Warm Word Arbitration for Multi-User Voice Assistant Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Assistant-enabled devices struggle to efficiently manage multiple users' preferences for warm words due to computational limitations and differing tolerance levels, leading to increased false positives and negatives in shared environments.
Innovation Solution
A method and system for detecting multiple users within an environment, obtaining respective active sets of warm words, and executing a warm word arbitration routine to enable a final set of warm words for detection, considering proximity, user identity, and resource constraints, while prioritizing shared and preferred warm words.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the device enables warm words for multiple users simultaneously, then user interaction efficiency is improved, but false positives and false negatives increase due to computational limitations
Solution Approach 1:
The system dynamically adjusts the set of enabled warm words based on detected user presence. When multiple users are detected, the arbitration routine selectively enables warm words from their respective active sets, creating a dynamic configuration that adapts to current environmental conditions rather than maintaining a static set of warm words
Solution Approach 2:
The warm word arbitration routine changes the parameter of which warm words are enabled for detection based on user presence and device state. This parameter adjustment allows the system to optimize between enabling more warm words (improving productivity) and maintaining detection accuracy (reliability) by selectively enabling only appropriate warm words for current conditions
2Adaptability or versatility
If the device enables more warm words for detection, then the number of actionable commands increases, but computational resources are exceeded
Solution Approach 1:
The arbitration routine extracts and enables only the necessary subset of warm words from the complete active sets, rather than enabling all available warm words. This selective extraction allows the system to maintain versatility with multiple user preferences while consuming manageable computational resources by enabling only the most relevant warm words for current conditions
Solution Approach 2:
The system enables a partial set of warm words rather than all available warm words from multiple users. This partial action approach provides sufficient versatility for current user needs while staying within computational resource constraints, avoiding the excessive action of enabling every possible warm word
3Adaptability or versatility
If the device maintains separate active warm word sets for each user, then user-specific preferences are preserved, but device complexity increases
Solution Approach 1:
The arbitration routine merges the separate active warm word sets from multiple users into a single final enabled set. This combining process manages device complexity by consolidating multiple user preferences into one unified configuration, while still preserving user-specific preferences through selective inclusion in the merged set
Solution Approach 2:
The warm word arbitration routine acts as an intermediary between multiple users' active warm word sets and the device's enabled warm words. This mediator reconciles conflicting preferences and manages complexity by making selective decisions about which warm words to enable, rather than directly managing separate configurations for each user
Data Source
AI summary
A method includes detecting a presence of multiple users within an environment of an assistant-enabled device (AED) and obtaining, for each user, a respective active set of warm words that each specify a respective action for a digital assistant to perform. Based on each respective active set of warm words, the method also includes executing a warm word arbitration routine to enable a final set of warm words for detection by the AED. Here, the final set of warm words includes warm words selected from the respective active set of warm words. While the final set of warm words are enabled, the method also includes receiving audio data corresponding to an utterance captured by the AED, detecting a warm word from the final set of warm words in the audio data, and instructing the digital assistant to perform the respective action specified by the detected warm word.


