| CPC G10L 15/22 (2013.01) [G06F 3/167 (2013.01); H04L 63/0876 (2013.01); G10L 2015/223 (2013.01)] | 19 Claims |

|
1. A computer implemented method, comprising:
receiving, by a shared device, first audio data that includes a first spoken query from a first user that is co-present with the shared device, wherein the first spoken query is directed to an automated assistant that is executing, at least in part, on the shared device;
receiving, by the shared device, second audio data that includes a second spoken query from a second user that is co-present with the shared device, wherein the second spoken query is directed to the automated assistant;
determining, based on sensor data generated by one or more sensors of the shared device, a first identifier of the first user;
determining a superset of actions to perform in response to the first spoken query and the second spoken query, wherein the superset includes a first subset of actions that are associated with the first spoken query and a second subset of actions that are associated with the second spoken query;
causing performance of the superset of actions to initiate via the shared device;
determining, based on the sensor data from the one or more sensors, that the first user is no longer co-present with the shared device;
identifying, based on the first identifier of the first user, an account that is associated with the first user;
identifying a personal device that is associated with the account;
causing performance of at least a portion of the first subset of actions via the personal device; and
removing, from the superset of actions, one or more actions of the first subset of actions that have not been performed.
|