Architecture for multi-domain natural language processing
Features are disclosed for processing a user utterance with respect to multiple subject matters or domains, and for selecting a likely result from a particular domain with which to respond to the utterance or otherwise take action. A user utterance may be transcribed by an automatic speech recognition (“ASR”) module, and the results may be provided to a multi-domain natural language understanding (“NLU”) engine. The multi-domain NLU engine may process the transcription(s) in multiple individual domains rather than in a single domain. In some cases, the transcription(s) may be processed in multiple individual domains in parallel or substantially simultaneously. In addition, hints may be generated based on previous user interactions and other data. The ASR module, multi-domain NLU engine, and other components of a spoken language processing system may use the hints to more efficiently process input or more accurately generate output.
1. A system comprising:
computer-readable memory storing executable instructions; and
one or more processors in communication with the computer-readable memory,
wherein the one or more processors are programmed by the executable instructions to:
receive interaction data representing one or more user interactions;
store history data regarding the one or more user interactions;
generate future hint data using the interaction data, wherein the future hint data represents at least a first natural language understanding (“NLU”) domain with which a future user utterance is likely to be associated, and wherein the first NLU domain is one of a plurality of NLU domains;
establish a natural language processing session subsequent to generating the future hint data;
receive natural language data representing a user utterance occurring during the natural language processing session;
access, in response to receiving the natural language data, the future hint data representing at least the first NLU domain;
determine, based at least partly on the future hint data, that the natural language data is associated with the first NLU domain;
generate first intent data based at least partly on the natural language data, wherein the first intent data represents a first intent associated with the first NLU domain;
generate second intent data based at least partly on the natural language data, wherein the second intent data represents a second intent associated with a second NLU domain;
rank the first intent data above the second intent data based at least partly on the future hint data; and
generate response data based at least partly on the first intent data.
2. The system of claim 1 , wherein the one or more processors are further programmed by the executable instructions to determine, based at least partly on the future hint data, that the natural language data is associated with the first intent, wherein the first intent is one of a plurality of intents associated with the first NLU domain.
3. The system of claim 1 , wherein the one or more user interactions includes a non-utterance interaction of a user.
4. The system of claim 1 , wherein the first NLU domain is associated with intents related to at least one of: phone dialing, shopping, getting directions, playing music, or performing a search.
5. The system of claim 1 , wherein the one or more processors are further programmed by the executable instructions to:
analyze the interaction data and the history data; and
determine, based at least partly on a result of analyzing the interaction data, that the future user utterance is likely to be associated with the first NLU domain.
6. The system of claim 1 , wherein the one or more processors are further programmed by the executable instructions to generate NLU result data using the natural language data and a first NLU component of a plurality of NLU components,
wherein the first NLU component is associated with the first NLU domain, and
wherein the NLU result data represents a named entity of a plurality of named entities corresponding to an executable action associated with the first intent.
7. The system of claim 1 , wherein the one or more processors are further programmed by the executable instructions to:
receive audio data representing the user utterance; and
generate the natural language data using the audio data and an automatic speech recognition (“ASR”) component.
8. The system of claim 1 , wherein the executable instructions to generate the response data comprise executable instructions to:
generate response text data using a natural language generation component; and
generate response audio data using the response text data and a text-to-speech component.
9. The system of claim 1 , wherein one or more processors are further programmed by the executable instructions to send the response data to a remote device configured to present a response using the response data.
10. A computer-implemented method comprising:
under control of one or more computing devices configured with specific computer-executable instructions,
receiving interaction data representing one or more user interactions;
storing history data regarding the one or more user interactions;
generating future hint data using the interaction data, wherein the future hint data represents at least a first natural language understanding (“NLU”) domain with which a future user utterance is likely to be associated, and wherein the first NLU domain is one of a plurality of NLU domains;
establishing a natural language processing session subsequent to generating the future hint data;
receiving natural language data representing a user utterance occurring during the natural language processing session;
accessing, in response to receiving the natural language data, the future hint data representing at least the first NLU domain;
determining, based at least partly on the future hint data, that the natural language data is associated with the first NLU domain;
generating first intent data based at least partly on the natural language data, wherein the first intent data represents a first intent associated with the first NLU domain;
generating second intent data based at least partly on the natural language data, wherein the second intent data represents a second intent associated with a second NLU domain;
ranking the first intent data above the second intent data based at least partly on the future hint data; and
generating response data based at least partly on the first intent data.
11. The computer-implemented method of claim 10 , further comprising determining, based at least partly on the future hint data, that the natural language data is associated with the first intent, wherein the first intent is one of a plurality of intents associated with the first NLU domain.
12. The computer-implemented method of claim 10 , wherein ranking the first intent data above the second intent data based at least partly on the future hint data comprises adjusting a rank of at least one of the first intent data or the second intent data.
13. The computer-implemented method of claim 10 , wherein receiving the interaction data comprises receiving data representing a non-utterance interaction of a user.
14. The computer-implemented method of claim 10 , wherein receiving the interaction data comprises receiving at least a portion of the interaction data in response to prompt.
15. The computer-implemented method of claim 10 , further comprising:
analyzing the interaction data and the history data; and
determining, based at least partly on a result of analyzing the interaction data, that the future user utterance is likely to be associated with the first NLU domain.
16. The computer-implemented method of claim 10 , further comprising generating NLU result data using the natural language data and a first NLU component of a plurality of NLU components,
wherein the first NLU component is associated with the first NLU domain, and
wherein the NLU result data represents a named entity of a plurality of named entities corresponding to an executable action associated with the first intent.
17. The computer-implemented method of claim 10 , further comprising:
receiving audio data representing the user utterance; and
generating the natural language data using the audio data and an automatic speech recognition (“ASR”) component.
18. The computer-implemented method of claim 10 , wherein the generating the response data comprises:
generating response text data using a natural language generation component; and
generating response audio data using the response text data and a text-to-speech component.