Single-Channel Noise Suppression Using Gaussian Mixture Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional noise suppression techniques fail to effectively differentiate and suppress various types of interfering sources in audio signals, such as stationary noise, non-stationary noise, and residual echo, leading to suboptimal noise-suppressed audio signals in communication devices.
Innovation Solution
The implementation of a back-end single-channel suppression system using statistical modeling, specifically Gaussian mixture models, to determine and apply noise suppression gains based on features associated with different types of interfering sources, enabling effective suppression of multiple types of noise in audio signals.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional noise suppression techniques are used, then the processing is simple, but the ability to differentiate and suppress various types of interfering sources is poor
Solution Approach 1:
The patent segments the noise suppression task by creating separate processing branches for different types of interfering sources (stationary noise, non-stationary noise, residual echo). Each branch is specialized to handle specific noise characteristics, enabling precise differentiation and suppression of various noise types while maintaining manageable system complexity through modular architecture.
Solution Approach 2:
The patent applies local quality by using different noise suppression algorithms and parameters tailored to specific types of interfering sources. Stationary noise suppression uses different techniques than non-stationary noise suppression, and each branch is optimized with locally appropriate processing characteristics to achieve superior differentiation precision for each noise category.
2Reliability
If conventional noise suppression techniques are used, then the system is simple to operate, but the quality of noise-suppressed audio signal is suboptimal
Solution Approach 1:
The patent implements dynamic noise suppression by continuously adapting the suppression gains based on real-time analysis of audio signal characteristics. The system dynamically switches between different suppression strategies and adjusts parameters according to the current noise environment, ensuring high reliability and optimal audio quality across varying conditions while managing complexity through adaptive control mechanisms.
Solution Approach 2:
The patent employs feedback mechanisms where the output of each noise suppression branch is analyzed and used to adjust the suppression gains and parameters. This closed-loop control ensures high reliability by continuously optimizing the noise suppression performance based on actual signal quality metrics, while the feedback-driven adaptation prevents the system from becoming overly complex through rule-based control.
3Device complexity
If single-channel suppression is implemented, then the device complexity is reduced, but the ability to suppress multiple types of noise effectively is limited
Solution Approach 1:
The patent achieves universality by designing a multi-functional single-channel suppression system that can handle multiple types of interfering sources (stationary noise, non-stationary noise, residual echo) within a unified architecture. The system uses a universal framework with specialized branches that can be selectively activated, providing versatile noise suppression capability while maintaining single-channel simplicity and manageable device complexity.
Data Source
AI summary
Techniques described herein are directed to performing back-end single-channel suppression of one or more types of interfering sources (e.g., additive noise) in an uplink path of a communication device. The back-end single-channel suppression techniques may suppress types(s) of additive noise using one or more suppression branches (e.g., a non-spatial (or stationary noise) branch, a spatial (or non-stationary noise) branch, a residual echo suppression branch, etc.). The non-spatial branch may be configured to suppress stationary noise from the single-channel audio signal, the spatial branch may be configured to suppress non-stationary noise from the single-channel audio signal and the residual echo suppression branch may be configured to suppress residual echo from the signal-channel audio signal. The spatial branch may be disabled based on an operational mode (e.g., single-user speakerphone mode or a conference speakerphone mode) of the communication device or based on a determination that spatial information is ambiguous.