Virtual reality conference information processing method and system

Abstract

The invention discloses a virtual reality conference information processing method and system, and belongs to the field of information processing, and the virtual reality conference information processing method comprises the steps: if a sudden notification is detected, summarizing conference information before the sudden notification, generating a conference text, and providing the conference text to a user for real-time reference; detecting whether the user marks the text in the conference text or not, and if yes, expanding the marked text based on the conference information; compared with the prior art, the method has the beneficial effects that even if the user is distracted in the virtual reality conference process and does not know the conference content, when the sudden notification occurs, conference information before the sudden notification is summarized, the conference text is generated for reference of the user, and when the user marks the text, extension is carried out based on the marked text, so that the user experience is improved. And the user is assisted to quickly know and process the sudden notification.

Description

Technical Field

[0001] The present invention belongs to the field of information processing, and in particular relates to a virtual reality conference information processing method and system. Background Art

[0002] Virtual reality conferences refer to online conferences held based on virtual reality (VR) technology. As an important tool for digital transformation, virtual reality conferences enhance interactivity and security, support multiple forms of interaction such as voice, body movements, and virtual whiteboards, and increase communication efficiency by 40% compared to traditional video conferences. They also automatically record the entire meeting process for easy subsequent review and information tracing.

[0003] During virtual reality meetings, some users may be distracted by viewing other applications (such as messages and emails) in the VR environment, resulting in missing meeting information. At this time, if there are sudden notifications (such as voting reminders), users will not know how to handle them, which needs improvement. Summary of the Invention

[0004] Based on this, it is necessary to provide a virtual reality conference information processing method and system to address the above problems.

[0005] The embodiment of the present invention is implemented as follows: a method for processing virtual reality conference information, comprising:

[0006] If an unexpected notification is detected, the meeting information before the unexpected notification is summarized (the virtual reality meeting will automatically record the entire meeting process to facilitate subsequent review and information tracing), and a meeting text is generated and provided to users for real-time reference;

[0007] Detect whether the user has marked text in the conference text. If there is marked text, expand the marked text based on the conference information.

[0008] In one embodiment, the present invention provides a method for processing virtual reality conference information. The method includes the following steps: if a sudden notification is detected, summarizing the conference information before the sudden notification, generating a conference text, and providing it to the user for real-time reference.

[0009] The meeting recording function (voice-to-text + motion capture) stores structured data (speech, voting, file sharing, etc.) by timeline. When an emergency notification is triggered, the system automatically captures the text content and associated operation logs of the previous N minutes (e.g., 2 minutes).

[0010] Use lightweight models (such as lightweight BERT) to extract key information from captured content, identifying core issues, resolution trends, and pending issues. Combined with voice sentiment analysis (such as tone and urgency) and shared document keywords, weighted summary text is generated with priority ranking.

[0011] The generated text is compressed into M short sentences (such as 3, 5, etc.) as the meeting text, and pushed to the user through a floating panel (upper right, the panel can be semi-transparent), while retaining the original data link for tracing back.

[0012] In one embodiment, the present invention provides a method for processing virtual reality conference information, wherein the step of detecting whether a user has marked text in a conference text and, if marked text exists, expanding the marked text based on the conference information specifically includes:

[0013] Capture the marked text selected by the user, record the timestamp of the marked text and its position in the meeting text; combine it with meeting records (voice transcription, shared documents, voting records) to extract the context of the marked content (such as speaker identity, discussion progress and related documents);

[0014] Provides the ability to select marking purposes, including questioning, supplementing, and asking questions. Matches relevant content in the meeting based on the marking purpose (e.g., after marking "budget risk" and selecting the marking purpose as supplement, pushes the preliminary cost analysis table and the corresponding discussion segment link).

[0015] An expansion box (which can be semi-transparent) is generated next to the marked position to display charts and summaries. The charts include file thumbnails and 3D data visualization charts.

[0016] In one embodiment, the present invention provides a method for processing virtual reality conference information. Before the step of detecting a sudden notification, summarizing the conference information before the sudden notification, generating a conference text, and providing it to a user for real-time reference, the method further includes:

[0017] In the UI (user interface) system, the rendering priority of notification pop-ups is set to the highest level, forcing them to appear above virtual scene elements (such as 3D models and PowerPoint presentations). The notification bar is fixed to the upper right corner of the user's field of view through an anchor system, dynamically adapting to the headset's viewing angle offset.

[0018] Add CSS animations (such as border pulse mutations and transparency gradients) to emergency notifications, automatically placing them on top and covering other non-critical operation interfaces when triggered; integrate directional spatial audio technology to make the prompt sound play in a direction corresponding to the position of the notification pop-up window (upper right), enhancing directional perception;

[0019] Detect user operation status (such as raising a hand or writing) in real time. If the hand model or tool interface overlaps with the notification area, adjust the notification area to ensure information visibility. Methods for adjusting the notification area include automatically reducing the notification size and temporarily making it translucent.

[0020] In one embodiment, the present invention provides a method for processing virtual reality conference information. Before the step of detecting a sudden notification, summarizing the conference information before the sudden notification, generating a conference text, and providing it to a user for real-time reference, the method further includes:

[0021] Crawl internal company documents (product manuals, compliance documents), historical meeting minutes, and industry standard libraries (such as the ISO official website) to extract terms (such as "ISO 9001"), clauses, and related cases. Use NLP tools (such as spaCy) to identify entities and relationships, store them in a graph database (Neo4j), and form a knowledge graph of terms, clauses, and cases.

[0022] Transcribe real-time speech in meetings into text and detect terminology using pre-trained entity recognition models (such as BERT-NER). Query the knowledge graph based on the terminology, return relevant clause summaries and case links, and weightedly display high-frequency citations.

[0023] A (semi-transparent) floating panel is generated on the side of the VR interface, displaying core terms (foldable), case thumbnails and one-click jump entrances in layers.

[0024] In one embodiment, the present invention provides a virtual reality conference information processing system, comprising:

[0025] The meeting text generation module is used to summarize the meeting information before the sudden notification if it is detected (the virtual reality meeting will automatically record the entire meeting process to facilitate subsequent review and information tracing), generate the meeting text, and provide it to users for real-time reference;

[0026] The markup extension module is used to detect whether the user has marked text in the conference text. If there is marked text, the marked text is expanded based on the conference information.

[0027] In one embodiment, the present invention provides a virtual reality conference information processing system, wherein the conference text generation module includes:

[0028] The content capture and association unit is used to store structured data (speech, voting, file sharing, etc.) by timeline through the meeting recording function (voice-to-text + motion capture). When an emergency notification is triggered, the system automatically captures the text content of the previous N minutes (for example, 2 minutes) and the associated operation logs;

[0029] The key information extraction unit uses a lightweight model (such as lightweight BERT) to extract key information from the intercepted content, identifying core issues, resolution trends, and pending issues. It also combines voice sentiment analysis (such as tone and urgency) with shared document keywords to generate weighted, prioritized summary text.

[0030] The text compression unit is used to compress the generated text into M short sentences (such as 3, 5, etc.) as the conference text, and push it to the user through the floating panel (upper right, the panel can be semi-transparent), while retaining the original data link for tracing back.

[0031] In one embodiment, the present invention provides a virtual reality conference information processing system, wherein the tag extension module includes:

[0032] The tag capture unit is used to capture the marked text selected by the user, record the timestamp of the marked text and its position in the meeting text; and extract the context of the marked content (such as speaker identity, discussion progress and related documents) by combining it with the meeting records (voice transcription, shared documents, voting records);

[0033] The related content matching unit is used to provide a selection of marking purposes, including questioning, supplementing, and asking questions. It also matches related content in the meeting based on the marking purpose (for example, after marking "budget risk" and selecting the marking purpose as supplement, the previous cost analysis table and the corresponding discussion segment link will be pushed).

[0034] The associated content display unit is used to generate an expansion box (the expansion box can be semi-transparent) next to the marked position to display charts and summaries. The charts include file thumbnails and 3D data visualization charts.

[0035] In one embodiment, the present invention provides a virtual reality conference information processing system, the virtual reality conference information processing system further comprising:

[0036] A priority setting module is used to set the rendering priority of notification pop-ups to the highest level in the UI (user interface) system, forcing notification pop-ups to appear above virtual scene elements (such as 3D models and PowerPoint presentations). The notification bar is fixed to the upper right corner of the user's field of view through an anchor system, and dynamically adapts to the headset's viewing angle offset.

[0037] Animation and audio prompt module, used to add CSS animations (such as sudden border pulses and transparency gradients) to sudden notifications. When triggered, it automatically stays on top and covers other non-critical operation interfaces. It also integrates directional spatial audio technology to make the prompt sound play in a direction that matches the position of the notification pop-up window (upper right), enhancing the sense of direction.

[0038] The operation status detection module is used to detect the user's operation status (such as raising a hand or writing) in real time. If the hand model or tool interface overlaps with the notification area, the notification area is adjusted to ensure information visibility. Methods for adjusting the notification area include automatically reducing the notification size and temporarily making it translucent.

[0039] In one embodiment, the present invention provides a virtual reality conference information processing system, the virtual reality conference information processing system further comprising:

[0040] A knowledge graph construction module is used to crawl internal enterprise documents (product manuals, compliance documents), historical meeting minutes, and industry standard libraries (such as the ISO official website) to extract terms (such as "ISO 9001"), clauses, and related cases. It uses NLP tools (such as spaCy) to identify entities and relationships, and stores them in a graph database (Neo4j) to form a knowledge graph of terms, clauses, and cases.

[0041] The knowledge graph matching module is used to transcribe real-time speech in meetings into text and detect terms using pre-trained entity recognition models (such as BERT-NER). It queries the knowledge graph based on terms, returns relevant term summaries and case links, and displays high-frequency citations in a weighted manner.

[0042] The knowledge graph display module is used to generate a (semi-transparent) floating panel on the side of the VR interface, displaying core terms (foldable) in layers, case thumbnails, and one-click jump entrances.

[0043] Compared with the prior art, the beneficial effect of the present invention is that even if the user is distracted during a virtual reality meeting and is unclear about the meeting content, when a sudden notification occurs, the present invention will summarize the meeting information before the sudden notification and generate a meeting text for the user's reference. When the user marks the text, it will be expanded based on the marked text to help the user quickly understand and handle the sudden notification. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 This is a flowchart of the first part of a method for processing virtual reality conference information provided by an embodiment of the present invention.

[0045] Figure 2 A schematic diagram of the process of generating conference text provided by an embodiment of the present invention.

[0046] Figure 3 A schematic diagram of the process of expanding marked text provided by an embodiment of the present invention.

[0047] Figure 4 This is a flowchart of the second part of a virtual reality conference information processing method provided by an embodiment of the present invention.

[0048] Figure 5 This is a flowchart of the third part of a virtual reality conference information processing method provided by an embodiment of the present invention.

[0049] Figure 6 This is a schematic diagram of the first part of a virtual reality conference information processing system provided by an embodiment of the present invention.

[0050] Figure 7 A schematic diagram of a conference text generation module provided in an embodiment of the present invention.

[0051] Figure 8 A schematic diagram of a tag extension module provided in an embodiment of the present invention.

[0052] Figure 9 This is a schematic diagram of the second part of a virtual reality conference information processing system provided by an embodiment of the present invention.

[0053] Figure 10 This is a schematic diagram of the third part of a virtual reality conference information processing system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0054] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0055] In one embodiment, Figure 1 As shown, a virtual reality conference information processing method includes:

[0056] Step S7: If an emergency notification is detected, summarize the meeting information before the emergency notification (the virtual reality meeting will automatically record the entire meeting process to facilitate subsequent review and information tracing), generate a meeting text, and provide it to the user for real-time reference;

[0057] Step S8: Detect whether the user has marked text in the conference text. If there is marked text, expand the marked text based on the conference information.

[0058] The core of the design of steps S7 and S8 is to solve two core problems derived from the immersive environment characteristics in virtual reality (VR) meetings: information gaps in sudden scenarios and insufficient knowledge depth in user active interaction.

[0059] In a single-task system architecture, some VR devices use an exclusive operation mechanism, forcing other applications into the background or suspended state. At this point, the conference application immediately loses control of the foreground. In a multi-task system, VR platforms that support multiple windows (such as Meta Quest Pro) allow split-screen operation. Therefore, in a multi-task system, while the user is busy with other tasks or distracted, this application can support this by maintaining the foreground work state.

[0060] Step S7: Real-time decision support for information gaps

[0061] The high immersion of VR meetings can easily cause users to lose focus due to unexpected notifications (such as voting or urgent topics). Therefore, step S7 is used to ensure decision consistency:

[0062] Instant context restoration: By automatically capturing and summarizing key discussions (such as controversial points and biased conclusions) before the notification is triggered, users' attention is quickly drawn back to the core of the topic, avoiding repeated communication or misjudgment caused by distraction.

[0063] Dynamic Priority Extraction: This system combines voice sentiment analysis (such as speaker urgency) with document keyword weights to generate a summary with implicit bias (such as "Most support Option B, but the cost is questionable"), helping users quickly grasp the focus of the conflict rather than mechanically retelling the text.

[0064] Lightweight information push: Through the translucent design of the floating panel in the upper right corner, key conclusions are delivered in a non-invasive manner while maintaining VR immersion, reducing cognitive switching costs.

[0065] Step S8: Strengthen the knowledge loop of active interaction

[0066] The information density in VR environments is high, but the passive reception mode can easily lead to "information overload and inefficient knowledge conversion." Therefore, step S8 is used to clarify the user's subjective information needs:

[0067] Externalizing and Structuring Demand: User tagging is essentially a proactive expression of ambiguous information needs. By categorizing and tagging purposes (questioning / supplementing / asking questions), unstructured needs are transformed into computable query logic, enabling precise content matching.

[0068] Cross-modal knowledge association: Dynamically bind text markers with non-text resources such as 3D models and shared documents in the meeting (for example, displaying a 3D data chart after clicking a marker), breaking through the information limitations of linear text and enhancing multi-dimensional cognition.

[0069] Personal knowledge anchor point construction: Marked content and extended information are automatically deposited into personal traceable notes, forming a complete knowledge chain from "meeting interaction" to "post-meeting execution", avoiding the loss of key information.

[0070] In one embodiment, Figure 2 As shown, a method for processing virtual reality conference information, wherein step S7, if a sudden notification is detected, summarizes the conference information before the sudden notification, generates a conference text, and provides it to the user for real-time reference, specifically includes:

[0071] Step S71: Use the meeting recording function (voice-to-text + motion capture) to store structured data (speech, voting, file sharing, etc.) according to the timeline. When an emergency notification is triggered, the system automatically captures the text content and associated operation logs of the previous N minutes (e.g., 2 minutes);

[0072] Step S72: Use a lightweight model (e.g., lightweight BERT) to extract key information from the intercepted content, identifying core issues, resolution trends, and pending issues. Combined with voice sentiment analysis (e.g., tone of voice urgency) and shared document keywords, a weighted summary text with priority ranking is generated.

[0073] Step S73, compress the generated text into M short sentences (such as 3, 5, etc.), as the conference text, and push it to the user through the floating panel (upper right, the panel can be semi-transparent), while retaining the original data link for backtracking.

[0074] For example, during a VR product review meeting, the team was discussing the cost of a new headset design. The moderator suddenly called a vote: "Should we adopt Plan B? Please vote within 5 minutes."

[0075] Step S71: Data capture: The system automatically captures the discussion content 2 minutes before the voting is triggered, including:

[0076] Speech transcription: "Option A has a 20% material cost overrun, but the sound quality is rated 9.1; Option B has a manageable cost and a score of 8.3."

[0077] Shared document: Cost comparison table (Plan A 1,500 yuan vs Plan B 1,200 yuan).

[0078] Action record: The project manager clicked on the 3D model of Plan B.

[0079] S72 Summary Generation: The NLP model extracts keywords: "cost overrun" and "Option B score 8.3." Combined with voice sentiment analysis, the project manager's tone of voice was more urgent (higher urgency) when mentioning Option B. After weighting, Option B was determined to be a potential preference. The output summary reads: "Focus of the last 2 minutes: Option A has a 20% cost overrun but excellent sound quality; Option B has manageable costs but a slightly lower score."

[0080] Most discussions leaned towards Option B.”

[0081] S73 push display: A translucent panel pops up in the upper right corner of the user, showing 3 summaries and a "Click to view the complete cost table" link at the bottom.

[0082] In one embodiment, Figure 3 As shown, a method for processing virtual reality conference information, wherein step S8 is detecting whether the user has marked text in the conference text, and if there is marked text, the step of expanding the marked text based on the conference information specifically includes:

[0083] Step S81: Capture the marked text selected by the user, record the timestamp of the marked text and its location in the meeting text; extract the context of the marked content (such as speaker identity, discussion progress, and related documents) in combination with the meeting records (voice transcription, shared documents, voting records);

[0084] Step S82 provides a selection of marking purposes, including questioning, supplementing, and asking questions; matching related content in the meeting based on the marking purpose (for example, after marking "budget risk" and selecting the marking purpose as supplement, the previous cost analysis table and the corresponding discussion segment link are pushed);

[0085] Step S83: Generate an expansion box (the expansion box may be semi-transparent) next to the marked position to display charts and summaries. The charts include file thumbnails and 3D data visualization charts.

[0086] The following description will continue with examples based on the example scenarios of steps S71 to S73.

[0087] S81 Mark Capture: The user uses the handle to select the text "Cost overrun 20%" in the summary, and the system records the timestamp of the mark (35 minutes into the meeting) and the associated document (cost table for Plan A).

[0088] S82 Purpose Match: The user selects the mark purpose as "Supplement." The system searches the knowledge base for related content: a cost optimization report for a similar project from three months ago, and the Finance Department's email feedback on the overspending of Plan A. Discussion Excerpt: At the 20th minute, a technician mentions that "the materials for Plan A can be replaced with domestic suppliers."

[0089] S83 Extended Display: A semi-transparent expansion box appears next to "20% Cost Overrun," with two columns inside. The left column summarizes the cost comparison with domestic suppliers (15% savings), a technical staff member's speech, and a link to the Finance Department's email regarding the overrun on Plan A. The right column charts a 3D bar chart comparing the cost structures of Plans A and B. Click the chart to view further details.

[0090] In one embodiment, Figure 4 As shown, a method for processing virtual reality conference information, in step S7, if a sudden notification is detected, summarizing the conference information before the sudden notification, generating a conference text, and providing it to the user for real-time reference, further includes:

[0091] Step S1: Setting the rendering priority of the notification pop-up window to the highest level in the UI (user interface) system, forcing the notification pop-up window to be displayed above virtual scene elements (such as 3D models and PowerPoint presentations), fixing the notification bar to the upper right corner of the user's field of view through an anchor system, and dynamically adapting the helmet's viewing angle offset;

[0092] Step S2: Add CSS animation (such as sudden pulse changes in the border and gradual transparency changes) to the sudden notification, automatically placing it on top and covering other non-critical operation interfaces when triggered; integrate directional spatial audio technology to make the prompt sound play in a direction corresponding to the position of the notification pop-up window (upper right), enhancing the sense of direction;

[0093] Step S3: Real-time detection of user operation status (such as raising a hand or writing). If the hand model or tool interface overlaps with the notification area, the notification area is adjusted to ensure information visibility. Methods for adjusting the notification area include automatically reducing the notification size and temporarily making it translucent.

[0094] Steps S1, S2, and S3 enable highlighting of sudden notifications (such as flashing borders, pop-up windows pinned to the top) and sound reminders (such as prompt sounds) in the VR conference platform settings, and fix the notification bar in an easily noticeable area of ​​the field of view (such as the upper right) to avoid being blocked by virtual scene elements. Ensure that key information is reached. The immersive characteristics of the VR environment can easily cause users to ignore sudden notifications. By fixing the position of the notification bar (anchored in the upper right field of view) and multi-sensory enhancement prompts (dynamic borders, spatial audio), and utilizing humans' physiological alertness to high-frequency changing stimuli, we can force a breakthrough in the immersive "information cocoon" and ensure that notifications are perceived in the first place.

[0095] In one embodiment, Figure 5 As shown, a method for processing virtual reality conference information, in step S7, if a sudden notification is detected, summarizing the conference information before the sudden notification, generating a conference text, and providing it to the user for real-time reference, further includes:

[0096] Step S4: Crawl internal enterprise documents (product manuals, compliance documents), historical meeting minutes, and industry standard libraries (such as the ISO official website) to extract terms (such as "ISO 9001"), clauses, and related cases; use NLP tools (such as spaCy) to identify entities and relationships, store them in a graph database (Neo4j), and form a knowledge graph of terms, clauses, and cases;

[0097] Step S5: transcribe the real-time speech in the meeting into text and detect terms using a pre-trained entity recognition model (such as BERT-NER); query the knowledge graph based on the terms, return relevant term summaries and case links, and display high-frequency citation content in a weighted manner;

[0098] Step S6: Generate a floating panel (semi-transparent) on the side of the VR interface, displaying core terms (foldable), case thumbnails, and a one-click jump entrance in layers.

[0099] In step S4, by crawling multi-source heterogeneous data (internal documents, industry standards) and constructing a knowledge graph, dispersed enterprise knowledge is transformed into an associative network. The graph structure of the graph database can efficiently express complex logical relationships between terms (such as the compliance mapping of ISO terms and cases), providing semantic support for subsequent real-time queries.

[0100] In step S5, speech recognition is combined with a pre-trained entity recognition model (such as BERT-NER) to dynamically match the meeting discussion content with the knowledge graph. Through entity extraction and graph traversal, abstract terms are automatically associated with specific cases, reducing manual search costs and ensuring standardized interpretation of professional terms.

[0101] In step S6, the floating panel adopts a layered, translucent design, avoiding obstruction of the 3D meeting scene while achieving a dynamic balance of information density through its foldable structure and thumbnails. The synergy between spatial audio and visual anchors aligns with users' spatial cognition in VR environments, allowing knowledge assistance to be naturally integrated into the immersive interaction process.

[0102] In one embodiment, Figure 6 As shown, a virtual reality conference information processing system includes:

[0103] Meeting text generation module 7 is used to summarize the meeting information before the sudden notification if an emergency notification is detected (the virtual reality meeting will automatically record the entire meeting process to facilitate subsequent review and information tracing), generate a meeting text, and provide it to users for real-time reference;

[0104] The markup expansion module 8 is configured to detect whether the user has marked text in the conference text, and if so, expand the marked text based on the conference information.

[0105] The Tag Extension Module 8 can also automatically link meeting role libraries (such as speakers and decision-makers) with task management systems, providing real-time feedback and to-do suggestions (such as "need confirmation from the finance manager") for those associated with the tag. It also recommends similar historical decision-making cases based on the knowledge graph and compares current differences in the extension box, helping users quickly build closed-loop decision chains and avoid redundant discussions.

[0106] In one embodiment, Figure 7 As shown, a virtual reality conference information processing system, the conference text generation module 7 includes:

[0107] The content capture and association unit 71 is used to store structured data (speech, voting, file sharing, etc.) according to the timeline through the conference recording function (speech-to-text + motion capture). When an emergency notification is triggered, the system automatically captures the text content of the previous N minutes (e.g., 2 minutes) and the associated operation log;

[0108] Key information extraction unit 72 is used to extract key information from the intercepted content using a lightweight model (such as lightweight BERT), identify core issues, resolution trends, and issues to be resolved; and combine voice sentiment analysis (such as tone urgency) and shared document keywords to generate a weighted summary text with priority ranking.

[0109] The text compression unit 73 is used to compress the generated text into M short sentences (such as 3, 5, etc.) as the conference text, and push it to the user through the floating panel (upper right, the panel can be semi-transparent), while retaining the original data link for backtracking.

[0110] In the text compression unit 73, a real-time user feedback mechanism is incorporated into the generation of short sentences. For example, users are allowed to instantly adjust the summary content through gestures (crossing out / highlighting) or voice ("Simplify the third item"). Based on this feedback data, the lightweight model is fine-tuned to make subsequent summaries more in line with user preferences. Furthermore, the summary's emphasis is automatically adjusted based on the user's role (e.g., manager, executor) (e.g., management focuses on resolution trends, while executives focus on unresolved issues).

[0111] In one embodiment, Figure 8 As shown, a virtual reality conference information processing system, the tag extension module 8 includes:

[0112] The tag capture unit 81 is used to capture the tagged text selected by the user, record the timestamp of the tagged text and its position in the meeting text; and extract the context of the tagged content (such as the speaker's identity, discussion progress, and related documents) in combination with the meeting records (voice transcription, shared documents, voting records);

[0113] The related content matching unit 82 is used to provide a selection of marking purposes, including questioning, supplementing, and asking questions; and matches related content in the meeting based on the marking purpose (for example, after marking "budget risk" and selecting the marking purpose as supplement, the previous cost analysis table and the corresponding discussion segment link are pushed);

[0114] The associated content display unit 83 is used to generate an expansion box (the expansion box can be semi-transparent) next to the marked position to display charts and summaries. The charts include file thumbnails and 3D data visualization charts.

[0115] The expanded frame of the associated content display unit 83 can also include a multi-person collaborative markup layer, allowing invited users to add annotations or highlight areas to charts, which are then synchronized with the perspectives of other participants in real time. Combined with spatial anchoring technology, users can drag annotations onto the surface of specific models (such as product prototypes) in the 3D meeting scene, forming a three-dimensional discussion focus, which is automatically recorded as a traceable decision node.

[0116] In one embodiment, Figure 9As shown, a virtual reality conference information processing system, the virtual reality conference information processing system also includes:

[0117] Priority setting module 1 is used to set the rendering priority of notification pop-ups to the highest level in the UI (user interface) system, forcing notification pop-ups to be displayed above virtual scene elements (such as 3D models and PowerPoint presentations). The notification bar is fixed to the upper right corner of the user's field of view through an anchor system, and dynamically adapts to the helmet's viewing angle offset.

[0118] Animation and audio prompt module 2 is used to add CSS animations (such as sudden border pulses and transparency gradients) to sudden notifications. When triggered, it automatically stays on top and covers other non-critical operation interfaces. It also integrates directional spatial audio technology to make the prompt sound play in a direction that follows the position of the notification pop-up window (upper right), enhancing directional perception.

[0119] The operation status detection module 3 is used to detect the user's operation status (such as raising a hand or writing) in real time. If the hand model or tool interface overlaps with the notification area, the notification area is adjusted to ensure information visibility. The methods for adjusting the notification area include automatically reducing the notification size and temporarily making it translucent.

[0120] Based on the real-time detection of the operation status detection module 3, the ambient light and user fatigue detection are superimposed. The helmet sensor obtains ambient brightness data and dynamically adjusts the transparency of the notification pop-up background (for example, reducing the brightness to prevent glare in dark environments). Combined with blink frequency analysis, if the user is determined to be distracted, the border animation intensity is enhanced or the prompt audio frequency is increased to ensure that the user receives key information.

[0121] In one embodiment, Figure 10 As shown, a virtual reality conference information processing system, the virtual reality conference information processing system also includes:

[0122] Knowledge graph construction module 4 is used to crawl internal enterprise documents (product manuals, compliance documents), historical meeting minutes, and industry standard libraries (such as the ISO official website) to extract terms (such as "ISO 9001"), clauses, and related cases; use NLP tools (such as spaCy) to identify entities and relationships, store them in a graph database (Neo4j), and form a knowledge graph of terms, clauses, and cases;

[0123] The knowledge graph matching module 5 is used to transcribe the real-time speech in the meeting into text and detect terms using a pre-trained entity recognition model (such as BERT-NER). It queries the knowledge graph based on the terms, returns relevant clause summaries and case links, and displays high-frequency citations in a weighted manner.

[0124] The knowledge graph display module 6 is used to generate a (semi-transparent) floating panel on the side of the VR interface, which displays core terms (foldable) in layers, case thumbnails, and one-click jump entrances.

[0125] User role weight analysis has been incorporated into the Knowledge Graph Matching Module 5. Based on historical user behavior (e.g., legal personnel frequently query compliance clauses), the information sorting within the Knowledge Graph Display Module 6 panel is dynamically adjusted, prioritizing frequently accessed categories. Interactive 3D explanation models are embedded for complex terms (e.g., clicking "Tolerance Standard" results in a floating display of a dynamic tolerance chain animation), lowering the threshold for professional understanding.

[0126] It should be understood that, although the various steps in the flow chart of each embodiment of the present invention are shown in sequence according to the indication of the arrows, these steps are not necessarily performed in sequence according to the order indicated by the arrows. Unless otherwise specified herein, the execution of these steps is not strictly limited in order, and these steps can be performed in other orders. Moreover, at least a portion of the steps in each embodiment may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these sub-steps or stages is not necessarily performed in sequence, but can be performed in turn or alternately with at least a portion of other steps or sub-steps or stages of other steps.

[0127] The technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above-mentioned embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0128] The above-described embodiments merely illustrate several implementations of the present invention, and while their descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art would be able to make numerous variations and improvements without departing from the spirit of the present invention, all of which fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be determined by the appended claims.

[0129] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

[0130] In addition, it should be understood that although this specification is described in terms of implementation methods, not every implementation method contains only one independent technical solution. This narrative method of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment can also be appropriately combined to form other implementation methods that can be understood by those skilled in the art.

Claims

1. A virtual reality conference information processing method, characterized in that: The virtual reality conference information processing method includes: If an emergency notification is detected, summarize the meeting information before the emergency notification, generate a meeting text, and provide it to the user for real-time reference; Detect whether the user has marked text in the conference text. If there is marked text, expand the marked text based on the conference information.

2. The virtual reality conference information processing method according to claim 1, characterized in that: If an unexpected notification is detected, summarizing the meeting information before the unexpected notification, generating a meeting text, and providing it to the user for real-time reference, specifically includes: The meeting record function stores structured data by timeline. When an emergency notification is triggered, the system automatically captures the text content and associated operation logs of the previous N minutes. A lightweight model is used to extract key information from intercepted content, identifying core issues, resolution trends, and pending issues. Voice sentiment analysis and shared document keywords are combined to generate weighted, prioritized summary text. The generated text is compressed into M short sentences as the meeting text and pushed to users through the floating panel, while retaining the original data link for reference.

3. The virtual reality conference information processing method according to claim 1, characterized in that: The step of detecting whether the user has marked text in the conference text, and if there is marked text, expanding the marked text based on the conference information specifically includes: Capture the marked text selected by the user, record the timestamp of the marked text and its position in the meeting text; and extract the context of the marked content in combination with the meeting records; Provides selection of marking purposes, including questioning, supplementing, and asking questions; matches relevant content in the meeting based on the marking purpose; Generate an expansion box next to the marked location to display charts and summaries. The charts include file thumbnails and 3D data visualization charts.

4. The virtual reality conference information processing method according to any one of claims 1 to 3, characterized in that: Before the step of, if an emergency notification is detected, summarizing the meeting information before the emergency notification, generating a meeting text, and providing it to the user for real-time reference, the method further includes: In the UI system, the rendering priority of notification pop-ups is set to the highest level, forcing them to appear above virtual scene elements. The notification bar is fixed to the upper right corner of the user's field of view through the anchor system, and dynamically adapts to the headset's viewing angle offset. Add CSS animations to emergency notifications, automatically placing them on top and covering other non-critical operation interfaces when triggered; integrate directional spatial audio technology to make the prompt sound play in a direction consistent with the location of the notification pop-up window, enhancing location awareness; Detect user operation status in real time. If the hand model or tool interface overlaps with the notification area, adjust the notification area to ensure information visibility. Methods for adjusting the notification area include automatically reducing the notification size and temporarily making it translucent.

5. The method for processing virtual reality conference information according to any one of claims 1 to 3, characterized in that: Before the step of, if an emergency notification is detected, summarizing the meeting information before the emergency notification, generating a meeting text, and providing it to the user for real-time reference, the method further includes: Crawl internal corporate documents, historical meeting minutes, and industry standard libraries to extract terms, clauses, and related cases; use NLP tools to identify entities and relationships, store them in a graph database, and form a knowledge graph of terms, clauses, and cases; Transcribe real-time speech in meetings into text and detect terms using a pre-trained entity recognition model. Query the knowledge graph based on the terms, return relevant clause summaries and case links, and weightedly display high-frequency citations. A floating panel is generated on the side of the VR interface, displaying core terms, case thumbnails and one-click jump entrances in layers.

6. A virtual reality conference information processing system, characterized in that: include: A meeting text generation module is used to summarize the meeting information before the sudden notification if an unexpected notification is detected, generate a meeting text, and provide it to the user for real-time reference; The markup extension module is used to detect whether the user has marked text in the conference text. If there is marked text, the marked text is expanded based on the conference information.

7. The virtual reality conference information processing system according to claim 6, characterized in that: The conference text generation module includes: The content capture and association unit is used to store structured data by timeline through the meeting record function. When an emergency notification is triggered, the system automatically captures the text content of the previous N minutes and the associated operation logs; The key information extraction unit uses a lightweight model to extract key information from intercepted content, identifying core issues, resolution trends, and pending issues. It also combines voice sentiment analysis and shared document keywords to generate weighted, prioritized summary text. The text compression unit is used to compress the generated text into M short sentences as the conference text, which is pushed to users through the floating panel while retaining the original data link for tracing back.

8. The virtual reality conference information processing system according to claim 6, characterized in that: The markup extension modules include: The tag capture unit is used to capture the tagged text selected by the user, record the timestamp of the tagged text and its position in the meeting text; and extract the context of the tagged content in combination with the meeting records; The related content matching unit is used to provide selection of marking purposes, including questioning, supplementing, and asking questions; and matches related content in the meeting based on the marking purpose; The associated content display unit is used to generate an expansion box next to the marked position to display charts and summaries. The charts include file thumbnails and 3D data visualization charts.

9. The virtual reality conference information processing system according to any one of claims 6 to 8, characterized in that: The virtual reality conference information processing system also includes: The priority setting module is used to set the rendering priority of notification pop-ups to the highest level in the UI system, forcing notification pop-ups to appear above virtual scene elements. The notification bar is fixed to the upper right corner of the user's field of view through the anchor system and dynamically adapts to the helmet's perspective offset. Animation and audio prompt module, used to add CSS animation to emergency notifications, automatically sticking them to the top and covering other non-critical operation interfaces when triggered; integrated directional spatial audio technology, so that prompt sounds are played in a direction that matches the location of the notification pop-up window, enhancing location awareness; The operation status detection module is used to detect the user's operation status in real time. If the hand model or tool interface overlaps with the notification area, the notification area is adjusted to ensure information visibility. Methods for adjusting the notification area include automatically reducing the notification size and temporarily making it translucent.

10. The virtual reality conference information processing system according to any one of claims 6 to 8, characterized in that: The virtual reality conference information processing system also includes: The knowledge graph construction module is used to crawl internal enterprise documents, historical meeting minutes, and industry standard libraries to extract terms, clauses, and related cases. It uses NLP tools to identify entities and relationships, stores them in a graph database, and forms a knowledge graph of terms, clauses, and cases. The knowledge graph matching module is used to transcribe real-time speech in meetings into text and detect terms using a pre-trained entity recognition model. It queries the knowledge graph based on terms, returns relevant clause summaries and case links, and displays high-frequency citations in a weighted manner. The knowledge graph display module is used to generate a floating panel on the side of the VR interface, displaying core terms, case thumbnails and one-click jump entrances in layers.

Images