Assistant Device Arbitration Using Interaction Cues
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In environments with multiple automated assistant devices, simultaneous detection of user utterances leads to wasted computing resources and suboptimal user experiences due to multiple devices responding to the same query, particularly affecting older and lower-resource devices.
Innovation Solution
Implementing arbitration between automated assistant devices based on interaction cues, such as user touch and gestural signals, to identify a target device for responding to user queries, suppressing other devices' actions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple automated assistant devices detect the same spoken utterance, then each device can respond to provide comprehensive service, but computing resources are wasted and user experience deteriorates due to redundant responses
Solution Approach 1:
The system performs preliminary arbitration before full processing by first detecting which devices have detected the hotword and are candidates for response. This preliminary identification of target devices allows the system to avoid unnecessary processing in devices that should not respond, thus preventing computing resource waste while maintaining comprehensive service coverage through proper device selection.
Solution Approach 2:
The arbitration mechanism uses feedback from interaction cues (touch events, gestural signals) to dynamically determine which device should respond to the utterance. This feedback loop allows the system to adapt the response behavior based on real-time user interaction context, ensuring resources are allocated only to devices that need to respond, thereby reducing waste while maintaining service comprehensiveness.
2Productivity
If multiple automated assistant devices respond to the same query, then user service coverage is improved, but processing costs increase and user experience deteriorates due to redundant responses
Solution Approach 1:
The system applies local quality by determining response eligibility individually for each device based on its specific interaction context and user engagement level. Not all devices that detect the hotword are treated equally; each device's response capability is locally evaluated based on factors like user touch interaction with that specific device, ensuring processing occurs only where necessary and avoiding redundant responses across all devices.
Solution Approach 2:
The arbitration process performs preliminary determination of which devices should respond before actual query processing begins. This preliminary action filters out devices that should not respond based on interaction cues, reducing the number of devices that undergo full processing and thereby lowering overall processing costs while maintaining adequate service response capability through selective device activation.
3Loss of energy
If arbitration is implemented to select a target device, then processing costs are reduced, but device complexity increases due to arbitration logic and interaction cue processing
Solution Approach 1:
The arbitration system leverages existing multi-functional capabilities of automated assistant devices, particularly their ability to detect various interaction cues (touch events, gestural signals, audio input). Rather than adding entirely new dedicated arbitration hardware, the system utilizes these existing multi-functional sensors and processors to perform arbitration, thereby reducing the increase in device complexity while still achieving processing cost reduction through intelligent device selection.
Solution Approach 2:
The arbitration mechanism acts as an intermediary layer between the hotword detection stage and the full query processing stage. This intermediary arbitration logic uses interaction cues as mediators to determine response eligibility, filtering out unnecessary processing before it occurs. By positioning arbitration at this intermediate stage rather than requiring complex processing in every device, the system reduces overall processing costs while limiting the complexity increase to a manageable intermediary layer.
Data Source
AI summary
Techniques are described herein for arbitration between automated assistant devices based on interaction cues. A method includes: receiving, via one or more microphones of a first computing device, first audio data that captures a spoken utterance of a user; determining that each of one or more additional computing devices has detected the spoken utterance of the user; determining that hotword arbitration is to be initiated between the first computing device and the one or more additional computing devices; for each of the first computing device and the one or more additional computing devices, identifying a similarity score for the computing device; selecting a target computing device, from the first computing device and the one or more additional computing devices, based on the similarity scores; and causing the target computing device to respond to a query that is included in the spoken utterance of the user.