Voice Command Detection in Mobile Devices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional voice response systems in mobile devices require explicit triggers, such as manual actions or specific voice commands, to engage the voice response system, which can be inconvenient, especially when the device is in a low power mode, limiting hands-free interaction and responsiveness.
Innovation Solution
A method and system that monitor the acoustic environment to detect voice commands without explicit triggers, using a multi-stage processing approach with low power processing stages and contextual cues to differentiate between voice commands and spurious acoustic activity, allowing the device to remain responsive in both active and low power modes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Use of energy by moving object
If the device operates in low power mode to conserve battery life, then energy consumption is reduced, but the voice response system becomes unresponsive and requires multiple manual wake-up actions
Solution Approach 1:
The patent segments the voice processing system into two distinct components: a low-power voice trigger detector that remains active during sleep mode to detect wake words, and a full voice response system that activates only when triggered. This segmentation allows the device to maintain voice responsiveness in low power mode by keeping only the essential trigger detection functionality running, rather than requiring full system wake-up for simple voice triggers.
Solution Approach 2:
The patent implements preliminary action by pre-configuring the device with a dedicated low-power voice trigger detector that is ready to detect wake words at any time, even when the device is in sleep mode. This preliminary detection capability eliminates the need for manual wake-up actions, as the system is already prepared to respond to voice triggers without requiring the user to first activate the device.
2Use of energy by moving object
If manual trigger actions are required to activate the voice response system, then power consumption is reduced, but user convenience deteriorates especially when hands are occupied
Solution Approach 1:
The patent implements self-service by enabling the voice trigger detector to autonomously detect wake words and initiate voice response system activation without requiring any manual user input. The system serves itself by automatically transitioning from low-power mode to active voice processing mode when it detects the appropriate trigger, providing true hands-free operation while maintaining power efficiency.
3Reliability
If explicit trigger words are required to engage the voice response system, then false positive detection is reduced, but the interaction becomes more complex and less natural
Solution Approach 1:
The patent extracts the trigger word detection functionality from the main voice response system and places it in a separate low-power detection module. This extraction allows the system to use simple keyword spotting for trigger detection, maintaining low computational complexity and low false positive rates, while the full voice response system remains dormant until activated. The separation enables natural, simple trigger phrases without requiring complex continuous voice processing.
Data Source
AI summary
According to some aspects, a method of monitoring an acoustic environment of a mobile device, at least one computer readable medium encoded with instructions that, when executed, perform such a method and/or a mobile device configured to perform such a method is provided. The method comprises receiving, by the mobile device, acoustic input from the environment of the mobile device, detecting whether the acoustic input includes a voice command from a user without requiring receipt of an explicit trigger from the user, and initiating responding to the detected voice command.