Google Inc.

Sharon Stovezky

Marcin M Nowak-Przygodzki

Vladimir Vuskovic

Joseph Lange

Mugurel Ionut Andreica

Behshad Behzadi

Yariv Adan

Radu Voroneanu

Ragnar Groot Koerkamp

Marcin Nowak-Przygodzki

Implementations are set forth herein for creating an order of execution for actions that were requested by a user, via a spoken utterance to an automated assistant. The order of execution for the requested actions can be based on how each requested action can, or is predicted to, affect other requested actions. In some implementations, an order of execution for a series of actions can be determined based on an output of a machine learning model, such as a model that has been trained according to supervised learning. A particular order of execution can be selected to mitigate waste of processing, memory, and network resources—at least relative to other possible orders of execution. Using interaction data that characterizes past performances of automated assistants, certain orders of execution can be adapted over time, thereby allowing the automated assistant to learn from past interactions with one or more users.

Orchestrating execution of a series of actions requested to be performed via an automated assistant

Implementations are set forth herein for creating an order of execution for actions that were requested by a user, via a spoken utterance to an automated assistant. The order of execution for the requested actions can be based on how each requested action can, or is predicted to, affect other requested actions. In some implementations, an order of execution for a series of actions can be determined based on an output of a machine learning model, such as a model that has been trained according to supervised learning. A particular order of execution can be selected to mitigate waste of processing, memory, and network resources&#x2014;at least relative to other possible orders of execution. Using interaction data that characterizes past performances of automated assistants, certain orders of execution can be adapted over time, thereby allowing the automated assistant to learn from past interactions with one or more users.

Implementations set forth herein relate to an automated assistant that operates according to a variety of different location-based biasing modes for rendering responsive content for a user and/or proactively suggesting content for the user. The user can provide condensed spoken utterances to the automated assistant, when the automated assistant is operating according to one or more location-based biasing modes, but nonetheless receive accurate responsive outputs from the automated assistant. A responsive output generated by biasing toward a subset of location characteristic data that has been prioritized over other subsets of location characteristic data. The biasing allows the automated assistant to compensate for any details that may be missing from a spoken utterance, but allows the user to provide shorter spoken utterances, thereby reducing an amount of language processing when processing inputs from the user.

Location-based mode(s) for biasing provisioning of content when an automated assistant is responding to condensed natural language inputs

Growing community of inventors

San Francisco, CA, United States of America

Sharon Stovezky