CONFIGURABLE ROLE PLAYING AI BOT, SYSTEM AND METHOD FOR REMOTE CONFERENCING SERVICES

Abstract

The invention relates to a method for AI-driven automated real-time task management in remote conferencing systems. It involves fetching various resources such as author-defined configurations, custom files, and web applications to perform AI-driven roles in a conference. These roles include analyzing and moderating the conference for scenario-driven tasks, performing interactive actions like correcting conversations, translating languages, and managing web applications. The method also includes managerial actions like muting participants and detecting emotions, as well as concluding actions such as summarizing the meeting and creating reports. The system can operate in standalone video conferences or those embedded in web applications, with capabilities for concurrent AI roles and real-time resource enhancement.

Specification

Description:[033] The accompanying drawings are included to provide a further understanding of the present disclosure and are incorporated in and constitute a part of this specification. The drawings illustrate exemplary embodiments of the present disclosure and, together with the description, serve to explain the principles of the present disclosure.
[034] In the figures, similar components and/or features may have the same reference label. Further various components of the same type may be distinguished by following the reference label with a second label that distinguishes among the similar components. If only the first reference label is used in the specification, the description is applicable to any of the similar components having the same reference label irrespective of the second reference label.
[035] FIG. 1 illustrates a basic block diagram of a system for AI-driven automated real-time task management in remote conferencing systems, according to an embodiment of the present invention.
[036] FIG. 2 illustrates a flow chart of a method for AI-driven automated real-time task management in remote conferencing systems, according to an embodiment of the present invention.
[037] FIG. 3 shows a flow chart depicting one or more ways by which a plurality of tasks may be triggered, according to an embodiment of the present invention.
[038] FIG. 4 illustrates a flow chart of a plurality of scenario-driven tasks, according to an exemplary embodiment of the present invention.

DETAILED DESCRIPTION OF THE INVENTION
[039] The present invention relates to a configurable bot, system and method for a generative AI driven automated real-time task management in remote conferencing systems for independently performing specific roles in the remote conferencing systems based on user preference and behavior analysis.
[040] Referring to FIG. 1, a basic block diagram of a system (100) for AI-driven automated real-time task management in remote conferencing systems is illustrated. This system (100) may include components such as one or more processors (101), one or more memory units (102), a server (103), a cloud storage server (104), Large Language Models (LLM) (105), an assistive bot (106), a transcription service (107), and an audio gateway module (108). These components may work together to perform a range of managerial actions, such as activating or deactivating the transcriber, adding/removing participants, muting or unmuting participants, converting speech to text, and converting text to speech. Additionally, the system (100) may detect faces and emotions, among other functions.
[041] As indicated in FIG. 2, a flowchart of a method for AI-driven automated real-time task management in remote conferencing systems is shown. The method may involve fetching a plurality of resources, including configurations, custom files, transcriptable audio modules, web applications, and application programming interfaces. The resources may be used to perform one or more artificial intelligence-driven roles in a remote conference involving participants. The system (100) may utilize a server (103) and a cloud storage server (104) to store and manage the resources, while Large Language Models (LLM) (105) may be employed to analyze conversations and direct task execution requests received via a transcription service (107). The server (103) may run continuously, handling voice transmissions and connecting to a transcription service (107) for speech-to-text conversion, which may be labeled to the speaker of the speech. This setup may allow the system (100) to leverage existing conferencing solutions, such as Jitsi or Google Meet, to develop configurable bots that can moderate the conference room.
[042] In some embodiments, the assistive bot (106), which may be configured by the bot author, may play various roles such as a teacher, project manager, or audio translator. The bot may utilize custom knowledge uploaded to the cloud storage server (104) to enhance its functionality. The bot may also be capable of managing multiple participants concurrently, a feature that distinguishes it from existing bots designed for one-to-one interactions. The bot's attention mechanism may be controlled to toggle its active state, reducing costs associated with keeping the bot active at all times. One or more trigger words (such as 'help', 'assist', 'listen', 'can you', 'mona', 'moon a', 'moan ah', 'my na', 'research' ) may be used to activate the bot, allowing it to listen and respond appropriately. The bot's proactivity mechanism may determine whether it should interrupt a conversation based on a proactivity factor, which may be configured by the bot author.
[043] A person skilled in the art would acknowledge the fact that the task of text-to-speech and speech-to-text conversion may be removed if the generative artificial intelligence is capable of traversing this action. This would allow the system (100) to function without these conversions, leveraging advanced Multi-Modal Large Language Models (LLM) (105) capabilities that recognize speech directly.
[044] Referring to FIG. 3, a flowchart depicting the trigger mechanisms is illustrated. The system (100) may detect a trigger from a participant's speech, which can activate a desired task based on the identified trigger. This step may enable the system (100) to respond dynamically to the participants' needs during a remote conference. The system (100) may utilize a combination of trigger words, which are pre-defined in the configuration, to activate the bot's attention. When these trigger words are detected in the speech, the bot may become attentive and ready to perform its role. The Large Language Models (LLM) (105) may be employed to analyze the conversation and determine the appropriate response or action to take. This mechanism may allow the bot to manage its active state effectively, ensuring that it only responds when necessary, thereby optimizing resource usage. The bot's proactivity mechanism may also play a role in this step, where the bot can decide whether to interrupt a conversation based on its proactivity factor. This factor may be a probabilistic value that determines the likelihood of the bot taking action. The system (100) may also leverage transcription services (107) to convert speech to text, which can then be analyzed by the Large Language Models (LLM) (105) to identify the trigger words. This integration of speech-to-text conversion and LLM analysis may enhance the bot's ability to respond accurately and promptly to the participants' needs. The server (103) and cloud storage server (104) may support this process by storing and managing the necessary configurations and custom knowledge that the bot may require to perform its tasks effectively.
[045] Referring to FIG. 4, a flowchart of scenario-driven tasks is shown. The system (100) may perform a variety of interactive actions, such as adding information, correcting conversations, and pointing out factual inaccuracies. The bot may also rephrase or clarify statements, translate conversations into other languages, and ask for clarifications or confirmations. Additionally, the bot may redirect or contradict conversations, summarize discussions, and prevent mistakes. It may show empathy, ask questions, and interrupt for time management. The bot may also supplement information, point out redundant information, and interrupt to balance conversations or bring attention to matters. The bot may prompt participants to respond or engage upon detecting a duration of silence. In some embodiments, managerial actions may also be performed, such as activating or deactivating the transcriber, removing participants, and muting or unmuting participants. The system (100) may manage a variety of web applications, opening and interacting with them, and optionally sharing the web application screen with participants. Smooth transitioning between video play mode and discussion mode may be facilitated, allowing participants to engage interactively. The bot may execute tabScripts to control web applications, synchronize webpage instances, and open one or more tabs to interact with it.
[046] As indicated in FIG. 4, concluding actions may include ending the meeting, summarizing the entire meeting, and creating extensive reports. The system (100) may control video playback and discuss mode, start the browser and open the video conferencing room URL, and handle text-to-speech conversion. The assistive bot (106) and one or more processors (101), and may be involved in these processes, ensuring a comprehensive and efficient management of the remote conferencing system (100).
[047] In some embodiments, a person having average skill in the art would acknowledge the fact that the method for AI-driven automated real-time task management in remote conferencing systems may involve the configuration of a remote conference as either a standalone video conference or one embedded onto a webpage or web application. This configuration may be facilitated by a server (103) and a cloud storage server (104), which may work in conjunction with Large Language Models (LLM) (105) to support the AI-driven roles within the conference. The server (103) may host the necessary resources and configurations, while the cloud storage server (104) may store custom knowledge that the AI bot can utilize. The Large Language Models (LLM) (105) may analyze conversations and user speech, leveraging transcription services (107) to convert speech to text, thereby enabling the AI bot to perform its roles effectively. The AI bot may play various roles, such as a teacher, scrum master, or project manager, and may moderate the conference by managing audio communication, turning the transcriber on or off, and muting participants as needed. The bot's attention may be toggled based on trigger words, and its proactivity may be determined by a proactivity factor, which may influence whether the bot interrupts ongoing conversations. This setup may allow for a flexible and dynamic conferencing environment, where the AI bot can adapt to different scenarios and participant interactions, enhancing the overall conference experience.
[048] In some embodiments, the method for AI-driven automated real-time task management in remote conferencing systems may involve the creation of a conference embedded onto a webpage or web application through a browser extension feature or any analogous functionality. This step may potentially utilize a server (103) and a cloud storage server (104), which may be integral to the system's (100) operation. The server (103) may facilitate the hosting of the conference, while the cloud storage server (104) may store necessary configurations and data. Large Language Models (LLM) (105) may be employed to analyze and moderate the conference, leveraging their capabilities to understand and process the ongoing interactions.
[049] In some embodiments, the browser extension feature may serve as a component, enabling the integration of the conference into a webpage or web application. This feature may allow participants to access the conference directly through their web browsers, enhancing accessibility and user experience. The extension may also facilitate the synchronization of web application instances, ensuring that all participants have a consistent view and interaction with the conference content.
[050] In some embodiments, the system (100) may further incorporate functionalities such as the execution of tabScripts to control web applications, enabling automated interactions and coordination of web app states. This capability may enhance the overall conference interaction, allowing for a more dynamic and engaging experience for participants. Additionally, the system (100) may manage audio communication and convert speech to text, utilizing transcription services (107) to provide real-time captions and enhance accessibility.
[051] In some embodiments, this step may describe a method for embedding a conference onto a webpage or web application, leveraging advanced technologies and features to facilitate seamless and interactive remote conferencing experiences.
[052] In some embodiments, the method for AI-driven automated real-time task management in remote conferencing systems may involve the performance of artificial intelligence-driven roles on a conference embedded onto a webpage or web application. This performance may occur upon receiving permission from one or more participants. The system (100) may utilize one or more processors (101) and one or more memory units (102) to facilitate this process, potentially involving a server (103), a cloud storage server (104), and Large Language Models (LLM) (105). The server (103) may run continuously, handling voice transmissions and connecting to a transcription service (107) for speech-to-text conversion. The Large Language Models (LLM) (105) may analyze conversations and direct task execution requests, responding according to its role and settings. The assistive bot (106) may moderate the room, manage audio communication, and perform various roles such as a teacher, scrum master, or project manager. The bot may leverage transcription services (107) within existing conferencing solutions, utilizing custom knowledge uploaded to the cloud storage server (104). The bot's attention may be toggled using trigger words, and its proactivity may be controlled by a proactivity factor, determining whether the bot should interrupt a conversation. The bot may also manage and moderate conference sessions, control its proactive behavior, and respond according to its role and settings. The system (100) may ensure that the AI-driven roles are executed effectively, analyzing and moderating the conference for scenario-driven tasks.
[053] In some embodiments, the process may involve detecting a trigger from a participant's speech, which can activate a desired task based on the identified trigger. This step may enable the system (100) to respond dynamically to the participants' needs during a remote conference. The system (100) may utilize a combination of trigger words, which are pre-defined in the configuration, to activate the bot's attention. When these trigger words are detected in the speech, the bot may become attentive and ready to perform its role. The Large Language Models (LLM) (105) may be employed to analyze the conversation and determine the appropriate response or action to take. This mechanism may allow the bot to manage its active state effectively, ensuring that it only responds when necessary, thereby optimizing resource usage. The bot's proactivity mechanism may also play a role in this step, where the bot can decide whether to interrupt a conversation based on its proactivity factor. This factor may be a probabilistic value that determines the likelihood of the bot taking action. The system (100) may also leverage transcription services (107) to convert speech to text, which can then be analyzed by the Large Language Models (LLM) (105) to identify the trigger words. This integration of speech-to-text conversion and LLM analysis may enhance the bot's ability to respond accurately and promptly to the participants' needs. The server (103) and cloud storage server (104) may support this process by storing and managing the necessary configurations and custom knowledge that the bot may require to perform its tasks effectively. Overall, this step may illustrate an interaction between various components of the system (100), enabling a responsive remote conferencing experience.
[054] In some embodiments, the process may involve comparing a value defined for a variety of scenario-driven tasks against a author-defined threshold. This comparison may serve as a basis for selectively choosing a task to interrupt an ongoing conversation in the meeting. The system (100) may utilize one or more processors (101) and one or more memory units (102) to facilitate this comparison, potentially leveraging a server (103), a cloud storage server (104), and Large Language Models (LLM) (105) to analyze the ongoing conversation and user speech. The assistive bot (106), configured as a Configurable Role-playing AI bot, may play a role in this process. The bot may utilize trigger words to activate its attention and may employ a proactivity factor to determine whether to interrupt the conversation. This mechanism may allow the bot to manage and moderate conference sessions effectively, ensuring that the conversation remains on track and relevant to the participants' needs. The bot's ability to toggle its attention and respond according to its role and settings may enhance the functionality of the remote conferencing system (100), providing an interactive experience for all participants involved.
[055] In some embodiments, the focus is on defining a variety of timeouts for managing and regulating real-time task management within a remote conferencing system (100). This step may involve the configuration of various timeouts for the operation of artificial intelligence-driven roles during a conference. The timeouts may serve to manage the bot's attention and responsiveness, ensuring that the system (100) remains efficient and responsive to the needs of the participants. The system (100) may utilize a proactivity factor to determine the bot's level of engagement, allowing it to decide when to interrupt or remain silent based on the ongoing conversation. The bot's attention mechanism may be toggled using pre-defined trigger words, which may activate or deactivate the bot's active listening mode. This mechanism may help in conserving resources by keeping the bot inactive when not needed, while still allowing for prompt engagement when required. The timeouts may also include bot gaps, which are specific intervals that dictate the bot's attentiveness and response timing. These gaps may be configured to ensure that the bot remains attentive for a certain duration after a response, allowing participants to interact naturally without unnecessary interruptions. Additionally, the system (100) may incorporate mechanisms to handle longer periods of silence, prompting participants to engage or providing feedback to maintain the flow of the conference. The implementation of these timeouts may be for maintaining a balanced interaction between the bot and the participants, ensuring that the AI-driven roles are performed in alignment with the conference's objectives.
[056] In some embodiments, the system (100) may be designed to facilitate the concurrent execution of multiple artificial intelligence-driven roles within a single meeting. This capability may be supported by a server (103) infrastructure that operates continuously in a cloud environment, alongside an audio gateway module (108) that manages voice transmissions and integrates with transcription services (107) for speech-to-text conversion. The system (100) may leverage Large Language Models (LLM) (105) to analyze conversations and respond according to pre-defined roles and settings. These roles may include, but are not limited to, a teacher, project manager, or audio translator, each potentially configured through small changes in environment and configuration files. The assistive bot (106), which may act as a moderator, can perform various managerial actions such as activating or deactivating the transcriber, muting or unmuting participants, and managing the overall conference session. The bot's attention mechanism may be controlled through trigger words, allowing it to become attentive and respond when necessary. Additionally, the bot's proactivity mechanism may enable it to interrupt user speech based on a probabilistic decision-making process, ensuring that the conversation remains on track and that all participants are engaged. The system's (100) architecture may allow for the integration of multiple bots, each capable of managing interactions with multiple participants simultaneously, thereby enhancing the efficiency and effectiveness of the remote conferencing experience.
[057] In some embodiments, the method for AI-driven automated real-time task management in remote conferencing systems may involve the execution of one or more roles, which can include but are not limited to a moderator, teacher, project manager, or scrum master. These roles may be facilitated by the system's (100) one or more processors (101) and one or more memory units (102), which are configured to perform a set of instructions that enable the artificial intelligence-driven roles to function within a remote conference. The system (100) may utilize a server (103), a cloud storage server (104), and Large Language Models (LLM) (105) to support these roles, allowing for the analysis and moderation of the conference based on various scenario-driven tasks. The roles may be designed to enhance the conference experience by performing interactive actions such as adding information, correcting conversations, and pointing out factual inaccuracies. Additionally, the roles may manage and moderate conference sessions by controlling the bot's behavior and responding according to pre-defined settings. The system (100) may also leverage transcription services (107) to convert speech to text, enabling the roles to analyze ongoing conversations and user speech. The roles may be configured to perform scenario-driven tasks for multiple participants simultaneously, ensuring that the conference is managed efficiently. The system's (100) ability to run multiple artificial intelligence-driven roles concurrently in a single meeting may further enhance its capability to manage complex conference scenarios.
[058] In some embodiments, the focus is on configuring roles to perform scenario-driven tasks for multiple participants simultaneously within a remote conferencing system (100). The system (100) may utilize one or more processors (101) and one or more memory units (102) to manage the execution of these tasks, leveraging a server (103) and a cloud storage server (104) to facilitate the process. The Large Language Models (LLM) (105) may play a role in analyzing conversations and user speech, enabling the system (100) to perform various roles such as a teacher, scrum master, project manager, or audio translator. The assistive bot (106), configured by the bot author, may utilize transcription services (107) to convert speech to text, allowing for real-time interaction and moderation of the conference. The bot's attention mechanism may be toggled using trigger words, ensuring that the bot remains attentive and responsive to the participants' needs. The proactivity mechanism may determine the bot's response, using a proactivity factor to decide when to interrupt or engage in the conversation. This setup may allow the system (100) to manage and moderate conference sessions, providing an experience for all participants involved.
[059] In some embodiments, the method for AI-driven automated real-time task management in remote conferencing systems may involve the analysis and moderation of the conference based on contextual awareness of conversation topics, temporal awareness of session duration, and social awareness of participant engagement. The system (100) may utilize one or more processors (101) and one or more memory units (102) to perform these tasks, leveraging a server (103), a cloud storage server (104), and Large Language Models (LLM) (105) to enhance its capabilities. The contextual awareness may allow the system (100) to understand the main topics and subtopics of the conversation, enabling it to moderate discussions. Temporal awareness may help the system (100) track the duration of the session, ensuring that the conference runs smoothly. Social awareness may enable the system (100) to monitor participant engagement, identifying who is contributing to the conversation and who may need encouragement to participate. The system (100) may also utilize a transcription service (107) to convert speech to text, providing a real-time record of the conversation that can be analyzed for further insights. Additionally, the system (100) may employ a bot attention mechanism to manage the bot's active state, ensuring that it responds appropriately to the ongoing conversation. The bot may use trigger words to activate its attention, allowing it to engage with participants when necessary. The proactivity mechanism may enable the bot to interrupt conversations when needed, using a proactivity factor to determine the likelihood of interruption. This approach to conference management may enhance the experience for participants, providing an interactive environment.
[060] In some embodiments, the focus is on the capability of participants to transfer additional resources to the AI-driven roles in real-time, thereby enhancing or specializing the knowledge of these roles within a remote conferencing system (100). This step may involve the interaction between participants and the system (100), where participants can provide supplementary data or configurations that the AI roles can utilize to improve their performance during the conference. The system (100) may be designed to accept various forms of input, such as documents, configurations, or other relevant data, which can be integrated into the AI's operational framework. This integration may allow the AI to adapt its behavior and responses based on the newly acquired information, thereby providing a tailored and contextually aware interaction within the conference. The process may involve the use of a cloud storage server (104) to store and manage these additional resources, ensuring that the AI roles have access to the most up-to-date information. The Large Language Models (LLM) (105) may play a role in analyzing and incorporating this data into the AI's decision-making processes, allowing for a dynamic and responsive conferencing experience. This capability may enhance the effectiveness of the AI-driven roles, enabling them to perform accurately in various scenarios presented during the conference.
[061] In some embodiments, the focus is on the method for AI-driven automated real-time task management in remote conferencing systems, specifically concerning the management of a variety of web applications. This step may involve the utilization of various web applications such as video, presentation, audio, document, image, file, social media app, messaging app, gaming app, online learning app, e-book, PDF reader, graphic design app, online whiteboard app, infographic, blog, or similar applications. The system (100) may be designed to handle these applications in a manner that allows for integration and interaction within a remote conferencing environment. The process may include opening and interacting with these web applications, optionally sharing the web application screen with participants in the meeting, and ensuring a transition between video play mode and discussion mode. This transition may be based on prompts activated by participants or configurations. The system (100) may allow participants to engage interactively while using the web application, enhancing the conferencing experience. The role of the system (100) may be to manage these applications, ensuring that they are utilized to their potential in a remote conferencing setting. The system (100) may employ various techniques, such as executing tabScripts to control web apps, synchronizing webpage instances, and opening a second tab to interact with it, to automate web app interactions, enhance conference interaction, and coordinate web app state. This approach may provide a solution for managing web applications in a remote conferencing environment, allowing for an interactive experience for participants.
[062] In some embodiments, the transition between video play mode and discussion mode may be facilitated based on prompts activated by participants or configurations. This transition can be managed by the system (100), which may include components such as a server (103), a cloud storage server (104), and Large Language Models (LLM) (105). The server (103) may host the conferencing system (100), while the cloud storage server (104) may store configurations and custom files that inform the bot's behavior. The Large Language Models (LLM) (105) may analyze conversations and respond appropriately, playing roles such as a teacher or project manager.
[063] In some embodiments, the transition between modes may involve managing web applications, where the system (100) may open and interact with web applications, share screens, and allow participants to engage interactively. This interaction may be automated through executable tabScripts, executable scripts that control web applications and synchronize webpage instances. The system (100) may open a second tab to interact with web applications, enhancing conference interaction and coordinating the web app state.
[064] In some embodiments, the transition may also involve the use of a video player, which may be controlled to pause and resume playback based on the discussion mode. The video player may transmit the current timestamp to the bot, which may use this information to manage the discuss mode session. The bot may engage participants by initiating prompts at author-defined intervals, encouraging discussion and interaction.
[065] In some embodiments, the system (100) may utilize a bi-directional communication protocol like websocket connection to facilitate communication between the main bot module and the video player, ensuring smooth transitions between video play and discussion modes. This setup may allow the bot to manage video sessions, control playback, and handle text-to-speech conversion, providing a seamless experience for participants.
[066] In some embodiments, the system (100) may perform a series of concluding actions to facilitate the closure of a meeting. The one or more processors (101) and one or more memory units (102) may be utilized to execute these actions, which can include ending the meeting, summarizing the entire meeting, and creating extensive reports. The main bot module may initiate the conference session by launching a client interface, such as an automated browser, to access the communication session URL. The speaker module may handle the conversion of text to speech, allowing for audio communication during the meeting. The video player may control video playback and manage the discuss mode, which can be activated to allow participants to converse and interact. The ppt.json file may provide structured data to guide the video player in pausing at specific points and initiating discussion prompts. These actions may be coordinated to ensure a transition from active meeting participation to the generation of meeting summaries and reports. The system (100) may leverage the capabilities of the one or more processors (101) and one or more memory units (102) to manage these tasks, ensuring that the meeting concludes with a record of the proceedings.
[067] According to an alternative embodiment, some embodiments of the present invention may be in the form of a non-transitory storage medium/platform having instructions stored thereon, that when executed by a machine, enables the machine to perform the afore-discussed method.
[068] While embodiments of the present disclosure have been illustrated and described, it will be clear that the disclosure is not limited to these embodiments only. Numerous modifications, changes, variations, substitutions, and equivalents will be apparent to those skilled in the art, without departing from the scope of the disclosure, as described in the claims.
, Claims:We Claim:
1) A method for AI-driven automated real-time task management in remote conferencing systems, comprising steps of:
fetching a plurality of resources including but not limited to a plurality of author-defined configurations, custom files, audio modules, web applications and application programming interfaces required to perform one or more artificial intelligence driven role(s) in a remote conference of one or more participant(s); and
based on one or more prompt(s) activated by the participant(s) and the plurality of resources, performing the artificial intelligence driven role(s) in the remote conference, wherein the role(s) analyze and moderate the conference for a plurality of scenario-driven tasks, further comprising:
performing a plurality of interactive actions such as adding information, correcting conversation, pointing at factual inaccuracies/errors, rephrasing/clarifying, translating into other languages, asking for clarifications and/or confirmations, redirecting and/or contradicting conversations, summarizing, preventing mistakes in conversations, showing empathy in conversations, asking questions, interrupting for time management, supplementing information, pointing out on redundant information, interrupting to balance conversations; interrupting to bring attention to something important, prompting the one or more participants to respond/engage upon spotting a duration of silence and alike;
performing a plurality of managerial actions such as activation/deactivation of the transcriber, kicking off abusive participants, muting/unmuting participants, speech-to-text conversion, and text-to-speech conversion, detecting faces/emotions, and alike;
managing a plurality of web applications, further comprising: opening and interacting with the web applications, optionally sharing the web application screen with the participants in the meeting; and smooth transitioning between video play mode and discussion mode while using the web application, and allowing the participants to engage interactively while using the web application; and
performing a plurality of concluding actions such as ending the meeting, summarizing the entire meeting and creating extensive reports and alike.
2) The method for AI-driven automated real-time task management in remote conferencing systems, as claimed in claim 1, wherein the remote conference may be a standalone video conference or a conference embedded onto a webpage/web application.
3) The method for AI-driven automated real-time task management in remote conferencing systems, as claimed in claim 2, wherein the conference embedded onto a webpage/web application is created through a browser extension feature or any analogous functionality.
4) The method for AI-driven automated real-time task management in remote conferencing systems, as claimed in claim 2, wherein the artificial intelligence driven role(s) perform the task(s) on the conference embedded onto a webpage/web application upon being permitted by the one or more participant(s).
5) The method for AI-driven automated real-time task management in remote conferencing systems, as claimed in claim 1, further comprising:
detecting a trigger from a speech by the participant(s); and
attending back with a desired task based on the trigger.
6) The method for AI-driven automated real-time task management in remote conferencing systems, as claimed in claim 1, further comprising:
comparing a value defined to the plurality of scenario-driven tasks against an author-defined threshold; and
selectively choosing the task to interrupt an ongoing conversation in the meeting.
7) The method for AI-driven automated real-time task management in remote conferencing systems, as claimed in claim 1, further comprising defining a plurality of timeouts of the role(s) for managing and regulating the real-time task management.
8) The method for AI-driven automated real-time task management in remote conferencing systems, as claimed in claim 1, wherein the method is capable of running multiple artificial intelligence driven roles concurrently in a single meeting.
9) The method for AI-driven automated real-time task management in remote conferencing systems, as claimed in claim 1, wherein the one or more role(s) may be including but not limited to as a moderator, teacher, project manager, scrum master, and the like.
10) The method for AI-driven automated real-time task management in remote conferencing systems, as claimed in claim 1, wherein the role(s) are configured to perform the scenario-driven tasks for multiple participants simultaneously.
11) The method for AI-driven automated real-time task management in remote conferencing systems, as claimed in claim 1, further comprising analysis and moderation of the conference based on contextual awareness of conversation topics, temporal awareness of session duration, social awareness of participant engagement and the like.
12) The method for AI-driven automated real-time task management in remote conferencing systems, as claimed in claim 1, wherein the participant(s) are capable of transferring additional resource(s) to the one or more role(s) in real-time to enhance/ specialize knowledge thereof.
13) The method for AI-driven automated real-time task management in remote conferencing systems, as claimed in claim 1, wherein the plurality of web applications may be video, presentation, audio, document, image, file, social media app, messaging app, gaming app, digital avatar, online learning app, e-book, PDF reader, graphic design app, online whiteboard app, infographic, blog or the like.
14) The method for AI-driven automated real-time task management in remote conferencing systems, as claimed in claim 1, further comprising transitioning between video play mode and discussion mode based on a prompt activated by the participant(s) and/or author-defined configurations.
15) A system (100) for AI-driven automated real-time task management in remote conferencing systems, comprising:
one or more processors (101) configured to perform a set of instructions; and
one or more memory units (102), operably associated with the one or more processors (101), configured to store the set of instructions;
wherein the one or more processors (101) are configured to perform a process comprising:
fetching a plurality of resources including but not limited to a plurality of author-defined configurations, custom files, audio modules, web applications and application programming interfaces required to perform one or more artificial intelligence driven role(s) in a remote conference of one or more participant(s); and
based on one or more prompt(s) activated by the participant(s) and the plurality of resources, performing the artificial intelligence driven role(s) in the remote conference, wherein the role(s) analyze and moderate the conference for a plurality of scenario-driven tasks, further comprising:
performing a plurality of interactive actions such as adding information, correcting conversation, pointing at factual inaccuracies/errors, rephrasing/clarifying, translating into other languages, asking for clarifications and/or confirmations, redirecting and/or contradicting conversations, summarizing, preventing mistakes in conversations, showing empathy in conversations, asking questions, interrupting for time management, supplementing information, pointing out on redundant information, interrupting to balance conversations; interrupting to bring attention to something important, prompting the one or more participants to respond/engage upon spotting a duration of silence and alike;
performing a plurality of managerial actions such as activation/deactivation of the transcriber, kicking off abusive participants, muting/unmuting participants, speech-to-text conversion, and text-to-speech conversion, detecting faces/emotions, and alike;
managing a plurality of web applications, further comprising: opening and interacting with the web applications, optionally sharing the web application screen with the participants in the meeting; and smooth transitioning between video play mode and discussion mode while using the web application, and allowing the participants to engage interactively while using the web application; and
performing a plurality of concluding actions such as ending the meeting, summarizing the entire meeting and creating extensive reports and alike.
16) A non-transitory storage medium/platform having instructions stored thereon, that when executed by a machine, enables the machine to perform a process, comprising steps of:
fetching a plurality of resources including but not limited to a plurality of author-defined configurations, custom files, audio modules, web applications and application programming interfaces required to perform one or more artificial intelligence driven role(s) in a remote conference of one or more participant(s); and
based on one or more prompt(s) activated by the participant(s) and the plurality of resources, performing the artificial intelligence driven role(s) in the remote conference, wherein the role(s) analyze and moderate the conference for a plurality of scenario-driven tasks, further comprising:
performing a plurality of interactive actions such as adding information, correcting conversation, pointing at factual inaccuracies/errors, rephrasing/clarifying, translating into other languages, asking for clarifications and/or confirmations, redirecting and/or contradicting conversations, summarizing, preventing mistakes in conversations, showing empathy in conversations, asking questions, interrupting for time management, supplementing information, pointing out on redundant information, interrupting to balance conversations; interrupting to bring attention to something important, prompting the one or more participants to respond/engage upon spotting a duration of silence and alike;
performing a plurality of managerial actions such as activation/deactivation of the transcriber, kicking off abusive participants, muting/unmuting participants, speech-to-text conversion, and text-to-speech conversion, detecting faces/emotions, and alike;
managing a plurality of web applications, further comprising: opening and interacting with the web applications, optionally sharing the web application screen with the participants in the meeting; and smooth transitioning between video play mode and discussion mode while using the web application, and allowing the participants to engage interactively while using the web application; and
performing a plurality of concluding actions such as ending the meeting, summarizing the entire meeting and creating extensive reports and alike.

Documents