Dual-Microphone Voice Detection With Adaptive Gain Thresholds
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Robust voice activity detection in noisy environments is challenging for small ear-mounted devices, often resulting in false positives and negatives due to reliance on fixed thresholds.
Innovation Solution
A method using two microphones, one ambient and one ear-canal, with band limiting or band pass filtering, calculating time-weighted power estimates, and adjusting gain based on the difference between fast and slow time-weighted ratios, with threshold adjustments for improved voice activity detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If fixed thresholds are used for voice activity detection, then the system is simple to implement, but false positives and false negatives increase in noisy environments
Solution Approach 1:
The patent implements dynamic threshold adjustment by continuously adapting the detection threshold based on the estimated noise level in the environment. The system calculates the noise floor and adjusts the voice activity detection threshold dynamically, allowing it to respond to changing acoustic conditions rather than relying on fixed thresholds. This resolves the contradiction by making the system adaptive to environmental variations while maintaining reasonable computational complexity.
Solution Approach 2:
The system changes the detection parameter (threshold) based on the operating conditions (noise level). By monitoring the acoustic environment and adjusting the threshold parameter accordingly, the system maintains high detection accuracy across different noise conditions. This parameter adaptation allows the simple fixed-threshold architecture to achieve reliable performance in varying environments.
2Measurement precision
If dynamic gain adjustment is implemented based on time-weighted ratios, then voice activity detection accuracy improves, but computational complexity increases
Solution Approach 1:
The patent segments the signal processing into distinct stages: noise estimation, ratio calculation, threshold comparison, and gain adjustment. By dividing the complex processing task into modular segments, the system achieves high detection accuracy through systematic analysis while keeping each individual processing stage computationally manageable. This segmentation allows implementation on resource-constrained ear-mounted devices.
Solution Approach 2:
The system applies partial processing by focusing computational resources on the most critical aspects of voice activity detection. Rather than analyzing all signal properties equally, it concentrates on time-weighted power ratios and noise floor estimation, achieving sufficient accuracy for voice-operated control without excessive computational burden.
3Reliability
If continuous monitoring of fast and slow time-weighted ratios is performed, then detection reliability improves, but power consumption increases
Solution Approach 1:
The system performs monitoring at periodic intervals rather than continuously, updating the fast and slow time-weighted ratios at controlled rates. This periodic updating maintains detection reliability by capturing voice activity events while significantly reducing the average power consumption compared to continuous monitoring. The system can adjust the update rate based on detected activity levels.
Solution Approach 2:
The system uses the incoming audio signal itself to drive the detection process, requiring minimal external processing power. By designing the algorithm to efficiently utilize the raw microphone input and perform localized computations, the system achieves reliable detection with low power consumption suitable for battery-operated ear-mounted devices.
Data Source
AI summary
An electronic device or method for adjusting a gain on a voice operated control system can include one or more processors and a memory having computer instructions. The instructions, when executed by the one or more processors causes the one or more processors to perform the operations of receiving a first microphone signal, receiving a second microphone signal, updating a slow time weighted ratio of the filtered first and second signals, and updating a fast time weighted ratio of the filtered first and second signals. The one or more processors can further perform the operations of calculating an absolute difference between the fast time weighted ratio and the slow time weighted ratio, comparing the absolute difference with a threshold, and increasing the gain when the absolute difference is greater than the threshold. Other embodiments are disclosed.