Dual-Microphone Noise Reduction via Transfer Function Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing noise reduction algorithms for communication devices are of high computational complexity and primarily single-channel, struggling to effectively estimate and suppress non-stationary noise, especially in dual-microphone systems.
Innovation Solution
A dual-channel noise power spectral density estimator that utilizes knowledge of noise field coherence, with low computational complexity, to enhance speech quality by exploiting a secondary microphone channel and combining with existing single-channel noise suppression systems.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If single-channel noise reduction algorithms are used in communication devices, then the device complexity is low, but the noise reduction effectiveness is insufficient especially for non-stationary noise
Solution Approach 1:
The patent transitions from single-channel to dual-channel noise reduction by utilizing a second microphone to capture additional spatial information. This dimensional expansion enables the system to estimate noise power spectral density more accurately by comparing signals from multiple spatial locations, thereby improving noise reduction effectiveness while maintaining computational feasibility through structured mathematical approaches.
Solution Approach 2:
The patent introduces transfer functions as an intermediary mathematical tool to model the relationship between the second microphone signal and the first microphone signal. This intermediary representation allows the system to leverage the second microphone's spatial information without directly processing complex multi-channel data, thus improving noise estimation accuracy while controlling algorithmic complexity.
2Reliability
If dual-microphone systems are implemented for noise reduction, then the noise estimation robustness is improved, but the computational complexity increases
Solution Approach 1:
The patent transforms the dual-channel noise reduction problem into a parameter estimation problem by focusing on computing the noise power spectral density through specific mathematical parameters (transfer functions and spectral ratios). This parameter-based approach consolidates the complex dual-channel processing into manageable computational steps, improving noise estimation robustness while controlling the computational burden through efficient parameter calculation.
3Reliability
If existing single-channel noise suppression systems are used, then the computational complexity is low, but the speech enhancement performance is limited in diffuse noise fields
Solution Approach 1:
The patent creates a scalable dual-channel noise reduction framework that can generalize from single-channel systems. The transfer function-based approach serves as a universal solution that works for both single and dual microphone configurations, allowing the system to adapt to different noise environments (including diffuse noise fields) while maintaining a unified computational structure that supports future extensions to more complex multi-microphone systems.
Data Source
AI summary
A method, system, and computer program product for managing noise in a noise reduction system, comprising: receiving a first signal at a first microphone; receiving a second signal at a second microphone; identifying noise estimation in the first signal and the second signal; identifying a transfer function of the noise reduction system using a ratio of a power spectral density of the second signal minus the noise estimation to a power spectral density of the first signal, wherein the noise estimation is removed from only the power spectral density of the second signal; and identifying a gain of the noise reduction system using the transfer function.