Amazon Technologies, Inc.

Vinaya Nadig

Samarth Bhargava

Ambika Babuji

Elad Refael Kassis

Rohan Mutagi

Tu Dien Do

He Lu

Neelesh Deo Dani

Zhuxuan Li

Shipra Agarwal Kanoria

Brian Alexander Oliver

Anjishnu Kumar

Bo Li

Christopher Geiger Parker

Iain A Kennedy

Ninad Anil Parkhi

Shawn Michael Banks

King Lung Chiu

Michael Richard Baglole

Ilana Rozanes

Omar Zia Khan

Supriya Medapati

Prakash Kumar Jha

Sulman Riaz

Hariharan Srinivasan

Eric Wei Hoong Ow

Nicholas Kalkouni

Tina Sonal Patel

Rafael Colorado Vargas

Anuj Kamra

Sarang Anil Ozarde

Zheng Zheng

Bhaskara Kiran Kumar Kommalapati

Adrien Montpellier

Mohit Nayak

Yugandhar Maram

Techniques for determining relevant and personalized supplemental content for a user are described. A system may prompt a language model to generate queries based on an entity (or entities) representing a user's affinity. In some examples, the language model may be prompted to generate an initial query corresponding to the entity and then generate one or more enriched queries based on the initial query. The one or more enriched queries may be used to search one or more content providers for supplemental content results. The supplemental content results may be aggregated, may be validated to ensure correspondence to the enriched query, and may be stored for future output. A system component may request supplemental content for output, at which time the system may select one or more supplemental content from storage, where the selection may be made to optimize relevancy for current user and situational context.

Determining supplemental content for output

Techniques for coordinating output of inferred content using various components and systems are described. A supplemental content system and a notification system may each receive inferred content to be output. When the supplemental content system or the notification system outputs the inferred content, the respective system stores a record of the output of the content in a historical output storage. Thereafter, when the other system is ready to output the inferred content, the other system may prevent the inferred content from being output based on the inferred content having already been output, as represented in the historical output storage.

Voice user interface notification rendering

A system is provided for determining customized notification output based on user preferences and event information. A user may request a system to provide a notification when a future event occurs. The system may determine a custom notification/output template for the notification, where a portion of the template is determined based on a user preference. When the event occurs, the system may generate a notification using the template, where portions of the notification are determined using event information and user preferences.

Generating event output

A system is provided for determining subscription data when a user requests to receive an output in the future when an event occurs. The system may determine an output type based on the capabilities of the output device and a trigger type. The system may determine a trigger type based on the priority of the triggering event. The system may also determine how many times the subscription is to be executed. Using this information, the system creates the subscription so that the user may receive a notification or an announcement when an event occurs.

Techniques for rendering notification and announcement content for read time are described. A notification system provides non-natural language notification content. Sometime thereafter, a user input to output the notification content is received. In response, template-based or natural language generation processing is performed to convert the non-natural language notification content into natural language notification content including updated time information. The natural language notification content is then output to the user. Alternatively, a notification system provides non-natural language announcement content. Sometime thereafter, the template-based or natural language generation processing is performed to convert the non-natural language announcement content into natural language announcement content including updated time information. The natural language announcement content is then proactively output to an intended recipient user or group of users.

Dynamically rendered notifications and announcements

A system is provided for determining privacy controls for output including sensitive data. A user may subscribe to receive an output in the future based on the occurrence of an event. The system may determine when the event is occurred triggering the output, and determine that the output includes outputting sensitive data. The system may determine output data that does not include the sensitive data, send the output data to a device, and may request the user to provide an authentication input to receive the sensitive data.

Sensitive data control

Techniques for ordering the output of notification summaries are described. A system may receive multiple notifications intended for a same user or group of users. In response to receiving a user input requesting output of notifications (or in response to multiple notifications expiring soon), the system may identify multiple notifications intended for the user or group of users. The system generates natural language summaries of the notifications, and orders the natural language summaries based on one or more default ordering rules, one or more user preferences, one or more notification provider preference, and/or user feedback. The system then outputs the ordered natural language summaries to the user.

Voice user interface notification ordering

Described are techniques for minimizing the performance of sets of actions in undesired situations. For example, a set of actions may be triggered based on a sensor detecting the presence of an individual, but the sensor may be faulty and trigger the set of actions to be performed based on false detections of an individual. A system may maintain a record of past executions of sets of actions. The system may analyze the past executions to determine a frequency with which a set of actions was executed, whether the past executions resulted in failed processing, etc. If the system determines, based on these determinations, that a set of actions should not be performed in the present instance, the system prevents the set of actions from being performed. The system may notify a user of such and provide the user with the ability to override the system's determination.

Reducing unintended automation events

Techniques for managing notifications to be presented by user devices are described. In an example, a computer system associates a first user account with a user device and stores first data indicating that a notification about a communications session via a communications network system is prohibited. The computer system receives a request associated with the first user account for communications with another device via the communications network system. Based at least in part on the other device, the computer system generates second data indicating that the notification about the communications session is to be presented within a period of time of the communications session being requested. The computer system associates the first user account with the second data.

Setting overrides

Techniques for using sending communication data using a first communication system and a second communication system are described. When a request is received to send communication data, it is determined that sending the communication data includes the second communication system. A request for updated identification information, such as a security token, is therefore sent to the second communication system prior to, for example, determining further account information related to the initiator of the communication and/or an identity of a recipient of the communication. While the request is pending, additional requests for the identification information are delayed. When the new identification information is received, the communication data is sent.

Communications token pre-fetch

Using a method of operating a system that includes remote servers, and at least one electronic device, a user verbally instructs the electronic device to activate a function. The system uses the remote servers and other parts of the system to determine that the function is not currently enabled and requires a user-acknowledgment to terms of use before the function is enabled. The system provides information on the terms of use and solicits a user-acknowledgment to the terms of use. The user provides a verbal acknowledgment of the terms of use and the verbal acknowledgment is received and stored in a persistent data store.

Functional prerequisites and acknowledgments

Growing community of inventors

Redmond, WA, United States of America

Vinaya Nadig