Abstract: Topic segments for a communication session are dynamically generated. In one embodiment, the system segments utterances of a transcript into one or more topic segments based on the topic; for each of the segments, determines whether the topic segment is related to the topic, and transmits, to one or more client devices, a list of the topic segments for the communication session.

Abstract: Automated content recommendation is used in conferencing. In one embodiment, a system receives a list of content recommendation actions. The system receives a number of utterances associated with the participants in real time. For each utterance, the system determines whether a prediction of relatedness is present between the utterance and one or more trigger phrases associated with a content recommendation action. Upon determining that a prediction of relatedness is present, the system performs the associated content recommendation action by transmitting, to one or more client devices, one or more pieces of content to be recommended.

Abstract: Example methods and systems for automatic task generation is provided. A communication platform receives an instruction to generate one or more task items associated with a virtual communication session related to a user. The communication platform accesses virtual communication data associated with the virtual communication session. The communication platform generates a task item based on the virtual communication data using a generative artificial intelligence (AI) model. The communication platform provides the task item to the user.

Abstract: Methods and systems provide for intelligent content recommendation within a communication session. In one embodiment, the system receives a list of content recommendation actions, each content recommendation action being associated with one or more trigger phrases constituting conditions for the content recommendation action to be performed, each trigger phrase being associated with a party the trigger phrase is to be uttered by. The system connects to a communication session with a plurality of participants, and receives a number of utterances associated with the participants in real time. For each utterance, the system determines whether a prediction of relatedness is present between the utterance and one or more trigger phrases associated with a content recommendation action. Upon determining that a prediction of relatedness is present, the system performs the associated content recommendation action by transmitting, to one or more client devices, one or more pieces of content to be recommended.

Abstract: Methods and systems provide for extracting filler words and phrases from a communication session. In one embodiment, the system receives a transcript of a conversation involving one or more participants produced during a communication session; extracts, from the transcript, utterances including one or more sentences spoken by the participants; identifies a subset of the utterances spoken by a subset of the participants associated with a prespecified organization; extracts filler phrases within the subset of utterances, the filler phrases each comprising one or more words representing disfluencies within a sentence, where extracting the filler phrases includes applying filler detection rules; and presents, for display at one or more client devices, data corresponding to the extracted filler phrases.

Abstract: Example methods and systems recommend and summarize unread messages for a user on an online chat platform. A client device accesses multiple unread messages associated with the user, and a machine learning (ML) model provided by the online chat platform. The client device selects one or more unread messages from the multiple unread messages using the ML model trained with historical chat data and user metadata. The client device also generates a summary of some or all of the unread messages using a second ML model trained with the historical chat data and the user metadata. The indication of the selected unread messages and the summary of unread messages are displayed on the client device.

Abstract: Methods and systems provide for extracting filler words and phrases from a communication session. In one embodiment, the system receives a transcript of a conversation involving one or more participants produced during a communication session; extracts, from the transcript, utterances including one or more sentences spoken by the participants; identifies a subset of the utterances spoken by a subset of the participants associated with a prespecified organization; extracts filler phrases within the subset of utterances, the filler phrases each comprising one or more words representing disfluencies within a sentence, where extracting the filler phrases includes applying filler detection rules; and presents, for display at one or more client devices, data corresponding to the extracted filler phrases.

Abstract: A machine learning model (e.g., including a deep learning neural network) with learned embeddings is applied to time series data with associated metadata to obtain predictions of the time series value. For example, a call volume in a period of time may be predicted based on call volume data for a sequence of time bins in a window of preceding time. Time bins may be associated with respective metadata, such as day of week, hour of day, day of month, holiday, part of business cycle, weather, and/or tide. These pieces of metadata may be mapped to embedding vectors using trained embedding functions. The resulting embedding vectors may be input to a neural network along with the corresponding time series data (e.g., call volumes) to make a prediction for future time bin. For example, the prediction may be used to provision servers in a network infrastructure.

Abstract: A sequence of call volume measurements is accessed, where each of the call volume measurements is associated with respective metadata. The respective metadata may provide information regarding a time period during which a call volume measurement was made. A window of the sequence of call volume measurements with the respective metadata is input to a machine learning model to obtain a prediction of a call volume. The machine learning model includes embedding functions that are applied to the respective metadata for the call volume measurements in the window.

Abstract: Methods and systems provide for intelligent content recommendation within a communication session. In one embodiment, the system receives a list of content recommendation actions, each content recommendation action being associated with one or more trigger phrases constituting conditions for the content recommendation action to be performed, each trigger phrase being associated with a party the trigger phrase is to be uttered by. The system connects to a communication session with a plurality of participants, and receives a number of utterances associated with the participants in real time. For each utterance, the system determines whether a prediction of relatedness is present between the utterance and one or more trigger phrases associated with a content recommendation action. Upon determining that a prediction of relatedness is present, the system performs the associated content recommendation action by transmitting, to one or more client devices, one or more pieces of content to be recommended.

Abstract: Methods and systems provide for extracting next step sentences from a communication session. In one embodiment, the system defines a set of annotation guidelines for labeling training data; receives a set of labeled training data including sentences from a transcript of a communication session, a subset of the sentences being associated with a positive label; organizes the labeled training data and trains a model with the labeled training data, the training including, for each of the sentences, inputting the sentence into a language model and a classification head to output a number of class probabilities, and inputting a classification token representing the sentence into a classification head; using a number of classifiers from the trained model to generate ensemble class scores; and using the ensemble class scores to predict one or more next step sentences from the sentences in the transcript.

Abstract: Methods and systems provide for dynamically generated topic segments for a communication session. In one embodiment, the system connects to a communication session with a number of participants; receives a list of topics; receives a transcript of a conversation between the participants produced during the communication session, the transcript including timestamps for a number of utterances associated with speaking participants; for each topic in the list of topics, segments the utterances into one or more topic segments based on the topic; for each of the segments, classifies whether the topic segment is related to the topic, and transmits, to one or more client devices, a list of the topic segments for the communication session.

Abstract: Methods and systems provide for dynamic conversation alerts within a communication session. In one embodiment, the system presents, to a client device associated with a user of a communication platform, a user interface (“UP”) including a prompt for the user to submit one or more alert phrases, each alert phrase being associated with a category; receives, from the client device, a list of submitted alert phrases; and receives a transcript of a communication session between participants. For each utterance in the transcript, the system determines whether one or more predictions of relatedness are present between the utterance and one or more alert phrases from the list of submitted alert phrases. The system then transmits, to the client device, a list of related categories, each related category including one or more timestamps of utterances for which a prediction of relatedness is present for an alert phrase associated with that category.

Abstract: Methods and systems provide for dynamically generated topic segments for a communication session. In one embodiment, the system connects to a communication session with a number of participants; receives a list of topics; receives a transcript of a conversation between the participants produced during the communication session, the transcript including timestamps for a number of utterances associated with speaking participants; for each topic in the list of topics, segments the utterances into one or more topic segments based on the topic; for each of the segments, classifies whether the topic segment is related to the topic, and transmits, to one or more client devices, a list of the topic segments for the communication session.

Abstract: Methods and systems provide for dynamically generated topic segments for a communication session. In one embodiment, the system connects to a communication session with a number of participants; receives a list of topics; receives a transcript of a conversation between the participants produced during the communication session, the transcript including timestamps for a number of utterances associated with speaking participants; for each topic in the list of topics, segments the utterances into one or more topic segments based on the topic; for each of the segments, classifies whether the topic segment is related to the topic, and transmits, to one or more client devices, a list of the topic segments for the communication session.

Abstract: Methods and systems provide for extracting engaging questions from a communication session. In one embodiment, the system connects to a communication session with a number of participants; receives a transcript of a conversation between the participants produced during the communication session; extracts, from the transcript, utterances including one or more sentences spoken by the participants; identifies a subset of the utterances spoken by a subset of the participants associated with a prespecified organization; extracts engaging questions within the subset of utterances, the engaging questions each including a question asked by the participant associated with the organization that is immediately answered in the following utterance by a participant not associated with the organization; and presents, for display at one or more client devices, data corresponding to the extracted engaging questions.

Abstract: Methods and systems provide for extracting next step sentences from a communication session. In one embodiment, the system connects to a communication session involving one or more participants; receives or generates a transcript of a conversation; extracts, from the transcript, a number of utterances including one or more sentences spoken by the participants; identifies a subset of the number of utterances spoken by a subset of the participants associated with a prespecified organization; extracts one or more next step sentences within the subset of the utterances, where the next step sentences each include an owner-action pair structure in which the action is an actionable verb in future tense or present tense; determines a set of analytics data corresponding to the next step sentences and the associated participants; and presents, to one or more users, at least a subset of the analytics data corresponding to the next step sentences.

Abstract: Methods and systems provide for extracting filler words and phrases from a communication session. In one embodiment, the system receives a transcript of a conversation involving one or more participants produced during a communication session; extracts, from the transcript, utterances including one or more sentences spoken by the participants; identifies a subset of the utterances spoken by a subset of the participants associated with a prespecified organization; extracts filler phrases within the subset of utterances, the filler phrases each comprising one or more words representing disfluencies within a sentence, where extracting the filler phrases includes applying filler detection rules; and presents, for display at one or more client devices, data corresponding to the extracted filler phrases.

Abstract: A sequence of call volume measurements is accessed, where each of the call volume measurements is associated with respective metadata. The respective metadata may provide information regarding a time period during which a call volume measurement was made. A window of the sequence of call volume measurements with the respective metadata is input to a machine learning model to obtain a prediction of a call volume. The machine learning model includes embedding functions that are applied to the respective metadata for the call volume measurements in the window.