Resonator Microphone Array for Low-Power Voice Triggering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing auto voice trigger methods for voice recognition and speaker authentication are inefficient due to high computational and power consumption, and accuracy is reduced in noisy environments, as they require continuous microphone usage and analyze wideband signals, which can be triggered by loud noise.
Innovation Solution
An auto voice trigger method utilizing a resonator microphone array with different frequency bandwidths, where only a subset of microphones is initially used to determine if a signal is a voice signal, and upon confirmation, the entire system is activated to analyze the wideband signal, reducing computational and power consumption while enhancing accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a microphone is always turned on and real-time signal analysis is performed, then voice trigger accuracy is improved, but power consumption and computational amount increase
Solution Approach 1:
The patent divides the frequency spectrum into multiple frequency bands and uses separate resonator microphones for each band. This segmentation allows the system to monitor only specific voice-relevant frequency ranges initially, reducing overall power consumption while maintaining voice detection accuracy. When voice is detected in a specific band, the system then activates wider band monitoring only for that frequency range.
Solution Approach 2:
The system performs partial action by initially monitoring only narrow frequency bands relevant to voice signals rather than analyzing the entire frequency spectrum continuously. This partial monitoring reduces computational load and power consumption. When voice is detected, the system then performs excessive action by activating full wideband analysis only for the specific frequency band where voice was detected, rather than maintaining full power consumption continuously.
2Ease of operation
If the entire frequency band signal energy is used for voice detection, then detection simplicity is improved, but accuracy is reduced in noisy environments
Solution Approach 1:
The patent segments the frequency spectrum into multiple bands and uses dedicated resonator microphones for each band. This segmentation allows the system to focus on voice-relevant frequency ranges while filtering out noise in other bands, improving detection accuracy without significantly increasing complexity. The system processes each frequency band separately and combines results.
Solution Approach 2:
The system applies local quality by using resonator microphones with specific frequency response characteristics tailored to each frequency band. Each resonator microphone is optimized for its specific band, providing enhanced sensitivity to voice signals in that range while naturally attenuating noise in other frequency ranges, thereby improving overall detection accuracy.
3Measurement precision
If frequency band separation and analysis is performed to increase noise resilience, then voice trigger accuracy is improved, but computational amount and power consumption increase
Solution Approach 1:
The patent uses physical segmentation through resonator microphones that are inherently tuned to specific frequency bands. This physical segmentation in the hardware layer simplifies the signal processing complexity compared to software-based frequency separation, as each microphone naturally outputs only its designated frequency band signal without requiring complex digital filtering.
Solution Approach 2:
The patent replaces complex digital signal processing mechanisms with mechanical resonator microphones that physically filter frequency bands through their resonant properties. This mechanical frequency selection substitutes for complex digital filtering and spectral analysis, reducing computational complexity while maintaining frequency band separation capabilities.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach reduces power consumption and computational load while maintaining high accuracy in voice signal detection, even in noisy environments, by selectively activating microphones and using a subset to determine voice signals before full system activation.
Implementation Method 1
receiving a signal by at least one resonator microphone included in an array of a plurality of resonator microphones with different frequency bandwidths
Data Source
AI summary
An auto voice trigger method and an audio analyzer employing the same are provided. The auto voice trigger method includes: receiving a signal by at least one resonator microphone included in an array of a plurality of resonator microphones with different frequency bandwidths; analyzing the received signal and determining whether the received signal is a voice signal; and when it is determined that the received signal is the voice signal, waking up a whole system to receive and analyze a wideband signal.