Audio Noise Suppression via Power Level Difference Correction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing audio signal processing techniques face challenges in accurately estimating noise and sound power level differences between primary and reference channels, leading to suboptimal noise suppression and distortion in audio signals due to differences in microphone sensitivity and orientation.

Innovation Solution

The method involves estimating frequency-dependent Noise Power Level Difference (NPLD) and Speech Power Level Difference (SPLD) using maximum likelihood techniques and data-driven recursive noise power estimation, which corrects noise variance estimates and modifies filtering processes to account for these differences.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Object-affected harmful factors

If noise suppression is performed using reference channel without correcting for power level differences, then noise removal is attempted, but distortion of targeted sound occurs due to over-suppression when reference noise level exceeds primary channel noise level

Engineering Contradiction:

Improvenoise in audio signalVSAvoiddistortion of targeted sound

Core Design Contradiction:

Object-affected harmful factorsVSManufacturing precision

Solution Approach 1:

The patent applies parameter changes by estimating and correcting power level differences between reference and primary channels. Specifically, it estimates noise power level difference (NPLD) and speech power level difference (SPLD) as correction parameters, then applies these to adjust the reference channel's noise variance estimate before using it for noise suppression, thereby preventing over-suppression distortion

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent implements feedback by continuously estimating NPLD and SPLD from the audio signals and using these estimates to dynamically adjust the noise suppression process. The estimated power level differences feed back into the noise variance correction, creating a closed-loop system that adapts to varying noise and speech conditions

Inventive Principle:

Principle #23Feedback

2Device complexity

If noise suppression is performed assuming reference channel noise level equals primary channel noise level, then processing is simplified, but under-suppression occurs when reference noise level is lower than primary channel noise level

Engineering Contradiction:

Improvenoise suppression processingVSAvoidresidual noise in audio signal

Core Design Contradiction:

Device complexityVSObject-affected harmful factors

Solution Approach 1:

The patent modifies the noise variance estimate of the reference channel by applying NPLD correction parameters. This parameter change ensures that even when the reference channel records lower noise levels, the corrected estimate accurately reflects the primary channel's noise level, enabling effective noise suppression without requiring complex alternative processing paths

Inventive Principle:

Principle #35Parameter changes

3Device complexity

If targeted sound present in reference channel is not accounted for, then noise estimation is simplified, but error in noise level estimate occurs leading to degraded audio quality

Engineering Contradiction:

Improvenoise estimation processVSAvoidnoise level estimate accuracy

Core Design Contradiction:

Device complexityVSMeasurement precision

Solution Approach 1:

The patent extracts and separates the speech component from the reference channel signal by estimating SPLD. It removes the speech power contribution from the reference channel's total power, isolating the noise component for accurate estimation. This extraction process eliminates the error caused by unaccounted targeted sound while maintaining reasonable processing complexity

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent applies parameter changes by estimating SPLD as a correction parameter and using it to adjust the reference channel's power level. This parameter adjustment transforms the contaminated noise estimate into an accurate noise-only estimate, improving measurement precision without substantially increasing device complexity

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS10332541B2Determining noise and sound power level differences between primary and reference channels

Publication Date: 2019.06.25 CIRRUS LOGIC INC

AI summary

A method for estimating and minimizing a noise power level difference (NPLD) between a primary channel and a reference channel of an audio device, includes receiving, by a primary channel, an audio signal that has a speech signal level and a noise signal level; receiving, by a reference channel, the audio signal with another speech signal level and another noise signal level; using the reference channel to estimate the noise signal level in the primary channel by reducing the another speech signal level; and compensating for a difference between the noise signal level and the another noise signal level to minimize a noise power level difference NPLD between the primary channel and the reference channel.