Zoom Communications, Inc.

Zoom Video Communications, Inc.

Vijay Parthasarathy

Bilung Lee

Min Xiao-Devins

Davide Giovanardi

Wei Ji

Xiaoli Song

Tao Huang

Tao Tang

Qiumin Dong

Xiao Song

Eunkyung Kim

Zhaoyi Liu

Chih-Kai Ting

Periyaven Naiken Gopalla

Dading Chong

Amarakota Madhu Vamsi

David Dharmendran Rajkumar

Daniar Imanbayev

Mark Yacovone

Shaoyang Wu

Yaling Song

Bing Zhao

Xiaojing Zhao

Nicholas Troy Johnson

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.

Automatic task generation

Methods and systems provide for presenting the results of talking speed analysis per topic segment in a communication session. 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, the transcript including timestamps for each utterance of a speaking participant; determines, based on analysis of the transcript, a meeting type for the communication session; generates, based on the meeting type, a number of topic segments for the conversation and respective timestamps for the topic segments; for each topic segment, determines a word count per unit of time for each speaking participant associated with a prespecified organization; and presents, to one or more client devices, data on the topic segments, each data including the word count per unit of time for each speaking participant within the topic segment.

Talking speed analysis per topic segment in a communication session

A contact center server obtains historical contact center data of a contact center by tracking contact center conditions. The contact center server trains, based on the historical contact center data, multiple modeling engines to generate agent demand data representing a number of agents working at a given time. The contact center server trains, based on the historical center contact center data and performance data of the multiple modeling engines, a combination engine to generate a combination of one or more modeling engines from the multiple modeling engines. The contact center server provides an output representing the trained combination engine and the multiple modeling engines.

Training modeling engines to predict contact center agent demand

Methods and systems provide for virtual meeting coaching with content-based evaluation. In one embodiment, the system receives a set of coaching items including a number of questions each associated with an expected answer; connects to a coaching session including one or more participants and a virtual coaching agent; for each question and for at least a subset of the participants: transmitting the question, by the virtual coaching agent, to the client device used by the participant; receiving an answer to the question by the participant, the answer including media of the participant; receiving text of utterances spoken by the participant during the answer; generating one or more evaluation scores for the answer based on evaluating at least the content of the answer to the question; and transmitting an overall evaluation score for each of the subset of participants based on the generated evaluation scores for that participant.

Virtual meeting coaching with content-based evaluation

Example methods and systems facilitate user actions during a chat session on an online chat platform. A client device is installed with a chat and video conference application. The chat and video conference application includes a machine learning (ML) model. The client device receives a chat message during a chat session and identifies an action item from the chat message using the ML model. One or more GUI elements can be generated associated with a functionality of an application corresponding to the action item. A GUI element can be activated to invoke the functionality of the application.

Facilitating user actions for virtual interactions

The accuracy of automatic speech recognition (ASR) tasks is improved using trained models. A speech recognition model is applied in a noisy environment where speech is spoken at a distance from the microphones. The techniques may include extracting speech features, data augmentation by adding feature perturbation, and/or a multi-domain end-to-end speech recognition model. In some implementations, the described technology includes using a teacher-group knowledge distillation strategy to train a deep end-to-end speech recognition model on original speech samples and the sample speech augmentation of the original speech samples, that outputs recognized text transcriptions corresponding to speech detected in the original speech samples and the sample speech augmentation.

Noisy far-field speech recognition

Methods and systems provide for presenting time distributions of participants across topic segments in 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, the transcript including timestamps for each utterance of a speaking participant; determines, based on analysis of the transcript, a meeting type for the communication session; generates a number of topic segments for the conversation and respective timestamps for the topic segments; for each participant, analyzes the time spent by the participant on each of the generated topic segments in the meeting; and presents, to one or more users, data on the time distribution of participants for each topic segment and across topic segments within the conversation.

Time distributions of participants across topic segments in a communication session

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.

Extracting filler words and phrases from a communication session

Growing community of inventors

San Jose, CA, United States of America

Vijay Parthasarathy