Abstract: Some embodiments relate to techniques performed at least in part by at least one voice agent executing on a computing device. The techniques comprise responsive to the at least one voice agent receiving input at least partially specifying a requested action to be performed at least partially by an application program, presenting visual feedback responsive to the input concurrently via a user interface of the at least one voice agent and a user interface of the application program.

Abstract: Some embodiments provide techniques performed by at least one voice agent. The techniques include receiving voice input; accessing contextual information related to an application program that has focus of the computing device when the voice input is received; and using the contextual information to interpret the received voice input.

Abstract: Some embodiments provide techniques performed by at least one voice agent. The techniques include receiving voice input from a user at least partially specifying a requested action to be performed at least in part by an application program, wherein the requested action requires a plurality of inputs to be fully specified; and in response to receiving the voice input, making the application program accessible to the user prior to completion of performance of the requested action, so as to enable the user to provide and/or edit at least one input of the plurality of inputs by directly interacting with the application program.