Speech processing performed with respect to first and second user profiles in a dialog session
Techniques for implementing a “volatile” user ID are described. A system receives first input audio data and determines first speech processing results therefrom. The system also determines a first user that spoke an utterance represented in the first input audio data. The system establishes a multi-turn dialog session with a first content source and receives first output data from the first content source based on the first speech processing results and the first user. The system causes a device to present first output content associated with the first output data. The system then receives second input audio data and determines second speech processing results therefrom. The system also determines the second input audio data corresponds to the same multi-turn dialog session. The system determines a second user that spoke an utterance represented in the second input audio data and receives second output data from the first content source based on the second speech processing results and the second user. The system causes the device to present second output content associated with the second output data.
1. A computer-implemented method comprising:
during a first time period:
receiving, from a device, first input audio data corresponding to a first utterance;
determining first audio characteristics corresponding to the first utterance;
determining the first audio characteristics correspond to first stored audio characteristic data associated with first user profile data;
performing automatic speech recognition (ASR) on the first input audio data to generate first input text data;
performing natural language understanding (NLU) on the first input text data to generate first speech processing results data;
determining a first speechlet device associated with the speech processing results;
sending, to the first speechlet device, the first user profile data, the first speech processing results data, and a dialog session identifier (ID);
receiving first output data and the dialog session ID from the first speechlet device; and
sending, to the device, the dialog session ID and the first output data; and
during a second time period after the first time period:
receiving, from the device, the dialog session ID and second input audio data corresponding to a second utterance;
performing ASR on the second input audio data to generate second input text data;
performing NLU on the second input text data to generate second speech processing results data;
determining second audio characteristics corresponding to the second utterance;
determining the second audio characteristics correspond to second stored audio characteristic data associated with second user profile data;
sending, to the first speechlet device, the second user profile data, the second speech processing results data, and the dialog session ID;
receiving second output data and the dialog session ID from the first speechlet device; and
sending to the device, the dialog session ID and the second output data.
2. The computer-implemented method of claim 1 , further comprising:
determining the first audio characteristics correspond to at least one individual under at least eighteen years of age; and
performing NLU on the first input text data to with respect to a subset of functions corresponding to content selected for at least one individual user under at least eighteen years of age.
3. The computer-implemented method of claim 1 , further comprising:
determining the first user profile data indicates a first user age of less than eighteen years;
determining the second user profile data indicates a second user age of at least eighteen years; and
based on the first user age being less than eighteen years, sending an indication that a user is a child to the first speechlet device.
4. The computer-implemented method of claim 1 , further comprising:
determining the first audio characteristics correspond to first stored audio characteristic data associated with at least one individual under at least eighteen years of age;
determining the second audio characteristics corresponding to second stored audio characteristic data associated with at least one individual eighteen years of age or older; and
sending an indication that a user is a child to the first speechlet device.
5. A system comprising:
at least one processor; and
at least one memory including instructions that, when executed by the at least one processor, cause the system to:
during a first time period:
receive, from a device, first input audio data corresponding to a first utterance;
determine first user profile data associated with the first input audio data;
determine first speech processing results data based on the first input audio data;
associate the first speech processing results data with a dialog session identifier (ID);
generate first output data based on the first speech processing results data and the first user profile data; and
send, to the device, the first output data and the dialog session ID; and
during a second time period after the first time period:
receive, from the device, the dialog session ID and second input audio data corresponding to a second utterance;
determine second speech processing results data based on the second input audio data;
associate the second speech processing results data with the dialog session ID
determine second user profile data associated with the second input audio data;
generate second output data based on the second speech processing results data and the second user profile data; and
send, to the device, the second output data and the dialog session ID.
6. The system of claim 5 , wherein the instructions, when executed by the at least one processor, further cause the system to:
determine first audio characteristics corresponding to the first utterance; and
determine the first audio characteristics correspond to stored audio characteristic data associated with at least one individual under at least eighteen years of age,
wherein the instructions further cause the system to determine the first speech processing results data with respect to a subset of functions corresponding to content selected for at least one individual user under at least eighteen years of age.
7. The system of claim 5 , wherein the instructions, when executed by the at least one processor, further cause the system to:
determine a device ID associated with the device; and
determine the device ID is associated with an indication of a user under at least eighteen years of age,
wherein the instructions further cause the system to determine the first speech processing results data with respect to a subset of functions corresponding to content selected for at least one individual user under at least eighteen years of age.
8. The system of claim 5 , wherein:
the first speech processing results data is associated with the dialog session ID based at least in part on the first utterance including a wakeword; and
the second speech processing results data is associated with the dialog session ID based at least in part on the dialog session ID being received with the second input audio data, the second utterance not including the wakeword.
9. The system of claim 5 , wherein the instructions, when executed by the at least one processor, further cause the system to:
determine the second speech processing results data correspond to a command inappropriate for at least one individual under at least eighteen years of age,
wherein the second output content corresponds to an indication that the command is inappropriate for at least one individual under at least eighteen years of age.
10. The system of claim 5 , wherein the instructions causing the system to generate the first output data further include instructions to:
send, to a first content source device, the first speech processing results data, the first user profile data, and the dialog session ID; and
receive, from the first content source, the first output data and the dialog session ID.
11. The system of claim 5 , wherein the instructions causing the system to generate the first output data further include instructions to:
determine the first user profile data indicates a first user age of less than eighteen years;
determine the second user profile data indicates a second user age of at least eighteen years; and
generate the second output data based at least in part on the first user age being less than eighteen years.
12. The system of claim 5 , wherein the instructions, when executed by the at least one processor, further cause the system to:
determine first audio characteristics corresponding to the first utterance;
determine the first audio characteristics correspond to first stored audio characteristic data associated with at least one individual under at least eighteen years of age;
determine second audio characteristics corresponding to the second utterance;
determine the second audio characteristics corresponding to second stored audio characteristic data associated with at least one individual eighteen years of age or older; and
generate the second output data based at least in part on the first audio characteristics corresponding to the first stored audio characteristic data.
13. A computer-implemented method comprising:
during a first time period:
receiving, from a device, first input audio data corresponding to a first utterance;
determining first user profile data associated with the first input audio data;
determining first speech processing results data based on the first input audio data;
associating the first speech processing results data with a dialog session identifier (ID);
generating first output data based on the first speech processing results data and the user profile data; and
sending, to the device, the first output data and the dialog session ID; and
during a second time period after the first time period:
receiving, from the device, the dialog session ID and second input audio data corresponding to a second utterance;
determining second speech processing results data based on the second input audio data;
associating the second speech processing results data with the dialog session ID;
determining second user profile data associated with the second input audio data;
generating second output data based on the second speech processing results data and the second user profile data; and
sending, to the device, the second output data and the dialog session ID.
14. The computer-implemented method of claim 13 , further comprising
determining first audio characteristics corresponding to the first utterance;
determining the first audio characteristics correspond to stored audio characteristic data associated with at least one individual under at least eighteen years of age; and
determining the first speech processing results data with respect to a subset of functions corresponding to content selected for at least one individual user under at least eighteen years of age.
15. The computer-implemented method of claim 13 , further comprising:
determining a device ID associated with the device;
determining the device ID is associated with an indication of a user under at least eighteen years of age; and
determining the first speech processing results data with respect to a subset of functions corresponding to content selected for at least one individual user under at least eighteen years of age.
16. The computer-implemented method of claim 13 , further comprising:
associating the first speech processing results data with the dialog session ID based at least in part on the first utterance including a wakeword; and
associating the second speech processing results data with the dialog session ID based at least in part on the dialog session ID being received with the second input audio data, the second utterance not including the wakeword.
17. The computer-implemented method of claim 13 , further comprising:
determining the second speech processing results data correspond to a command inappropriate for at least one individual under at least eighteen years of age,
wherein the second output content corresponds to an indication that the command is inappropriate for at least one individual under at least eighteen years of age.
18. The computer-implemented method of claim 13 , wherein generating the first output audio data further comprises:
sending, to a first content source device, the first speech processing results data, the first user profile data, and the dialog session ID; and
receiving, from the first content source, the first output data and the dialog session ID.
19. The computer-implemented method of claim 13 , further comprising:
determining the first user profile data indicates a first user age of less than eighteen years;
determining the second user profile data indicates a second user age of at least eighteen years; and
generating the second output data based at least in part on the first user age being less than eighteen years.
20. The computer-implemented method of claim 13 , further comprising:
determining first audio characteristics corresponding to the first utterance;
determining the first audio characteristics correspond to first stored audio characteristic data associated with at least one individual under at least eighteen years of age;
determining second audio characteristics corresponding to the second utterance;
determining the second audio characteristics corresponding to second stored audio characteristic data associated with at least one individual eighteen years of age or older; and
generating the second output data based at least in part on the first audio characteristics corresponding to the first stored audio characteristic data.