Automated Assistant Response Adaptation for Resource-Aware Interaction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing automated assistants provide static responses to repeated user queries, regardless of the user's level of interaction, leading to inefficient use of resources and suboptimal user experience.
Innovation Solution
An automated assistant that modifies response features and processing based on a determined interaction measure, which quantifies the degree of interaction between the user and the assistant, by considering temporal proximity of utterances and adapting response robustness and processing techniques such as ASR and voice filtering.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the automated assistant provides static responses with the same amount of detail for all queries, then the system is simple to operate and consistent, but the response robustness and user engagement are insufficient
Solution Approach 1:
The patent implements dynamic response generation by adjusting the amount of detail in responses based on the determined level of interaction. The system transitions from static to dynamic behavior by modifying response robustness according to interaction metrics such as temporal proximity of utterances, enabling the assistant to provide more detailed responses when appropriate while maintaining simplicity otherwise.
Solution Approach 2:
The system changes the parameter of response detail level based on the interaction measure. By calculating temporal proximity between utterances and determining an interaction level, the system dynamically adjusts response parameters (amount of content, detail level) to optimize both robustness and resource usage.
2Reliability
If the automated assistant provides detailed robust responses for all queries, then the response robustness is improved, but the computational resources and network bandwidth are wasted
Solution Approach 1:
The system dynamically changes the response detail parameter based on the interaction level determined from temporal proximity analysis. When interaction level is low (isolated queries), the system reduces response detail to conserve resources. When interaction level is high (conversational context), the system increases response detail to improve robustness, optimizing the balance between quality and resource consumption.
Solution Approach 2:
The system applies partial action by providing only the necessary amount of detail in responses based on context. Instead of always providing maximum detail, the system provides just enough information appropriate to the interaction level, avoiding excessive resource consumption while maintaining adequate response quality.
3Productivity
If the automated assistant provides concise responses, then the resource consumption is reduced, but the user engagement and likelihood of follow-on queries increase
Solution Approach 1:
The system dynamically adjusts response length and detail based on the interaction level. For users showing high engagement (frequent temporal proximity), the system provides more detailed responses that may reduce follow-on queries. For users with low engagement, the system provides concise responses to conserve resources, accepting that additional interaction time may be needed.
4Adaptability or versatility
If the automated assistant adapts responses based on interaction level, then the user engagement is improved, but the processing complexity and determination of interaction measures increase
Solution Approach 1:
The system implements dynamic adaptability by determining interaction levels based on temporal proximity of utterances and adjusting response characteristics accordingly. This dynamic approach enables the system to adapt to different user contexts without requiring complex user modeling, achieving versatility through relatively simple temporal analysis.
Data Source
AI summary
Implementations set forth herein relate to an automated assistant that provides a response for certain user queries based on a level of interaction of the user with respect to the automated assistant. Interaction can be characterized by sensor data, which can be processed using one or more trained machine learning models in order to identify parameters for generating a response. In this way, the response can be limited to preserve computational resources and/or ensure that the response is more readily understood given the amount of interaction exhibited by the user. In some instances, a response that embodies information that is supplemental, to an otherwise suitable response, can be provided when a user is exhibiting a particular level of interaction. In other instances, such supplemental information can be withheld when the user is not exhibiting that particular level of interaction, at least in order to preserve computational resources.