Message response routing

1. A computer-implemented method comprising:

receiving, from a first device associated with a first phone number and with a sender, first input audio data;

performing

speech processing on the first audio data to determine a request to send a first message to a recipient, the first message including a content portion indicating a location;

determining, using a contact list associated with a user profile of the sender, a second phone number corresponding to the location;

determining that the second phone number is different from the first phone number;

creating message data comprising a portion of the first input audio data and metadata including reply destination data corresponding to the second phone number;

sending the message data to a third device associated with the recipient;

receiving, from the third device, second input audio data;

processing the second input audio data to determine

a request to reply to the sender;

identifying, using the metadata, the second phone number; and

routing, using the metadata, a reply communication to the second device.

2. The computer-implemented method of claim 1 , further comprising:

identifying a contact list associated with a user profile of the recipient;

identifying, in the contact list, a contact entry corresponding to the sender, the contact entry including the first phone number; and

storing the second phone number with the contact entry.

3. The computer-implemented method of claim 1 , wherein the first text data does not expressly include the second phone number.

4. A computer-implemented method comprising:

determining first message data comprising first text data corresponding to content of a message and first metadata including a source contact identifier;

processing the first text data to determine that the first text data indicates a location;

determining, using the first text data, a second contact identifier corresponding to the location, wherein the second contact identifier is different from the source contact identifier;

creating second message data comprising data corresponding to the content and second metadata including reply destination data corresponding to the second contact identifier;

receiving a request to reply to the message; and

determining, based on the second metadata, to route a reply communication to a device associated with the second contact identifier.

5. The computer-implemented method of claim 4 , wherein the method further comprises:

causing the second message data to be sent to a second device associated with a recipient of the message;

receiving, from the second device, input audio data;

performing automatic speech recognition (ASR) processing on the input audio data to determine ASR result data comprising second text data;

performing NLU processing on the ASR result data to determine NLU result data comprising the second text data;

determining the NLU result data corresponds to the request to reply to the message; and

sending to the first device the reply communication.

6. The computer-implemented method of claim 4 , wherein the method further comprises:

receiving, from a second device associated with the source contact identifier, input audio data corresponding to the message;

performing automatic speech recognition (ASR) on the input audio data to determine the first text data;

performing text-to-speech (TTS) processing on the first text data to determine output audio data; and

including the output audio data in the second message data.

7. The computer-implemented method of claim 4 , further comprising:

identifying a contact list associated with a user profile of a recipient of the message;

identifying, in the contact list, a contact entry corresponding to a sender of the message, the contact entry including the source contact identifier; and

associating the second contact identifier with the contact entry.

8. The computer-implemented method of claim 7 , further comprising:

performing NLU processing on the first text data to determine a time period associated with the second contact identifier; and

storing, with the contact entry, a time indicator corresponding to the time period.

9. The computer-implemented method of claim 4 , wherein the first text data does not expressly include the second contact identifier.

10. The computer-implemented method of claim 4 , wherein the request to reply to the message does not include the second contact identifier.

11. The computer-implemented method of claim 4 , wherein:

the first metadata includes the source contact identifier as a reply to address; and

creating the second message data comprises replacing the source contact identifier with the second contact identifier as the reply: to address.

12. The computer-implemented method of claim 4 , wherein routing a reply communication to the device includes causing a call to be initiated with the device.

13. The computer-implemented method of claim 4 , wherein routing a reply communication to the device includes sending a reply message to the device.

14. A computing system comprising:

at least one processor;

a memory including instructions operable to be executed by the at least one processor to configure the system to:

determine first message data comprising first text data corresponding to content of a message and first metadata including a source contact identifier;

process the first text data to determine that the first text data indicates a location;

determine, using the first portion, a second contact identifier corresponding to the location, wherein the second contact identifier is different from the source contact identifier;

create second message data comprising data corresponding to the content and second metadata including reply destination data corresponding to the second contact identifier;

receive a request to reply to the message; and

determine, based on the second metadata, to route a reply communication to a device associated with the second contact identifier.

15. The computing system of claim 14 , wherein the memory further includes instructions that further configure the system to:

cause the second message data to be sent to a second device associated with a recipient of the message;

receive, from the second device, input audio data;

perform automatic speech recognition (ASR) processing on the input audio data to determine ASR result data comprising second text data;

perform NLU processing on the ASR result data to determine NLU result data comprising the second text data;

determine the NLU result data corresponds to the request to reply to the message; and

send to the first device the reply communication.

16. The computing system of claim 14 , wherein the memory further includes instructions that further configure the system to:

receive, from a second device associated with the source contact identifier, input audio data corresponding to the message;

perform automatic speech recognition (ASR) processing on the input audio data to determine the first text data; and

perform text-to-speech (TTS) processing on the first text data to determine output audio data; and

include the output audio data in the second message data.

17. The computing system of claim 14 , wherein the memory further includes instructions that further configure the system to:

identify a contact list associated with a user profile of a recipient of the message;

identify, in the contact list, a contact entry corresponding to a sender of the message, the contact entry including the source contact identifier; and

associate the second contact identifier with the contact entry.

18. The computing system of claim 17 , wherein the memory further includes instructions that further configure the system to:

perform NLU on the first text data to determine a time period associated with the second contact identifier; and

store, with the contact entry, a time indicator corresponding to the time period.

19. The computing system of claim 14 , wherein the first text data does not expressly include the second contact identifier.