Voice to voice natural language understanding processing

1. A computer-implemented method, the method comprising:

receiving, from a first device associated with a user profile, first audio data corresponding to a first utterance;

performing speech processing on the first audio data to determine first text data;

determining that the first text data represents an order intended for a business;

identifying business profile data corresponding to the business;

determining, using the business profile data, that first data and second data are needed to complete the order;

determining that a first portion of the first text data corresponds to the first data;

determining second text data corresponding to a first request for the second data;

performing text-to-speech (TTS) processing on the second text data to generate second audio data;

sending, to the first device, the second audio data;

receiving, from the first device, third audio data corresponding to a second utterance;

performing speech processing on the third audio data to determine third text data;

determining that a second portion of the third text data corresponds to the second data;

generating fourth text data corresponding to the order, the fourth text data including at least:

a second request for the order,

the first portion of the first text data, and

the second portion of the third text data;

identifying a second device associated with the business profile data;

performing TTS processing on the fourth text data to generate fourth audio data;

sending the fourth audio data to the second device;

receiving, from the second device, fifth audio data corresponding to a third utterance;

performing speech processing on the fifth audio data to determine fifth text data;

generating sixth text data using at least a portion of the fifth text data;

performing TTS processing on the sixth text data to generate sixth audio data; and

sending the sixth audio data to the first device.

2. The computer-implemented method of claim 1 , wherein determining that the first portion of the first text data corresponds to the first data further comprises:

determining that the first text data indicates a first item from a list;

determining customization options available for the first item;

selecting, using the user profile, a first option of the customization options; and

determining a first portion of the first data, wherein the first portion of the first data indicates the first item and the first option.

3. The computer-implemented method of claim 1 , further comprising, prior to receiving the first audio data:

receiving a third request to generate the business profile data;

sending an indication identifying a plurality of functions;

receiving first input data selecting a first function;

receiving second input data indicating first content associated with the first function;

receiving third input data selecting a second function;

receiving fourth input data indicating second content associated with the second function; and

generating the business profile data including the first function, the first content, the second function, and the second content.

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

initiate a communication session with the second device;

sending, as part of the communication session, the fourth audio data to the second device;

receiving, from the second device as part of the communication session, the fifth audio data;

determining that the fifth text data represents confirmation of the order and indicates an amount of time before the order is ready; and

generating the sixth text data indicating that the order was placed and the amount of time.

5. A computer-implemented method comprising:

receiving, from a first device, first audio data corresponding to a first utterance;

determining, based on the first audio data, that the first utterance corresponds to a transaction;

determining first profile data associated with the transaction;

determining, based on the first profile data, that first data is needed for the transaction;

determining that the first audio data corresponds to at least a portion of the first data;

identifying a second device associated with the first profile data;

generating second audio data including first synthesized speech indicating at least some of the first data;

sending the second audio data to the second device;

receiving, from the second device, third audio data corresponding to a second utterance;

performing speech processing on the third audio data to determine second data;

generating, using at least a portion of the second data, fourth audio data including second synthesized speech; and

sending the fourth audio data to the first device.

6. The computer-implemented method of claim 5 , further comprising:

performing speech processing on the first audio data to determine first text data;

determining an intent associated with the first text data;

determining, based on the intent, that the first text data corresponds to the transaction;

identifying a first entity included in the first text data;

determining, based on the first entity, the first profile data;

determining, based on the first profile data, the first data;

determining a second entity included in the first text data; and

determining that the second entity corresponds to a first portion of the first data.

7. The computer-implemented method of claim 5 , further comprising:

performing speech processing on the first audio data to determine first text data;

identifying a first entity included in the first text data;

identifying a second entity included in the first text data;

determining that the first entity corresponds to a first portion of the first data;

determining that the second entity corresponds to a second portion of the first data;

generating second text data indicating at least the first entity and the second entity; and

generating the second audio data based on the second text data.

8. The computer-implemented method of claim 5 , further comprising:

identifying user profile data associated with the first device;

determining that a portion of the user profile data is associated with the transaction; and

determining a second portion of the first data based on the portion of the user profile data.

9. The computer-implemented method of claim 5 , further comprising:

determining an identification number associated with the first data;

identifying a third device associated with the first profile data;

sending the first data to the third device via a data network; and

generating the second audio data, wherein the first synthesized speech indicates at least the identification number.

10. The computer-implemented method of claim 5 , further comprising:

initiating a communication session with the second device;

determining first text data corresponding to at least some of the first data;

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

sending, as part of the communication session, the second audio data to the second device.

11. The computer-implemented method of claim 5 , further comprising:

performing speech processing on the third audio data to determine the second data, the second data including first text data;

determining an intent associated with the first text data;

determining second text data based on the intent and the first text data; and

performing text-to-speech (TTS) processing on the second text data to generate the fourth audio data.

12. The computer-implemented method of claim 5 , further comprising:

receiving, from the second device, fifth audio data corresponding to a third utterance;

determining that the third utterance corresponds to a request for second data;

generating sixth audio data including third synthesized speech requesting the second data;

sending the sixth audio data to the first device;

receiving, from the first device, seventh audio data;

determining, based on the seventh audio data, the second data;

generating eighth audio data corresponding to the second data; and

sending the eighth audio data to the second device.

13. The computer-implemented method of claim 5 , further comprising:

determining second profile data associated with the transaction;

identifying a third device associated with the second profile data;

sending the second audio data to the third device;

receiving, from the second device, fifth audio data corresponding to a third utterance;

determining that the third utterance corresponds to third data;

receiving, from the third device, sixth audio data corresponding to a fourth utterance;

determining that the fourth utterance corresponds to fourth data;

generating seventh audio data based on the third data and the fourth data; and

sending the seventh audio data to the first device.

14. The computer-implemented method of claim 5 , further comprising, prior to receiving the first audio data:

receiving a request to generate the first profile data;

sending an indication identifying a plurality of functions;

receiving first input data selecting a first function;

receiving second input data indicating first content associated with the first function;

receiving third input data selecting a second function;

receiving fourth input data indicating second content associated with the second function; and

generating the first profile data including the first function, the first content, the second function, and the second content.

15. The computer-implemented method of claim 5 , further comprising, prior to generating the second audio data:

determining that the first audio data corresponds to a first portion of the first data;

generating fifth audio data including third synthesized speech requesting a second portion of the first data;

sending, to the first device, the fifth audio data;

receiving, from the first device, sixth audio data corresponding to a third utterance; and

determining, based on the sixth audio data, the second portion of the first data.

16. A system comprising:

at least one processor; and

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

receive, from a first device, first audio data corresponding to a first utterance;

determine, based on the first audio data, that the first utterance corresponds to a transaction;

determine first profile data associated with the transaction;

determine, based on the first profile data, that first data is needed for the transaction;

determine that the first audio data corresponds to at least a portion of the first data;

identify a second device associated with the first profile data;

generate second audio data including first synthesized speech indicating at least some of the first data;

send the second audio data to the second device;

receive, from the second device, third audio data corresponding to a second utterance;

perform speech processing on the third audio data to determine second data;

generate, using at least a portion of the second data, fourth audio data including second synthesized speech; and

send the fourth audio data to the first device.

17. The system of claim 16 , wherein the memory further comprises instructions that, when executed by the at least one processor, further cause the system to:

perform speech processing on the first audio data to determine first text data;

determine an intent associated with the first text data;

determine, based on the intent, that the first text data corresponds to the transaction;

identify a first entity included in the first text data;

determine, based on the first entity, the first profile data;

determine, based on the first profile data, the first data;

determine a second entity included in the first text data; and

determine that the second entity corresponds to a first portion of the first data.

18. The system of claim 16 , wherein the memory further comprises instructions that, when executed by the at least one processor, further cause the system to:

perform speech processing on the first audio data to determine first text data;

identify a first entity included in the first text data;

identify a second entity included in the first text data;

determine that the first entity corresponds to a first portion of the first data;

determine that the second entity corresponds to a second portion of the first data;

generate second text data indicating at least the first entity and the second entity; and

generate the second audio data based on the second text data.

19. The system of claim 16 , wherein the memory further comprises instructions that, when executed by the at least one processor, further cause the system to:

initiate a communication session with the second device;

determine first text data corresponding to at least some of the first data;

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

send, as part of the communication session, the second audio data to the second device.

20. The system of claim 16 , wherein the memory further comprises instructions that, when executed by the at least one processor, further cause the system to:

perform speech processing on the third audio data to determine the second data, the second data including first text data;

determine an intent associated with the first text data;

determine second text data based on the intent and the first text data; and

perform text-to-speech (TTS) processing on the second text data to generate the fourth audio data.

21. The system of claim 16 , wherein the memory further comprises instructions that, when executed by the at least one processor, further cause the system to:

receive, from the second device, fifth audio data corresponding to a third utterance;

determine that the third utterance corresponds to a request for second data;

generate sixth audio data including third synthesized speech requesting the second data;

send the sixth audio data to the first device;

receive, from the first device, seventh audio data;

determine, based on the seventh audio data, the second data;

generate eighth audio data corresponding to the second data; and

send the eighth audio data to the second device.