Calibrated Dual Omnidirectional Microphone Array Noise Cancellation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional multi-microphone noise suppression systems suffer from poor noise cancellation performance and significant speech distortion due to inadequate calibration of omnidirectional microphones, leading to inconsistent noise suppression across different environments and headsets.
Innovation Solution
A calibration method that accurately calibrates omnidirectional microphones in both amplitude and phase using a DSP-based compensation technique, forming virtual microphones with similar noise responses and dissimilar speech responses, thereby improving the signal-to-noise ratio (SNR) of desired speech.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional multi-microphone noise suppression systems are used, then noise cancellation is attempted by steering nulls to noise sources, but noise cancellation performance is poor and speech distortion is significant
Solution Approach 1:
The patent applies parameter changes by calibrating the amplitude and phase responses of omnidirectional microphones to match a reference response. This involves adjusting frequency-dependent gain and phase shift parameters across the frequency spectrum to ensure consistent noise suppression performance across different headsets and environmental conditions, thereby improving noise cancellation reliability while minimizing speech distortion
Solution Approach 2:
The patent implements preliminary action through a calibration process performed before noise suppression operation. The calibration establishes reference amplitude and phase responses for each microphone, enabling the system to pre-compensate for individual microphone characteristics. This preliminary calibration ensures that subsequent noise suppression operations produce consistent results across different headsets without causing speech distortion
2Adaptability or versatility
If omnidirectional microphones are used without calibration, then system simplicity is maintained, but noise suppression varies significantly across different headsets and environments
Solution Approach 1:
The patent replaces mechanical calibration adjustments with digital signal processing. Instead of physically adjusting microphone characteristics, the system uses DSP algorithms to compute and apply amplitude and phase correction filters. This substitution maintains system simplicity while achieving adaptability across different headsets, as the calibration can be performed software-based without mechanical modifications
Solution Approach 2:
The patent introduces a reference response as an intermediary standard for calibration. Each microphone's response is compared against this reference, and correction filters are designed to match the reference characteristics. This intermediary reference enables consistent noise suppression across different headsets without requiring complex pairwise calibration between microphones, balancing adaptability with system simplicity
Data Source
AI summary
Systems and methods are described by which microphones comprising a mechanical filter can be accurately calibrated to each other in both amplitude and phase.