Adaptive interface in a voice-activated network
The systems and methods of the present disclosure generally relate to a data processing system that can identify and surface alternative requests when presented with ambiguous, unclear, or other requests to which a data processing system may not be able to respond. The data processing system can improve the efficiency of network transmissions to reduce network bandwidth usage and processor utilization by selecting alternative requests that are responsive to the intent of the original request.
1. A method implemented by one or more processors, the method comprising:
receiving, from a first client device, a first input audio signal;
determining, based on the first input audio signal, a first plurality of candidate requests, each of the first plurality of candidate requests corresponding to a different semantic meaning of the first input audio signal;
determining an interface type of the first client device;
generating a first action data structure for a first candidate request of the first plurality of candidate requests and a second action data structure for a second candidate request of the first plurality of candidate requests, the first action data structure comprising a response to the first candidate request of the first plurality of candidate requests and the second action data structure comprising a response to the second candidate request of the first plurality of candidate requests; and
transmitting, to the first client device, the first action data structure and the second action data structure to be rendered in series by the first client device, wherein the transmitting is performed to cause rendering in series based on the interface type of the first client device.
2. The method according to claim 1 , wherein the transmitting is operable to cause the first client device to:
render, on a display of the first client device, the response to the first candidate request included in the first action data structure; and
subsequent to the first client device receiving user input to clear the response to the first candidate request from the display of the first client device, render, on the display of the first client device, the response to the second candidate request, instead of the response to the first candidate request.
3. The method according to claim 1 , further comprising:
receiving, from a second client device, a second input audio signal;
determining, based on the second input audio signal, a second plurality of candidate requests, each of the second plurality of candidate requests corresponding to a different semantic meaning of the second input audio signal;
determining an interface type of the second client device;
generating a third action data structure for a first candidate request of the second plurality of candidate requests and a fourth action data structure for a second candidate request of the second plurality of candidate requests, the third action data structure comprising a response to the first candidate request of the second plurality of candidate requests and the fourth action data structure comprising a response to the second candidate request of the second plurality of candidate requests; and
transmitting, to the second client device, the third action data structure and the fourth action data structure to be rendered in parallel by the second client device, wherein the transmitting is performed to cause rendering in parallel based on the interface type of the second client device.
4. The method according to claim 1 , further comprising:
parsing the first input audio signal to identify a primary request in the first input audio signal; and
determining a confidence score for a semantic meaning of the primary request,
wherein determining the first plurality of candidate requests is in response to the confidence score failing to satisfy a predetermined threshold.
5. The method according to claim 1 , wherein the interface type of the first client device indicates at least one of a display format of the first client device, a display size of the first client device, a display availability of the first client device, or a client device type of the first client device.
6. The method according to claim 1 , further comprising:
identifying a term in the first input audio signal having a plurality of interpretations; and
determining the first plurality of candidate requests based on the term in the first input audio signal having the plurality of interpretations.
7. The method according to claim 1 , further comprising determining the first plurality of candidate requests using a log of previously received input audio signals.
8. The method accordion to claim 1 , further comprising:
transmitting an audio signal request comprising a prompt; and
receiving, from the first client device, a second input audio signal from, the second input audio signal generated in response to the prompt,
wherein determining the first plurality of candidate requests is based the second input audio signal.
9. A computer program product comprising one or more computer-readable storage media having program instructions collectively stored on the one or more computer-readable storage media, the program instructions executable to:
receive, from a first client device, a first input audio signal;
determine, based on the first input audio signal, a first plurality of candidate requests, each of the first plurality of candidate requests corresponding to a different semantic meaning of the first input audio signal;
determine an interface type of the first client device;
generate a first action data structure for a first candidate request of the first plurality of candidate requests and a second action data structure for a second candidate request of the first plurality of candidate requests, the first action data structure comprising a response to the first candidate request of the first plurality of candidate requests and the second action data structure comprising a response to the second candidate request of the first plurality of candidate requests; and
transmit, to the first client device, the first action data structure and the second action data structure to be rendered in series by the first client device, wherein the transmitting is performed to cause rendering in series based on the interface type of the first client device.
10. The computer program product according to claim 9 , wherein the transmitting is operable to cause the first client device to:
render, on a display of the first client device, the response to the first candidate request included in the first action data structure; and
subsequent to the first client device receiving user input to clear the response to the first candidate request from the display of the first client device, render, on the display of the first client device, the response to the second candidate request, instead attic response to the first candidate request.
11. The computer program product according to claim 9 , wherein the program instructions are further executable to:
receive, from a second client device, a second input audio signal;
determine, based on the second input audio signal, a second plurality of candidate requests, each of the second plurality of candidate requests corresponding to a different semantic meaning of the second input audio signal;
determine an interface type of the second client device;
generate a third action data structure for a first candidate request of the second plurality of candidate requests and a fourth action data structure for a second candidate request of the second plurality of candidate requests, the third action data structure comprising a response to the first candidate request of the second plurality of candidate requests and the fourth action data structure comprising a response to the second candidate request of the second plurality of candidate requests; and
transmit, to the second client device, the third action data structure and the fourth action data structure to be rendered in parallel by the second client device, wherein the transmitting is performed to cause rendering in parallel based on the interface type of the second client device.
12. The computer program product according to claim 9 , wherein the program instructions are further executable to:
parse the first input audio signal to identify a primary request in the first input audio signal; and
determine a confidence score for a semantic meaning of the primary request,
wherein determining the first plurality of candidate requests is in response to the confidence score failing to satisfy a predetermined threshold.
13. The computer program product according to claim 9 , wherein the interface type of the first client device indicates at least one of a display format of the first client device, a display size of the first client device, a display availability of the first client device, or a client device type of the first client device.
14. The computer program product according to claim 9 , wherein the program instructions are further executable to:
identify a term in the first input audio signal having a plurality of interpretations; and
determine the first plurality of candidate requests based on the term in the first input audio signal having the plurality of interpretations.
15. The computer program product according to claim 9 , wherein the program instructions are further executable to determine the first plurality of candidate requests using a log of previously received input audio signals.
16. The computer program product according to claim 9 , wherein program instructions are further executable to:
transmit an audio signal request comprising a prompt; and
receive, from the first client device, a second input audio signal from, the second input audio signal generated in response to the prompt,
wherein determining the first plurality of candidate requests is based the second input audio signal.
17. A system comprising:
a processor, a computer-readable memory, one or more computer-readable storage media, and program instructions collectively stored on the one or more computer-readable storage media, the program instructions executable to:
receive, from a first client device, a first input audio signal;
determine, based on the first input audio signal, a first plurality of candidate requests, each of the first plurality of candidate requests corresponding to a different semantic meaning of the first input audio signal;
determine an interface type of the first client device;
generate a first action data structure for a first candidate request of the first plurality of candidate requests and a second action data structure for a second candidate request of the first plurality of candidate requests, the first action data structure comprising a response to the first candidate request of the first plurality of candidate requests and the second action data structure comprising a response to the second candidate request of the first plurality of candidate requests; and
transmit, to the first client device, the first action data structure and the second action data structure to be rendered in series by the first client device, wherein the transmitting is performed to cause rendering in series based on the interface type of the first client device.
18. The system according to claim 17 , wherein the transmitting is operable to cause the first client device to:
render, on a display of the first client device, the response to the first candidate request included in the first action data structure; and
subsequent to the first client device receiving user input to clear the response to the first candidate request from the display of the first client device, render, on the display of the first client device, the response to the second candidate request, instead of the response to the first candidate request.
19. The system according to claim 17 , wherein the program instructions are further executable to:
receive, from a second client device, a second input audio signal;
determine, based on the second input audio signal, a second plurality of candidate requests, each of the second plurality of candidate requests corresponding to a different semantic meaning of the second input audio signal;
determine an interface type of the second client device;
generate a third action data structure for a first candidate request of the second plurality of candidate requests and a fourth action data structure for a second candidate request of the second plurality of candidate requests, the third action data structure comprising a response to the first candidate request of the second plurality of candidate requests and the fourth action data structure comprising a response to the second candidate request of the second plurality of candidate requests; and
transmit, to the second client device, the third action data structure and the fourth action data structure to be rendered in parallel by the second client device, wherein the transmitting is performed to cause rendering in parallel based on the interface type of the second client device.
20. The system according to claim 17 , wherein the program instructions are further executable to:
parse the first input audio signal to identify a primary request in the first input audio signal; and
determine a confidence score for a semantic meaning of the primary request,
wherein determining the first plurality of candidate requests is in response to the confidence score failing to satisfy a predetermined threshold.