| CPC G10L 15/22 (2013.01) [G06F 16/3329 (2019.01); G10L 15/083 (2013.01); G10L 15/1815 (2013.01); G10L 15/1822 (2013.01); G10L 15/26 (2013.01); G10L 15/30 (2013.01); G10L 25/48 (2013.01); G10L 15/18 (2013.01); G10L 2015/223 (2013.01); G10L 2015/228 (2013.01)] | 14 Claims |

|
1. A method implemented by one or more processors and comprising:
activating a speech acquiring mode of the dialog system in response to a user speaking an activation phrase;
receiving, via the speech acquiring mode after activating the speech acquiring mode, a first spoken input of a user;
deactivating, after receiving the first spoken input of the user, the speech acquiring mode;
processing the first spoken input, using an automatic speech recognizer, to generate a recognized input;
processing the recognized input, using a natural language processing module, to determine a meaning representation for the recognized input;
obtaining, based on the meaning representation, a response associated with the first spoken input;
delivering the response in response to receiving the first spoken input;
selecting, from a plurality of candidate time periods, a particular time period for the speech acquiring mode to last when the speech acquiring mode is automatically reactivated responsive to the first spoken input;
determining, in response to a conversation mode being activated, to automatically re-activate the speech acquiring mode responsive to the first spoken input; and
automatically re-activating the speech acquiring mode responsive to the first spoken input and in response to determining to automatically re-activate the speech acquiring mode,
wherein automatically re-activating the speech acquiring mode comprises causing the speech acquiring mode to last for the selected particular time period.
|