Abstract: An audio stream is subjected to speech-to-text processing in order to obtain a textual representation of the audio stream. Hierarchical topic extraction is performed on the textual representation to obtain a multi-level hierarchical topic representation of the textual representation. A user interface actuator is generated, which allows a user to search through the audio stream. Different levels of the multi-level hierarchical topic representation are displayed to the user, based upon the speed of actuation of the user interface actuator.

Abstract: A meeting application server and a method for real-time content of interest detection and notification for a meeting are described herein. The meeting application server includes a processor and a computer-readable storage medium operatively coupled to the processor. The computer-readable storage medium includes computer-executable instructions that cause the processor to receive, via a remote computing system, content of interest data for a meeting. The computer-executable instructions also cause the processor to analyze the content of interest data to determine a theme of interest for the meeting, train a classification model for the theme of interest, and generate real-time meeting data for the meeting. The computer-executable instructions further cause the processor to determine the probability that the real-time meeting data relate to the theme of interest using the classification model and, if the probability exceeds a threshold value, transmit a content of interest alert to the remote computing system.

Abstract: Systems and methods for enhanced classification of sequences of objects based on clique similarity and metadata associated with the sequences are presented. Sequences are received. Events are detected based on analyzing k-skip-n-grams included in the sequences. For each event of the detected plurality of events, a graph is generated. The graph for a particular event includes z-cliques that correspond to portions of the k-skip-n-grams that are included in the sequences that are associated with the particular event. A first sequence, which is separate from the other sequences, is received. The first sequence includes a first plurality of k-skip-n-grams. A trained classifier is employed to classify the first sequence as being associated with a first event of the detected events. Classifying the first sequence is based on a comparison between the first plurality of k-skip-n-grams and the z-cliques of the graph that is generated for the first event.

Abstract: A meeting application server and a method for real-time content of interest detection and notification for a meeting are described herein. The meeting application server includes a processor and a computer-readable storage medium operatively coupled to the processor. The computer-readable storage medium includes computer-executable instructions that cause the processor to receive, via a remote computing system, content of interest data for a meeting. The computer-executable instructions also cause the processor to analyze the content of interest data to determine a theme of interest for the meeting, train a classification model for the theme of interest, and generate real-time meeting data for the meeting. The computer-executable instructions further cause the processor to determine the probability that the real-time meeting data relate to the theme of interest using the classification model and, if the probability exceeds a threshold value, transmit a content of interest alert to the remote computing system.

Abstract: An intrusion detection system detects when an unexpected person enters the environment of a user who is in a meeting. A privacy protection action which is an action that is to be taken in response to the detected intrusion, is identified. Audio and/or video systems are then controlled to perform the privacy protection action. Machine learning can be used, based upon user interactions, to improve intrusion detection and other parts of the system.

Abstract: Computerized systems are provided for applying data indicative of a personal style to a feature of a user represented in one or more images based on determining or estimating the personal style. In operation, embodiments can receive a first image of a first user that indicates the personal style of the first user. The first image can then be fed to one or more machine learning models in order to learn and capture the personal style of the first user. Subsequently, some embodiments capture the first user in another image or set of images. Some embodiments can then detect one or more features of the first user in these other images and based on the determining of the user's personal style in the first image, can apply data indicative of the personal style of the first user to the one or more features of the user in these other images.

Abstract: Computerized systems are provided for determining an identity of one or more users that use a same audio source, such as a microphone. The identity of one or more users that use a same audio source can based on generating a list of participant candidates who are likely to participate in an associated event, such as a meeting. For instance, embodiments can generate one or more network graphs of a meeting invitee any only voice input samples of the meeting invitee's N closest connections are compared to an utterance to determine the identity of the user associated with the utterance. One or more indicators that identify the users who are using the same audio source, as well as additional information or metadata associated with the identified user can be caused to be presented.

Abstract: An audio stream is subjected to speech-to-text processing in order to obtain a textual representation of the audio stream. Hierarchical topic extraction is performed on the textual representation to obtain a multi-level hierarchical topic representation of the textual representation. A user interface actuator is generated, which allows a user to search through the audio stream. Different levels of the multi-level hierarchical topic representation are displayed to the user, based upon the speed of actuation of the user interface actuator.

Abstract: In the present disclosure, artificial intelligence (AI) processing is trained and leveraged to learn user-specific insights that are contextually relevant to a state of a user communication. Contextual information about a state of a user communication may be collected and analyzed. That contextual information may be cross-referenced with an extensive knowledge graph that is constructed from user context data. Exemplary AI processing may further be trained to apply a relevance analysis to assist with processing described herein including generation and curation of data insights that are most relevant to a state of a user communication. In some examples, the data insight generation process may be augmented by pre-generating data insights that may be relevant to a user communication prior to occurrence of the user communication. Further technical examples pertain to the rendering and presentation of representations of data insights through a graphical user interface (GUI).

Abstract: An audio stream is subjected to speech-to-text processing in order to obtain a textual representation of the audio stream. Hierarchical topic extraction is performed on the textual representation to obtain a multi-level hierarchical topic representation of the textual representation. A user interface actuator is generated, which allows a user to search through the audio stream. Different levels of the multi-level hierarchical topic representation are displayed to the user, based upon the speed of actuation of the user interface actuator.

Abstract: An intrusion detection system detects when an unexpected person enters the environment of a user who is in a meeting. A privacy protection action which is an action that is to be taken in response to the detected intrusion, is identified. Audio and/or video systems are then controlled to perform the privacy protection action. Machine learning can be used, based upon user interactions, to improve intrusion detection and other parts of the system.

Abstract: Techniques performed by a data processing system for predicting availability of a user include receiving, from a first computing device over a network connection, a first request for predicted availability of a first user to participate in an online communication session, wherein the first request includes an identifier associated with the first user and a time period for the predicted availability of the user, in response to receiving the first request, determining a first predicted availability of the first user over the predicted time period using a first machine learning model trained with user information from a plurality of data sources, the user information being indicative of when the user is likely to be available to participate in the online communication session, and sending, to the first computing device over the network connection, availability information including the first predicted availability of the first user.

Abstract: Computerized systems are provided for determining an identity of one or more users that use a same audio source, such as a microphone. The identity of one or more users that use a same audio source can based on generating a list of participant candidates who are likely to participate in an associated event, such as a meeting. For instance, embodiments can generate one or more network graphs of a meeting invitee any only voice input samples of the meeting invitee's N closest connections are compared to an utterance to determine the identity of the user associated with the utterance. One or more indicators that identify the users who are using the same audio source, as well as additional information or metadata associated with the identified user can be caused to be presented.

Abstract: Techniques performed by a data processing system for predicting availability of a user include receiving, from a first computing device over a network connection, a first request for predicted availability of a first user to participate in an online communication session, wherein the first request includes an identifier associated with the first user and a time period for the predicted availability of the user, in response to receiving the first request, determining a first predicted availability of the first user over the predicted time period using a first machine learning model trained with user information from a plurality of data sources, the user information being indicative of when the user is likely to be available to participate in the online communication session, and sending, to the first computing device over the network connection, availability information including the first predicted availability of the first user.

Abstract: In the present disclosure, artificial intelligence (AI) processing is trained and leveraged to learn user-specific insights that are contextually relevant to a state of a user communication. Contextual information about a state of a user communication may be collected and analyzed. That contextual information may be cross-referenced with an extensive knowledge graph that is constructed from user context data. Exemplary AI processing may further be trained to apply a relevance analysis to assist with processing described herein including generation and curation of data insights that are most relevant to a state of a user communication. In some examples, the data insight generation process may be augmented by pre-generating data insights that may be relevant to a user communication prior to occurrence of the user communication. Further technical examples pertain to the rendering and presentation of representations of data insights through a graphical user interface (GUI).

Abstract: Technologies are provided for the monitoring, detection, and notification of emerging, related issues within a system, which may indicate a problem. Within a computing-security system, a sudden increase in the frequency of events associated with unauthorized logon attempts signal a real-time and ongoing security risk. A method monitors system-related events and generates a vector representation for each event based on event features. Clusters of related events are determined, and a state automaton is employed to determine a strength of temporal “bursty” activity for each cluster. Hypothesis testing is performed on each cluster to determine a likelihood that the cluster is a temporally emergent cluster. Clusters with a bursting likelihood above a threshold are determined to be an emergent cluster associated with an anomalous issue. A notification regarding the detected anomaly is provided. A remedial action addressing the anomaly is performed.

Abstract: An audio stream is subjected to speech-to-text processing in order to obtain a textual representation of the audio stream. Hierarchical topic extraction is performed on the textual representation to obtain a multi-level hierarchical topic representation of the textual representation. A user interface actuator is generated, which allows a user to search through the audio stream. Different levels of the multi-level hierarchical topic representation are displayed to the user, based upon the speed of actuation of the user interface actuator.

Abstract: Technologies are provided for the monitoring, detection, and notification of emerging, related issues within a system, which may indicate a problem. Within a computing-security system, a sudden increase in the frequency of events associated with unauthorized logon attempts signal a real-time and ongoing security risk. A method monitors system-related events and generates a vector representation for each event based on event features. Clusters of related events are determined, and a state automaton is employed to determine a strength of temporal “bursty” activity for each cluster. Hypothesis testing is performed on each cluster to determine a likelihood that the cluster is a temporally emergent cluster. Clusters with a bursting likelihood above a threshold are determined to be an emergent cluster associated with an anomalous issue. A notification regarding the detected anomaly is provided. A remedial action addressing the anomaly is performed.

Abstract: The present invention enables a system to efficiently maintain an ancillary pane containing a projection of document content, and content derived therefrom, while the underlying document is being edited. In one embodiment, an acronym pane displays a list of acronyms used in a document and one or more expansions of those acronyms. A data store containing maps is used to index which paragraphs contain which acronyms, and which acronyms are associated with which expansions. As sections of content are added, modified, or deleted, acronyms and expansions are added to and/or removed from the data store, causing the acronym pane to be updated.

Abstract: Systems and methods for enhanced classification of sequences of objects based on clique similarity and metadata associated with the sequences are presented. Sequences are received. Events are detected based on analyzing k-skip-n-grams included in the sequences. For each event of the detected plurality of events, a graph is generated. The graph for a particular event includes z-cliques that correspond to portions of the k-skip-n-grams that are included in the sequences that are associated with the particular event. A first sequence, which is separate from the other sequences, is received. The first sequence includes a first plurality of k-skip-n-grams. A trained classifier is employed to classify the first sequence as being associated with a first event of the detected events. Classifying the first sequence is based on a comparison between the first plurality of k-skip-n-grams and the z-cliques of the graph that is generated for the first event.