| CPC G10L 17/22 (2013.01) [G10L 15/08 (2013.01); G10L 15/1822 (2013.01); G10L 15/30 (2013.01); G10L 21/0208 (2013.01); H04R 1/08 (2013.01); H04R 1/323 (2013.01); H04R 1/326 (2013.01); H04R 1/345 (2013.01); H04R 3/002 (2013.01); H04R 3/005 (2013.01); G10L 15/20 (2013.01); G10L 2021/02082 (2013.01); G10L 2021/02166 (2013.01); H04R 3/12 (2013.01); H04R 27/00 (2013.01); H04R 2227/003 (2013.01); H04R 2420/07 (2013.01)] | 20 Claims |

|
1. A device, comprising:
a housing;
a microphone;
a network interface configured to communicate over a network;
one or more processors; and
non-transitory computer-readable media storing instructions that, when executed by the one or more processors, cause the device to perform operations comprising:
receiving audio at the microphone;
generating first audio data representing a processed instance of an audio signal corresponding to the audio;
generating second audio data as a result of digital signal processing performed on the first audio data;
causing a first speech recognition component to determine if the second audio data represents a first wake word;
causing a second speech recognition component to determine if the second audio data represents a second wake word, the first speech recognition component differing from the second speech recognition component; and
sending the second audio data to a remote speech processing system based at least in part on the first speech recognition component determining that the second audio data represents the first wake word.
|