Wakeword detection
Techniques for implementing multiple wakeword detectors on a single device are described. A digital signal processor (DSP) of the device may implement a wakeword detection component to detect when captured speech includes a wakeword. A companion application installed on the device may implement a wakeword detection component trained using speech of a user of the device. If the DSP's wakeword detection component detects a wakeword in speech, the companion application's wakeword detection component may be used to determine whether the wakeword was spoken by the user of the device. If the companion application's wakeword detection component determines the user spoke the wakeword, audio data representing the speech may be sent to at least one server(s) for processing.
1 . A computer-implemented method, comprising:
receiving first data representing user presence information;
processing the first data to determine the first data indicates a presence of a first user;
based at least in part on the presence of the first user being indicated by the first data, implementing a first machine learning model;
receiving input data representing a user input;
processing the input data using the first machine learning model to determine output data; and
performing an action using the output data.
2 . The computer-implemented method of claim 1 , wherein receiving the input data comprises receiving audio data representing user speech.
3 . The computer-implemented method of claim 1 , wherein implementing the first machine learning model occurs prior to receiving the input data.
4 . The computer-implemented method of claim 1 , wherein the first machine learning model comprises a wakeword detection model.
5 . The computer-implemented method of claim 1 , further comprising:
selecting the first machine learning model from a plurality of machine learning models including at least the first machine learning model and a second machine learning model different from the first machine learning model.
6 . The computer-implemented method of claim 1 , wherein the first data corresponds to a device associated with the first user.
7 . The computer-implemented method of claim 6 , wherein device corresponds to a key.
8 . The computer-implemented method of claim 1 , wherein the first machine learning model is associated with a first speech processing component of a plurality of speech processing components.
9 . The computer-implemented method of claim 1 , wherein the first machine learning model is associated with a first companion application of a plurality of applications.
10 . The computer-implemented method of claim 1 , further comprising:
processing the output data to determine that the user input originated from the first user.
11 . A system comprising:
at least one processor; and
at least one memory comprising instructions that, when executed by the at least one processor, cause the system to:
receive first data representing user presence information;
process the first data to determine the first data indicates a presence of a first user;
based at least in part on the presence of the first user being indicated by the first data, implement a first machine learning model;
receive input data representing a user input;
process the input data using the first machine learning model to determine output data; and
perform an action using the output data.
12 . The system of claim 11 , wherein receiving the input data comprises receiving audio data representing user speech.
13 . The system of claim 11 , wherein implementing the first machine learning model occurs prior to receiving the input data.
14 . The system of claim 11 , the first machine learning model comprises a wakeword detection model.
15 . The system of claim 11 , wherein the at least one memory further comprises instructions that, when executed by the at least one processor, further cause the system to:
select the first machine learning model from a plurality of machine learning models including at least the first machine learning model and a second machine learning model different from the first machine learning model.
16 . The system of claim 11 , wherein the first data corresponds to a device associated with the first user.
17 . The system of claim 16 , wherein device corresponds to a key.
18 . The system of claim 11 , wherein the first machine learning model is associated with a first speech processing component of a plurality of speech processing components.
19 . The system of claim 11 , wherein the first machine learning model is associated with a first companion application of a plurality of applications.
20 . The system of claim 11 , wherein the at least one memory further comprises instructions that, when executed by the at least one processor, further cause the system to:
process the output data to determine that the user input originated from the first user.