Sound Processing Method for Diffuse Noise Suppression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional noise suppression methods in mobile phones, such as adaptive blocking matrix and adaptive noise canceller, are ineffective in diffuse noise environments due to their inability to handle multiple spatial reflections and rapid changes in transfer functions between microphone channels, resulting in poor noise elimination and low signal-to-noise ratio.
Innovation Solution
A sound processing method using two microphones, where a vector of a residual signal is determined from input signals of both microphones, and a gain function is calculated to differentially process the signals, enhancing the signal-to-noise ratio by offsetting noise and voice signals, thereby improving noise elimination and obtaining a purer voice signal.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If adaptive filter is used to eliminate noise, then noise elimination is attempted, but the effect on noise elimination is poor and signal-to-noise ratio remains low
Solution Approach 1:
The patent introduces a noise reference signal as an intermediary element. This reference signal, obtained from a noise reference microphone or processed from the main channel, serves as a mediator to help the adaptive filter identify and eliminate noise components more effectively. The reference signal provides additional information about the noise characteristics, enabling the filter to distinguish noise from voice signals more accurately and improve the overall noise elimination effect.
2Object-affected harmful factors
If traditional adaptive blocking matrix or adaptive noise canceller is used, then noise suppression is attempted, but the method is ineffective in diffuse noise environments due to inability to handle multiple spatial reflections and rapid changes in transfer functions
Solution Approach 1:
The patent employs dynamic adaptive filtering where the filter coefficients are continuously updated to track rapid changes in transfer functions. The adaptive algorithm adjusts the filter parameters in real-time based on the changing acoustic environment, enabling the system to adapt to diffuse noise conditions with multiple spatial reflections. This dynamic adaptation allows the noise suppression method to remain effective even when the noise characteristics change rapidly.
Solution Approach 2:
The patent implements feedback mechanisms where the output of the adaptive filter is continuously monitored and fed back to adjust the filter coefficients. This feedback loop enables the system to learn from its performance and continuously improve its noise suppression capability. The feedback mechanism allows the system to adapt to diffuse noise environments by adjusting to the complex interference patterns created by multiple spatial reflections.
3Object-affected harmful factors
If single microphone is used, then device complexity is low, but noise elimination capability is insufficient
Solution Approach 1:
The patent combines multiple signal sources (main channel microphone and noise reference microphone) into a unified processing framework. By merging these signals and processing them through a coordinated adaptive filtering system, the patent achieves superior noise elimination capability. The merging of multiple microphones allows the system to exploit spatial information and correlations between signals to effectively separate noise from voice, while the integrated processing approach manages the complexity efficiently.
Data Source
AI summary
A sound processing method includes: determining a vector of a first residual signal according to a first signal vector and a second signal vector, the first signal vector including a first voice signal and a first noise signal input into the first microphone, the second signal vector including a second voice signal and a second noise signal input into the second microphone, and the first residual signal including the second noise signal and a residual voice signal; determining a gain function of a current frame according to the vector of the first residual signal and the first signal vector; and determining a first voice signal of the current frame according to the first signal vector and the gain function of the current frame.