Multi-Assistant Speech Selection for Single-Device Response
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Managing natural user interaction interfaces across multiple computing devices in a shared environment to provide a positive user experience is challenging due to the potential for multiple devices to respond to the same user, leading to disorientation and unnecessary resource consumption.
Innovation Solution
An intelligent assistant system that records speech, determines a self-selection score, and compares it with a remote-selection score to decide which device should respond, blocking other devices until a disengagement metric exceeds a threshold.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple intelligent assistant devices are deployed in a shared environment to serve multiple users, then the system coverage and user service capability are improved, but the risk of multiple devices responding to the same user simultaneously increases, causing user disorientation and resource waste
Solution Approach 1:
The patent introduces a coordinator service as an intermediary component that manages communication between multiple intelligent assistant devices. This coordinator receives requests from devices, determines which device should handle each user request based on various criteria, and routes requests appropriately. This mediator prevents multiple devices from responding simultaneously by centralizing the decision-making process, thus resolving the contradiction between system coverage and response coordination reliability.
Solution Approach 2:
The system implements feedback mechanisms where devices report their state, user presence, and response status to the coordinator. The coordinator uses this feedback information to make informed decisions about request routing. This feedback loop ensures that when one device is already engaged with a user, other devices can detect this state and refrain from responding, maintaining reliable coordinated responses while preserving multi-device deployment benefits.
2Ease of operation
If all intelligent assistant devices continuously monitor and respond to user inputs, then the responsiveness and user experience are improved, but the energy consumption and computational resources are unnecessarily depleted
Solution Approach 1:
The patent implements dynamic power management where devices can transition between active and low-power states based on their engagement status and environmental context. When a device is not the selected responder or when no user presence is detected, it can enter a lower-power mode while maintaining basic listening capabilities. This dynamic state adjustment maintains responsiveness when needed while significantly reducing energy consumption during idle periods, resolving the contradiction between responsiveness and energy usage.
3Productivity
If multiple devices independently process user requests without coordination, then the processing speed and individual device autonomy are improved, but the user experience deteriorates due to conflicting responses and system confusion
Solution Approach 1:
The system performs preliminary actions by pre-establishing communication channels, device capability profiles, and coordination protocols before user interactions begin. Devices register themselves with the coordinator service in advance, sharing their capabilities and current state information. This preliminary setup enables rapid coordinated response during actual user interactions, as the framework for collaboration is already in place, resolving the contradiction between processing speed and user experience by preparing the coordination infrastructure beforehand.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An intelligent assistant records speech spoken by a first user and determines a self-selection score for the first user. The intelligent assistant sends the self-selection score to another intelligent assistant, and receives a remote-selection score for the first user from the other intelligent assistant. The intelligent assistant compares the self-selection score to the remote-selection score. If the self-selection score is greater than the remote-selection score, the intelligent assistant responds to the first user and blocks subsequent responses to all other users until a disengagement metric of the first user exceeds a blocking threshold. If the self-selection score is less than the remote-selection score, the intelligent assistant does not respond to the first user.