Voice detection optimization using sound metadata

A method to be performed by a network microphone device (NMD), the method comprising:

• at a first time, capturing first sound data via one or more microphones of the NMD, wherein the first sound data includes a voice input;
• analyzing the first sound data to detect a wake word;
• obtaining an intent based on the voice input;
• performing a command based on the intent;
• at a second time, capturing second sound data via the one or more microphones;
• capturing metadata associated with the second sound data, wherein the second sound data is not derivable from the metadata;
• and
• based on the metadata, modifying at least one performance parameter of the NMD.

A tangible, non-transitory, computer-readable medium having stored therein instructions executable by one or more processors to cause a network microphone device (NMD) to perform operations comprising:

• at a first time, capturing first sound data via one or more microphones of the NMD, wherein the first sound data includes a voice input;
• analyzing the first sound data to detect a wake word;
• obtaining an intent based on the voice input;
• performing a command based on the intent;
• at a second time, capturing second sound data via the one or more microphones;
• capturing metadata associated with the second sound data, wherein the second sound data is not derivable from the metadata;
• and
• based on the metadata, modifying at least one performance parameter of the NMD.