Input detection windowing

A method to be performed by a playback device comprising at least one microphone configured to detect sound, the method comprising:

• detecting a first input sound data stream comprising a wake word and a voice input;
• performing an action based on the voice input;
• after performing the action, detecting an event, the event being associated with the playback device or a media playback system comprising the playback device, wherein the event is not based on sound data captured via the at least one microphone;
• responsive to the event, opening an input detection window for a given time period;
• while the input detection window is open for the given time period:
• receiving a second input sound data stream from the at least one microphone before the input detection window ends, wherein the second input sound data stream represents sound detected by the at least one microphone;
• analyzing the second input sound data stream by the at least one processor for a plurality of keywords supported by the playback device;
• determining, based on the analysis, that the second input sound data stream includes voice input data comprising a keyword;
• wherein the keyword is one of the plurality of keywords supported by the playback device;
• and
• responsive to the determining that the second input sound data stream includes voice input data comprising a keyword, causing a command corresponding to the keyword to be performed.

A non-transitory computer-readable medium having instructions stored thereon that are executable by one or more processors to cause a playback device to perform functions, the playback device comprising at least one microphone configured to detect sound, the functions comprising:

• detecting a first input sound data stream comprising a wake word and a voice input;
• performing an action based on the voice input;
• after performing the action, detecting an event, the event being associated with the playback device or a media playback system comprising the playback device, wherein the event is not based on sound data captured via the at least one microphone;
• responsive to the event, opening an input detection window for a given time period;
• while the input detection window is open for the given time period:
• receiving a second input sound data stream from the at least one microphone, wherein the second input sound data stream represents sound detected by the at least one microphone;
• analyzing the second input sound data stream by the at least one processor for a plurality of keywords supported by the playback device;
• determining, based on the analysis, that the second input sound data stream includes voice input data comprising a keyword;
• wherein the keyword is one of the plurality of keywords supported by the playback device;
• and
• responsive to the determining that the second input sound data stream includes voice input data comprising a keyword, causing a command corresponding to the keyword to be performed.