Routing queries based on carrier phrase registration
In general, the subject matter described in this specification can be embodied in methods, systems, and program products for receiving s voice query at a mobile computing device and generating data that represents content of the voice query. The data is provided to a server system. A textual query that has been determined by a speech recognizer at the server system to be a textual form of at least part of the data is received at the mobile computing device. The textual query is determined to include a carrier phrase of one or more words that is reserved by a first third-party application program installed on the computing device. The first third-party application is selected, from a group of one or more third-party applications, to receive all or a part of the textual query. All or a part of the textual query is provided to the selected first application program.
1. A method comprising:
storing a predefined phrase on memory hardware of a computing device, the predefined phrase designating a particular application to perform an operation specified by a voice command;
receiving, at the computing device, a voice input comprising a beginning portion that includes a particular phrase and a remaining portion that includes a particular voice command; and
based on a determination that the particular phrase matches the predefined phrase:
selecting, by the computing device, a language model for use by a speech recognizer to convert the remaining portion of the voice input into a textual representation of the voice command;
converting, by the computing device, using the speech recognizer and the selected language model, the remaining portion of the voice input into the textual representation of the voice command without using the selected language model to convert the beginning portion of the voice input; and
instructing, by the computing device, based on the textual representation of the voice command, the particular application to perform a particular operation specified by the particular voice command of the voice input.
2. The method of claim 1 , wherein:
the computing device is associated with a user; and
receiving the voice input comprises:
recording, using a microphone, an audio input signal characterizing the voice input spoken by the user; and
encoding, using a voice query encoder residing on the computing device, the audio input signal into a digital representation of the voice input.
3. The method of claim 2 , wherein the speech recognizer resides on the computing device and is configured to convert the digital representation of the voice input into the textual representation of the voice command.
4. The method of claim 1 , further comprising, after receiving the voice input:
determining, by the computing device, that the particular phrase of the voice input matches the predefined phrase stored on the memory hardware of the computing device; and
in response to determining that the particular phrase of the voice input matches the predefined phrase, identifying, by the data processing, the particular application designated by the predefined phrase to perform the operation specified by the voice command.
5. The method of claim 4 , wherein identifying the particular application comprises:
accessing a particular user account associated with a user of the computing device; and
determining that the particular application is assigned to the particular user account and that the predefined phrase is reserved by the particular application for invoking the particular application when the predefined phrase is spoken by the user.
6. The method of claim 1 , wherein the particular application is installed on the computing device.
7. The method of claim 1 , wherein the particular application executes on a remote server system in communication with the computing device.
8. The method of claim 7 , wherein instructing the particular application to perform the particular operation specified by the particular voice command of the voice input comprises transmitting a resource request to the remote server system, the resource request comprising a query parameter associated with the textual representation of the voice command, the resource request, when received by the remote server system, causing the remote server system to:
extract the query parameter from the resource request; and
cause the particular application to perform the particular operation based on a content of the query parameter.
9. The method of claim 1 , wherein the predefined phrase comprises one or more words.
10. The method of claim 1 , wherein:
the particular application comprises a music player application; and
the selected language model is trained one musical artists and songs.
11. A user computing device comprising:
data processing hardware; and
memory hardware in communication with the data processing hardware and storing instructions that when executed by the data processing hardware cause the data processing hardware to perform operations comprising:
storing a predefined phrase on the memory hardware, the predefined phrase designating a particular application to perform an operation specified by a voice command;
receiving a voice input comprising a beginning portion that includes a particular phrase and a remaining portion that includes a particular voice command; and
based on a determination that the particular phrase matches the predefined phrase:
selecting a language model for use by a speech recognizer to convert the remaining portion of the voice input into a textual representation of the voice command;
converting, using the speech recognizer and the selected language model, the remaining portion of the voice input into the textual representation of the voice command without using the selected language model to convert the beginning portion of the voice input; and
instructing, based on the textual representation of the voice command, the particular application to perform a particular operation specified by the particular voice command of the voice input.
12. The user computing device of claim 11 , further comprising:
a microphone; and
a voice query encoder,
wherein receiving the voice input comprises:
recording, using the microphone, an audio input signal characterizing the voice input spoken by a user of the user computing device; and
encoding, using the voice query encoder, the audio input signal into a digital representation of the voice input.
13. The user computing device of claim 12 , wherein the speech recognizer resides on the user computing device and is configured to convert the digital representation of the voice input into the textual representation of the voice command.
14. The user computing device of claim 11 , wherein the operations further comprise, after receiving the voice input:
determining that the particular phrase of the voice input matches the predefined phrase stored on the memory hardware; and
in response to determining that the particular phrase of the voice input matches the predefined phrase, identifying the particular application designated by the predefined phrase to perform the operation specified by the voice command.
15. The user computing device of claim 14 , wherein identifying the particular application comprises:
accessing a particular user account associated with a user of the user computing device; and
determining that the particular application is assigned to the particular user account and that the predefined phrase is reserved by the particular application for invoking the particular application when the predefined phrase is spoken by the user.
16. The user computing device of claim 11 , wherein the particular application is installed on the user computing device.
17. The user computing device of claim 11 , wherein the particular application executes on a remote server system in communication with the user computing device.
18. The user computing device of claim 17 , wherein instructing the particular application to perform the particular operation specified by the particular voice command comprises transmitting a resource request from the user computing device to the remote server system, the resource request comprising a query parameter associated with the textual representation of the voice command, the resource request, when received by the remote server system, causing the remote server system to:
extract the query parameter from the resource request; and
cause the particular application executing on the remote server system to perform the particular operation based on a content of the query parameter.
19. The user computing device of claim 11 , wherein the predefined phrase comprises one or more words.
20. The user computing device of claim 11 , wherein:
the particular application comprises a music player application; and
the selected language model is trained one musical artists and songs.