CPC G06F 3/0484 (2013.01) [G06F 3/0481 (2013.01); G06F 3/167 (2013.01); H04M 1/72454 (2021.01); H04M 1/72457 (2021.01)] | 20 Claims |
1. A method performed by one or more data processing apparatus, the method comprising:
receiving request data specifying a request to launch a virtual assistant application from a lock screen of a mobile device;
in response to receiving the request data, obtaining a plurality of input signals, each of the input signals being of an input type different from each other input type of each other input signal, the input types for the input signals including an image type and an audio type;
evaluating each of the plurality of input signals to respectively determine a plurality of confidence scores for a plurality of input modes corresponding to the plurality of input types;
selecting, from the plurality of input modes, an input mode for the virtual assistant application based on the plurality of confidence scores generated for the plurality of input modes; and
presenting, by the virtual assistant application, content to a user of the mobile device, the presented content being selected based on the input signal received using the selected input mode.
|