Single Microphone Noise Detection Using Segmentation and Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional noise detectors and audio noise compensation systems are slow, inaccurate, and costly, often misidentifying speech as noise, leading to undesirable volume adjustments and requiring multiple microphones that become mismatched over time, increasing costs and discomfort.
Innovation Solution
A system using a single microphone with a novel noise filter arrangement and statistics generator to analyze noise power levels, distinguishing between noise and speech, and providing quick and accurate noise level measurements for effective noise compensation through system gain control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the attack time of power estimator is reduced to avoid responding to caller's speech, then the noise detection accuracy is improved, but the response time becomes unacceptably slow
Solution Approach 1:
The patent segments the noise detection process into multiple independent components: a noise detector that continuously monitors for noise events, a speech detector that identifies speech segments, and a controller that integrates both signals. This segmentation allows each component to operate optimally without compromising the other, resolving the contradiction between fast response and accurate noise detection.
Solution Approach 2:
The patent implements feedback mechanisms where the speech detector continuously monitors the audio signal and provides feedback to the controller. When speech is detected, the controller adjusts the noise detection sensitivity, creating a closed-loop system that adapts to changing acoustic conditions and maintains both fast response and accurate measurement.
2Measurement precision
If multiple microphones are used to detect caller's voice and background noise separately, then the noise level measurement accuracy is improved, but the device complexity and cost increase
Solution Approach 1:
The patent makes the single microphone multi-functional by implementing both a noise detector and a speech detector that process the same audio signal. The noise detector identifies noise events while the speech detector identifies speech segments, allowing one microphone to perform functions that would traditionally require multiple specialized microphones.
Solution Approach 2:
The system uses self-service processing where the audio signal is analyzed by multiple detection algorithms simultaneously. The noise detector and speech detector both process the same microphone input independently, and their results are combined by the controller, eliminating the need for additional microphones while maintaining measurement accuracy.
3Measurement precision
If multiple microphones are used to improve noise detection, then the noise level measurement is improved, but the microphones become mismatched over time resulting in degraded operation
Solution Approach 1:
The patent segments the detection functions into separate algorithms (noise detector and speech detector) that process the same signal independently. This segmentation eliminates the need for multiple physical microphones and their associated matching problems, as the algorithms can be software-based and thus more stable over time.
Solution Approach 2:
The patent replaces the mechanical system of multiple physical microphones with a computational system using signal processing algorithms. The noise detector and speech detector are implemented as software modules that analyze the audio signal, eliminating the physical mismatch issues that arise with multiple microphones aging at different rates.
4Measurement precision
If bone conduction microphone is used to detect caller's voice, then the voice detection accuracy is improved, but the user comfort is reduced due to skin contact requirements
Solution Approach 1:
The patent extracts the voice detection function from the acoustic microphone system and implements it through signal processing algorithms. The speech detector analyzes the audio signal captured by the regular microphone to identify speech segments, eliminating the need for bone conduction technology and its associated discomfort.
Solution Approach 2:
The patent replaces the mechanical bone conduction system with a computational approach. Instead of using a bone conduction microphone that requires skin contact, the system uses digital signal processing algorithms to detect and remove speech from the audio signal, achieving the same functional goal without the physical discomfort.
Data Source
AI summary
Systems and methods are described which facilitate quick and accurate extraction of the true noise level from a noise signal that includes additional signals, such as speech, in a cost effective implementation. Aspects of the invention allow the use of one microphone to simultaneously detect background noise as well as speech, while avoiding problems associated with artificially high background noise indication due to inclusion of the speech component in the noise determination. Additionally, systems and methods are described for altering system gain based on accurate noise level determinations.