Generating input alternatives

1. 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 input data corresponding to a user input;

based at least in part on a similarity of sound between the user input and a first representation, generate first output data proposing the first representation to be used to respond to the user input;

receive a response confirming the first representation is to be used;

determine first application data associated with the first representation; and

generate second output data based on the first application data.

2. The system of claim 1 , wherein the at least one memory further comprises instructions that, when executed by the at least one processor, further cause the system to:

performing speech processing on the input data to determine results data,

wherein generation of the first output data is based at least in part on the results data.

3. The system of claim 2 , wherein the speech processing comprises natural language processing and the results data comprises a confidence score.

4. The system of claim 1 , wherein the at least one memory further comprises instructions that, when executed by the at least one processor, further cause the system to:

perform automatic speech recognition (ASR) processing using the input data to determine ASR confidence data; and

determine the ASR confidence data fails to satisfy a condition,

wherein generation of the first output data is based at least in part on the ASR confidence data failing to satisfy the condition.

5. The system of claim 4 , wherein the ASR confidence data is based at least in part on the similarity of sound.

6. The system of claim 1 , wherein the first representation corresponds to an alternative for the input data.

7. The system of claim 1 , wherein the at least one memory further comprises instructions that, when executed by the at least one processor, further cause the system to:

query a storage using at least a portion of the input data to determine a plurality of candidate input representations, the storage storing data representing a mapping between sample input data and system actionable data; and

identify the first representation based on the plurality of candidate input representations.

8. The system of claim 7 , wherein the at least one memory further comprises instructions that, when executed by the at least one processor, further cause the system to:

determine the plurality of candidate input representations based at least in part on a keyword represented in the user input.

9. The system of claim 7 , wherein the plurality of candidate input representations correspond to past inputs made by a plurality of users.

10. The system of claim 1 , wherein the at least one memory further comprises instructions that, when executed by the at least one processor, further cause the system to:

perform speech processing using the input data to determine speech processing results data; and

process the input data and the speech processing results data using a machine learning component to determine model output data representing a first likelihood that the input data corresponds to a false wake of a first device,

wherein generation of the first output data is based at least in part on the model output data.

11. A computer-implemented method, comprising:

receiving input data representing an utterance spoken to a first device;

performing speech processing using the input data to determine speech processing results data;

processing the input data and the speech processing results data using a machine learning component to determine model output data representing a first likelihood that the input data corresponds to a false wake of the first device;

generating first output data requesting confirmation of an intended wake of the first device;

receiving a response to the first output data; and

performing an action based at least in part on the response.

12. The computer-implemented method of claim 11 , further comprising:

receiving, from an automatic speech recognition component, the speech processing results data, wherein the speech processing results data comprises score data.

13. The computer-implemented method of claim 12 , wherein the score data represents a likelihood a user is conversing with the first device.

14. The computer-implemented method of claim 11 , further comprising:

receiving, from a natural language understanding component, the speech processing results data.

15. The computer-implemented method of claim 11 , further comprising:

prior to generating the first output data, determining the first likelihood satisfies a condition.

16. The computer-implemented method of claim 11 , further comprising:

receiving dialog data corresponding to the first device,

wherein the machine learning component further processes the dialog data to determine the model output data.

17. The computer-implemented method of claim 11 , wherein the first output data requesting confirmation comprises a request for a repeat of the utterance.

18. The computer-implemented method of claim 11 , further comprising:

receiving latency data corresponding to the utterance,

wherein the machine learning component further processes the latency data to determine the model output data.

19. The computer-implemented method of claim 11 , further comprising:

determining the response corresponds to an indication to perform processing;

determining second output data responsive to the utterance; and

causing presentation of the second output data.

20. The computer-implemented method of claim 11 , further comprising:

determining the response corresponds to an indication to stop processing; and

termination processing with regard to the input data.