The server 164 determines based on an analysis of the motion stream that a visitor is approaching the entryway, and performs a facial recognition operation based on one or more frames of the motion stream. The server 164 then determines based on an outcome of the facial recognition operation whether the person is known to the smart home environment. A time window is initiated in response to the determination that a visitor is approaching, and the time window is optionally predefined to correspond to a reasonable amount of time for the visitor to complete approaching/reaching the entryway. During the time window, the server 164 obtains contextual information from one or more sensors of the smart home environment 100, and determines whether an action from the visitor e. g. , a doorbell press, a door knock is detected within the time window. In accordance with an action being detected within the time window, a first type of response is initiated. Examples of the first type of response include sending a first type of notification to a user of a client device. In accordance with no action being detected within the time window, a second type of response is initiated. Examples of the second type of response include sending a second type of notification to a user of a client device. In some implementations, when the client device 220 loads the user mobile application 624 associated with the doorbell camera 106, the application 624 initiates a live video and/or audio connection to the doorbell camera, and allows video and audio information to stream within a predetermined duration of time e.