Local Device Arbitration for Speech Interface Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing speech interface systems face challenges in efficiently arbitrating which device should respond to user speech when multiple devices detect an utterance, especially in environments with varying sound capture latency and internet connectivity issues.
Innovation Solution
Implementing a time-based local device arbitration technique that designates the most appropriate speech interface device to perform an action by comparing wakeword occurrence times and using a pre-calculated time offset value to synchronize devices, allowing for effective device selection even without internet connectivity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If cloud-based services are accessed through speech interface devices, then service functionality is improved, but dependency on internet connectivity increases
Solution Approach 1:
The system performs preliminary actions by designating a master speech interface device in advance using time-based arbitration. This master device is pre-configured to handle speech processing tasks, ensuring that the system can immediately respond to user speech even when internet connectivity is unavailable or slow.
Solution Approach 2:
The patent introduces an intermediary mechanism (time-based arbitration system) that operates locally on speech interface devices to determine which device should process speech. This intermediary layer enables devices to autonomously make decisions about speech processing without requiring real-time cloud connectivity, thus reducing dependency on internet while maintaining service functionality.
2Adaptability or versatility
If multiple speech interface devices are deployed in proximity, then device coverage and user convenience are improved, but device arbitration complexity increases
Solution Approach 1:
The system changes the parameter used for arbitration from complex multi-factor evaluation to a simple time-based parameter. By measuring and comparing the timing of wakeword detection across multiple devices, the system achieves efficient arbitration with minimal computational complexity, enabling multiple devices to be deployed without proportionally increasing system complexity.
Solution Approach 2:
The patent replaces complex mechanical or algorithmic arbitration mechanisms with a simpler time-synchronization approach. Instead of using complex decision-making algorithms or hierarchical device structures, the system uses time stamps and clock synchronization to automatically determine which device detected the wakeword first, significantly reducing arbitration complexity.
3Measurement precision
If time-based arbitration is used to designate the closest device, then device selection accuracy is improved, but synchronization precision requirements increase
Solution Approach 1:
The system performs preliminary clock synchronization between all speech interface devices before arbitration occurs. This preliminary action ensures that all devices operate from a common time reference, allowing accurate time-based comparison of wakeword detection events without requiring extremely high-precision hardware clocks or frequent synchronization during operation.
Data Source
AI summary
A device can perform device arbitration, even when the device is unable to communicate with a remote system over a wide area network (e.g., the Internet). Upon detecting a wakeword in an utterance, the device can wait a period of time for data to arrive at the device, which, if received, indicates to the device that another speech interface device in the environment detected an utterance. If the device receives data prior to the period of time lapsing, the device can determine the earliest-occurring wakeword based on multiple wakeword occurrence times, and may designate whichever device that detected the wakeword first as the designated device to perform an action with respect to the user speech. To account for differences in sound capture latency between speech interface devices, a pre-calculated time offset value can be applied to wakeword occurrence time(s) during device arbitration.