment strategies includes:

[0064] After the meeting ends, the meeting content is presented through the meeting traceability path. According to the speaking time sequence between each meeting task sub-node or between meeting task nodes within the meeting traceability path, the speaking data of each meeting task sub-node and between meeting task nodes are sorted out to obtain intelligent meeting minutes.

[0065] A threshold for active speaking frequency is set. The number of meeting task sub-nodes and the total number of meeting task nodes contained within each seat coordinate in the meeting participant seating chart are counted. If the count is less than the active speaking frequency threshold, the participant at that position has a low participation rate in speaking. If the count is greater than or equal to the active speaking frequency threshold, the personal information data of the corresponding position coordinate is compiled to obtain the speaking data. Using AI technology, the meeting seating chart is analyzed in conjunction with the speaking data to provide a strategy for adjusting seats in the next meeting, thus obtaining a seating adjustment strategy.

[0066] Example 2: The AI-based intelligent meeting system proposed in this invention is applied to the AI-based intelligent meeting method described in Example 1. Specifically, it includes a management center, which is communicatively connected to a data acquisition module, a meeting analysis module, a meeting processing module, and a meeting management module.

[0067] The data acquisition module is used to set up the intelligent meeting platform, acquire identity information data and meeting participation data, and analyze the identity information data and meeting participation data in conjunction with the intelligent meeting platform to construct an identity matrix seating chart.

[0068] The meeting analysis module is used to update the identity matrix seating chart by combining the meeting check-in process and the end of check-in, and obtain the participant matrix seating chart; it also analyzes the meeting process by combining the participation matrix seating chart to obtain the meeting minutes path.

[0069] The meeting processing module is used to analyze the meeting minutes path by combining meeting plan data to obtain the meeting agenda path; and to analyze the meeting agenda path to construct the meeting traceability path.

[0070] The meeting management module is used to analyze the meeting tracing path by combining the meeting participant matrix and seating chart, and to obtain intelligent meeting minutes and seating adjustment strategies.

[0071] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited thereto. Various changes can be made within the scope of knowledge possessed by those skilled in the art without departing from the spirit of the present invention.

Claims

1. An AI-based intelligent meeting method, characterized in that, Includes the following steps: S1. Set up an intelligent meeting platform to obtain identity information data and meeting participation data. Combine the intelligent meeting platform to analyze the identity information data and meeting participation data, and construct an identity matrix seating chart. S2. Based on the meeting check-in process and the end of check-in, update the identity matrix seating chart to obtain the participant matrix seating chart; By combining the seating chart with the participation matrix, the meeting process can be analyzed to obtain the meeting minutes path; S3. Analyze the meeting minutes path by combining meeting plan data to obtain the meeting agenda path; analyze the meeting agenda path to construct a meeting tracing path; S4. Analyze the meeting tracing path by combining the meeting participant matrix and seating chart to obtain intelligent meeting minutes and seating adjustment strategies.

2. The AI-based intelligent meeting method according to claim 1, characterized in that, The process of setting up an intelligent meeting platform, acquiring identity information data and meeting participation data, and then analyzing the identity information data and meeting participation data using the intelligent meeting platform to construct an identity matrix seating chart includes: The intelligent meeting platform is equipped with a participant data collection terminal, a participant authentication terminal, a meeting information terminal, and a meeting recording terminal; The participant data collection terminal is used to acquire the identity information data of pre-participants and the meeting participation data; the identity information data includes facial data, audio data, and personal information data; the meeting participation data includes meeting plan data and meeting notification documents; The meeting information terminal is used to store all meeting distribution data, including meeting room numbers and seating charts. The meeting recording terminal is used to record the voice recordings of all participants throughout the meeting using recording equipment. By combining AI technology with identity information data and meeting participation data obtained from the participant collection terminal, the meeting room and seats of the pre-participants are assigned, and the allocation number of the meeting room and the seating relationship of the participants are obtained; combined with the meeting distribution data and the allocation number, the meeting seat distribution map corresponding to the allocation number is obtained and recorded as the allocation seat distribution map; By combining the personnel seating allocation relationship and the allocated seating distribution map, the identity information data of the pre-participants is mapped to the corresponding seats in the allocated seating distribution map, and the seat coordinates of the corresponding seats are obtained to construct an identity matrix seating map.

3. The AI-based intelligent conferencing method according to claim 2, characterized in that, The process of updating the identity matrix seating chart and obtaining the participant seating chart by combining the meeting check-in process and the check-in completion includes: When the meeting is checking in, if the facial data and personal information data of the person checking in match the facial data and personal information data of the attendees at the data collection terminal, seats are assigned to the person who has passed the identity verification through the identity matrix seating chart. If the facial data of the person checking in does not match the facial data collected by the attendee's collection terminal, and the person has a meeting notification document, the person checking in will be recorded as a temporary attendee, and a seat will be assigned to the temporary attendee through the identity matrix seating chart; if the facial data of the person checking in does not match the facial data collected by the attendee's collection terminal, and the person does not have a meeting notification document, the person checking in will not be allowed to enter the meeting. If any assigned seats in the identity matrix seating chart become available between the end of the meeting registration period and the start of the meeting, temporary attendees will be reassigned to their corresponding seats. The changed seats will then be updated in the identity matrix seating chart to obtain the meeting attendee seating chart.

4. The AI-based intelligent conferencing method according to claim 3, characterized in that, The process of analyzing the meeting by combining the participation matrix seating chart and obtaining the meeting minutes path is as follows: When the meeting begins, the speech feature extraction technology is combined with the speech recordings from the meeting recording terminal to identify the personnel related to the audio data in the seating chart of the participant matrix and record them as speakers; meeting task nodes are constructed based on the speakers in the seating chart of the participant matrix. The system converts the continuously identified speech records of speakers into text, and uses natural language processing technology to process the converted text to obtain the speech data of the meeting participants. The speech data of the meeting participants is then stored in the meeting task node. Based on the acquired speech records, the system identifies the speaking order of the speakers and their corresponding times, and constructs the speaking time sequence link between the meeting task nodes according to the speaking order and their corresponding times. If there is subsequent speech data of meeting participants in the already constructed meeting task node, then a meeting task sub-node is constructed within the meeting task node, and the speaking order of the participants and their corresponding times are identified. A speaking time sequence sub-link is constructed between the meeting task sub-nodes, and the corresponding speech data of the participants is stored in the meeting task sub-node. The corresponding meeting task sub-nodes and meeting task nodes are linked through the speaking time sequence sub-link and speaking time sequence link respectively to construct the meeting minutes path.

5. The AI-based intelligent meeting method according to claim 4, characterized in that, The process of analyzing meeting minutes paths by combining meeting plan data to obtain meeting agenda paths includes: By combining meeting plan data and using natural language processing technology, keywords for meeting topics are extracted from all meeting task sub-nodes and meeting participants' speech data within the meeting task nodes in the meeting minutes path. If the same keyword content exists in meeting task nodes with consecutive speaking timelines or in meeting task sub-nodes with consecutive speaking timelines, the corresponding speaking timeline and meeting task node or speaking timeline sub-node and meeting task sub-node are extracted to obtain the meeting topic path.

6. The AI-based intelligent meeting method according to claim 5, characterized in that, The process of analyzing the meeting agenda path and constructing a meeting tracing path includes: If the same meeting topic keyword exists within the meeting task sub-nodes or meeting task nodes across different meeting topic paths, the meeting topic path with the most identical meeting topic keywords will be designated as the main meeting topic path, and the rest will be designated as sub-meeting topic paths. The sub-meeting topic paths will be associated with the main meeting topic path based on the speaking time sequence and the time order of the speaking time sequence. The same meeting task sub-nodes or meeting task nodes associated with the main meeting topic paths will be designated as task jump nodes. The main meeting topic paths will be associated through task jump nodes to construct a meeting traceability path.

7. The AI-based intelligent conferencing method according to claim 6, characterized in that, The process of analyzing meeting tracing paths by combining the meeting participant matrix and seating chart to obtain intelligent meeting minutes and seating adjustment strategies includes: After the meeting ends, according to the speaking time sequence between each meeting task sub-node or between meeting task nodes within the meeting traceability path, the speaking data of each meeting task sub-node and between meeting task nodes are sorted out to obtain intelligent meeting minutes. Set an active speaking frequency threshold, and count the number of meeting task sub-nodes and the total number of meeting task nodes contained within each seat coordinate in the meeting participant matrix seating chart. If the counted number is greater than or equal to the active speaking frequency threshold, organize the personal information data of the corresponding position coordinates to obtain the meeting speaking data. Use AI technology to analyze the meeting participant matrix seating chart in combination with the meeting speaking data to obtain a seating adjustment strategy.

8. An AI-based intelligent meeting system, specifically applied to the AI-based intelligent meeting method according to any one of claims 1 to 7, comprising a management center, characterized in that, The management center communication connection includes a data acquisition module, a meeting analysis module, a meeting processing module, and a meeting management module. The data acquisition module is used to set up the intelligent meeting platform, acquire identity information data and meeting participation data, and analyze the identity information data and meeting participation data in conjunction with the intelligent meeting platform to construct an identity matrix seating chart. The meeting analysis module is used to update the identity matrix seating chart by combining the meeting check-in process and the end of check-in, and obtain the meeting participant matrix seating chart. By combining the seating chart with the participation matrix, the meeting process can be analyzed to obtain the meeting minutes path; The meeting processing module is used to analyze the meeting minutes path by combining meeting plan data to obtain the meeting agenda path; and to analyze the meeting agenda path to construct the meeting traceability path. The meeting management module is used to analyze the meeting tracing path by combining the meeting participant matrix and seating chart, and to obtain intelligent meeting minutes and seating adjustment strategies.

Images