Systems and method to resolve audio-based requests in a networked environment
Techniques are described herein for enabling an automated assistant to adjust its behavior depending on a detected vocabulary level or other vocal characteristics of an input utterance provided to an automated assistant. The estimated vocabulary level or other vocal characteristics may be used to influence various aspects of a data processing pipeline employed by the automated assistant. In some implementations, one or more tolerance thresholds associated with, for example, grammatical tolerances or vocabulary tolerances, may be adjusted based on the estimated vocabulary level or vocal characteristics of the input utterance.
1. A system to resolve requests in audio-based networked environments, comprising:
a data processing system having one or more processors coupled with memory to:
parse an input audio acquired via a sensor at a client device to determine a characteristic, the characteristic including at least one of a vocal characteristic or a vocabulary level;
select, from a plurality of interaction models, an interaction model based on the characteristic and a tolerance of the interaction model;
identify a digital component based on one or more keywords identified from the input audio in accordance with the interaction model; and
provide, in response to the input audio, the digital component to present via the client device.
2. The system of claim 1 , comprising the data processing system to:
identify a trigger to disambiguate the parsing of the input audio; and
select, responsive to the identification of the trigger to disambiguate, the interaction model from the plurality of interaction models to process the input audio.
3. The system of claim 1 , comprising the data processing system to:
identify an account profile associated with the client device via which the input audio is acquired; and
select, from the plurality of interaction models, the interaction model based on the account profile.
4. The system of claim 1 , comprising the data processing system to:
parse, using a natural language processor, the input audio to identify a plurality of candidate intents;
select, from the plurality of candidate intents, an intent based on the one or more keywords of the input audio identified in accordance with the interaction model; and
identify the digital component based on the intent identified from the plurality of candidate intents.
5. The system of claim 1 , comprising the data processing system to:
identify an intent based on at least one of the one or more keywords identified from the input audio in accordance with the interaction model, the intent defined for a third-party application interfacing with a digital assistant application executing on the client device; and
identify the digital component based on the intent defined for the third-party application.
6. The system of claim 1 , comprising the data processing system to:
identify an intent based on a mapping defined by the interaction model between at least one of the one or more keywords to one of a plurality of intents; and
identify the digital component based on the intent defined by the mapping.
7. The system of claim 1 , comprising the data processing system to:
identify a first attribute from the input audio using the interaction model; and
determine, using an entity data structure defining relationships among a plurality of attributes, a second attribute associated with first attribute to identify the digital component.
8. The system of claim 1 , comprising the data processing system to:
select the interaction model from the plurality of interaction models, each of the plurality of interaction models having a tolerance level for at least one of a pronunciation, a grammar, and a vocabulary in processing the input audio; and
modify a first keyword parsed from the input audio to a second keyword in accordance with the tolerance level of the interaction model to identify the digital component.
9. The system of claim 1 , comprising the data processing system to:
select, from a plurality of voice synthesis models, a voice synthesis model based on the characteristic determined from the input audio; and
provide the digital component to present via the client device using the voice synthesis model.
10. The system of claim 1 , comprising the data processing system to
identify a second digital component associated with the digital component based on the one or more keywords identified from the input audio in accordance with the interaction model; and
provide, in response to the input audio, the digital component and the second digital component to present via the client device.
11. A method of resolving requests in audio-based networked environments, comprising:
parsing, by a data processing system, an input audio acquired via a sensor at a client device to determine a characteristic, the characteristic including at least one of a vocal characteristic or a vocabulary level;
selecting, by the data processing system, from a plurality of interaction models, an interaction model based on the characteristic and a tolerance of the interaction model;
identifying, by the data processing system, a digital component based on one or more keywords identified from the input audio in accordance with the interaction model; and
providing, by the data processing system, in response to the input audio, the digital component to present via the client device.
12. The method of claim 11 , comprising:
identifying, by the data processing system, a trigger to disambiguate the parsing of the input audio; and
selecting, by the data processing system, responsive to the identification of the trigger to disambiguate, the interaction model from the plurality of interaction models to process the input audio.
13. The method of claim 11 , comprising:
identifying, by the data processing system, an account profile associated with the client device via which the input audio is acquired; and
selecting, by the data processing system, from the plurality of interaction models, the interaction model based on the account profile.
14. The method of claim 11 , comprising:
parsing, by the data processing system, using a natural language processor, the input audio to identify a plurality of candidate intents;
selecting, by the data processing system, from the plurality of candidate intents, an intent based on the one or more keywords of the input audio identified in accordance with the interaction model; and
identifying, by the data processing system, the digital component based on the intent identified from the plurality of candidate intents.
15. The method of claim 11 , comprising:
identifying, by the data processing system, an intent based on at least one of the one or more keywords identified from the input audio in accordance with the interaction model, the intent defined for a third-party application interfacing with a digital assistant application executing on the client device; and
identifying, by the data processing system, the digital component based on the intent defined for the third-party application.
16. The method of claim 11 , comprising:
identifying, by the data processing system, an intent based on a mapping defined by the interaction model between at least one of the one or more keywords to one of a plurality of intents; and
identifying, by the data processing system, the digital component based on the intent defined by the mapping.
17. The method of claim 11 , comprising:
identifying, by the data processing system, a first attribute from the input audio using the interaction model; and
determining, by the data processing system, using an entity data structure defining relationships among a plurality of attributes, a second attribute associated with first attribute to identify the digital component.
18. The method of claim 11 , comprising:
selecting, by the data processing system, the interaction model from the plurality of interaction models, each of the plurality of interaction models having a tolerance level for at least one of a pronunciation, a grammar, and a vocabulary in processing the input audio; and
modifying, by the data processing system, a first keyword parsed from the input audio to a second keyword in accordance with the tolerance level of the interaction model to identify the digital component.
19. The method of claim 11 , comprising:
selecting, by the data processing system, from a plurality of voice synthesis models, a voice synthesis model based on the characteristic determined from the input audio; and
providing, by the data processing system, the digital component to present via the client device using the voice synthesis model.
20. The method of claim 11 , comprising:
identifying, by the data processing system, a second digital component associated with the digital component based on the one or more keywords identified from the input audio in accordance with the interaction model; and
providing, by the data processing system, in response to the input audio, the digital component and the second digital component to present via the client device.