Amazon Technologies, Inc.

Katherine M Ryan

Jules Cook Graybill

Marshall Friend Tappen

Maren Marie Costa

Kristina Lisa Klinkner

Chao Wang

Viktor Rozgic

Siddhartha Reddy Jonnalagadda

Erik Bergman

Christer Garbis

Alexandros Potamianos

Jeffrey Pratt

Brian Kropf

Edward Michael Leonard

Robert Wadia Moses

Avani Parakh

Alper Aras

James Rosenblum

Jera Vincelli

Gregory J Bartlett

The present application provides techniques for implementing a skill component, configured to perform an assessment of a user, as part of a speech processing system. The system may receive a natural language user input requesting assistance. The skill component may, using one or more machine learning models, determine at least one characteristic of the natural language input (e.g., lexical embedding, acoustic embedding, topic, tone, etc.). The skill component may determine state data for a present session, where the state data indicates a topic of the natural language user input and/or a user state associated with the natural language user input. The skill component may determine past state data of one or more past sessions, and generate a question to the user based on the state data for the natural language user input and the past state data.

Conversation-based skill component for assessing a user's state

Systems and methods to improve video conferencing may include various improvements and modifications to image and audio data to simulate mutual presence among users. For example, automated positioning and presentation of image data of users via presentation devices may present users at a position, size, and scale to simulate mutual presence of users within a same physical space. In addition, rendering image data of users from virtual camera positions via presentation devices may simulate mutual eye gaze and mutual presence of users. Further, blending a background, or accentuating a user, within image data that is presented via presentation devices may match visual characteristics of the environment or the user to simulate mutual presence of users. Moreover, dynamic steering of individual audio transducers of an audio transducer array, and simulating audio point sources using a combination of speakers, may simulate direct communication and mutual presence of users.

Image and audio data processing to create mutual presence in a video conference

Systems and methods are described herein for providing predictive user interface elements. A computing system may train a machine-learning model to identify issues that are likely being experienced by users contacting a customer service system based at least in part on historical user account data of a plurality of user accounts. When a request for assistance is received, user account data corresponding to the request may be obtained and provided to the model to identify issues likely experienced by a user. A number of graphical user interface (GUI) elements (e.g., 'match cards'), each corresponding to one of the identified issues, may be generated, ranked, and presented in accordance with the ranking. Each GUI element may be selectable. Upon selection additional data likely to be pertinent to the selected issue may be presented alleviating a need to search for this data as would be the case in conventional systems.

Techniques for providing predictive interface elements

Growing community of inventors

Seattle, WA, United States of America

Katherine M Ryan