Automated Assistant Failover for Unfulfilled User Requests
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Multiple automated assistants on a device or nearby devices often fail to handle user requests optimally, leading to suboptimal or inaccurate responses, despite a better assistant being available.
Innovation Solution
An automated assistant detects failures in other assistants and offers to handle the user's request, either automatically or upon user request, by processing cached audio data to provide a fulfilling response.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple automated assistants are running on the same device or nearby devices, then the system provides more options for handling user requests, but the complexity of detecting and managing failures across assistants increases
Solution Approach 1:
The patent introduces an intermediary component (failure detection component) that mediates between multiple automated assistants and the user. This intermediary monitors assistant responses, detects failures, and coordinates failover to alternative assistants, thereby managing the complexity of multi-assistant coordination without requiring each assistant to independently monitor others.
Solution Approach 2:
The system implements feedback mechanisms where the failure detection component continuously monitors responses from automated assistants and provides feedback about success or failure. This feedback loop enables automatic detection of assistant failures and triggers appropriate failover actions, simplifying the management of multiple assistants through centralized monitoring and response coordination.
2Reliability
If the automated assistant continuously monitors other assistants for failures, then the reliability of request fulfillment improves, but the energy consumption and processing resources increase
Solution Approach 1:
The failure detection component performs preliminary monitoring of assistant responses to detect failures before they impact user experience. By proactively detecting failures in other assistants, the system can prepare failover responses in advance, ensuring reliable request fulfillment without requiring continuous high-energy processing during normal operation.
Solution Approach 2:
Each automated assistant independently processes its own requests and generates its own responses, while the failure detection component monitors these self-generated responses for failure conditions. This self-service approach allows assistants to operate autonomously with minimal energy overhead for monitoring, as each assistant manages its own execution while the system collectively ensures reliability through distributed failure detection.
Data Source
AI summary
Techniques are described herein for detecting and handling failures in other automated assistants. A method includes: executing a first automated assistant in an inactive state at least in part on a computing device operated by a user; while in the inactive state, determining, by the first automated assistant, that a second automated assistant failed to fulfill a request of the user; in response to determining that the second automated assistant failed to fulfill the request of the user, the first automated assistant processing cached audio data that captures a spoken utterance of the user comprising the request that the second automated assistant failed to fulfill, or features of the cached audio data, to determine a response that fulfills the request of the user; and providing, by the first automated assistant to the user, the response that fulfills the request of the user.