Hotword recognition

A system comprising:

• one or more computers;
• and
• one or more computers and one or more storage devices storing instructions that are operable, when executed by the one or more computers, to cause the one or more computers to perform operations comprising:
• receiving, by a first computing device, first audio data of a first utterance;
• determining, by the first computing device, that the first utterance includes a particular, predefined hotword;
• based on determining that the first utterance includes the particular, predefined hotword, determining, by the first computing device, whether the first audio data includes an audio signal previously outputted by a second computing device while the second computing device received second audio data of a second utterance that included the particular, predefined hotword;
• and
• based on determining whether the first audio data includes the audio signal, determining, by the first computing device, whether to perform a command that is included in the first utterance and that follows the particular, predefined hotword.

A non-transitory computer-readable medium storing software comprising instructions executable by one or more computers which, upon such execution, cause the one or more computers to perform operations comprising:

• receiving, by a first computing device, first audio data of a first utterance;
• determining, by the first computing device, that the first utterance includes a particular, predefined hotword;
• based on determining that the first utterance includes the particular, predefined hotword, determining, by the first computing device, whether the first audio data includes an audio signal previously outputted by a second computing device while the second computing device received second audio data of a second utterance that included the particular, predefined hotword;
• and
• based on determining whether the first audio data includes the audio signal, determining, by the first computing device, whether to perform a command that is included in the first utterance and that follows the particular, predefined hotword.