Hearing Aid SNR Estimation via Adaptive Non-Linear Smoothing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current audio processing devices, such as hearing aids, face challenges in accurately estimating the signal-to-noise ratio (SNR) in noisy environments, particularly in time-frequency domains, leading to suboptimal noise reduction and speech intelligibility.
Innovation Solution
The implementation of a non-linear smoothing method using a recursive algorithm with adaptive low-pass filtering to estimate the a priori SNR from a posteriori SNR estimates, incorporating bias and smoothing parameters optimized through supervised learning, and adaptive time constants to improve noise reduction in hearing aids.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If non-linear smoothing with adaptive low-pass filtering is applied to estimate a priori SNR from a posteriori SNR estimates, then noise reduction capability is improved, but device complexity increases
Solution Approach 1:
The patent implements adaptive low-pass filtering where the filtering parameters (cut-off frequency, time constant) are dynamically adjusted based on the estimated SNR conditions. The filter adapts its characteristics in real-time to match the acoustic environment, providing optimal noise reduction for different SNR scenarios while managing computational complexity through selective adaptation rather than fixed complex processing
Solution Approach 2:
The patent changes the parameters of the low-pass filter (cut-off frequency, time constant, smoothing factor) based on the estimated a posteriori SNR values. By adjusting these parameters dynamically, the system achieves improved noise reduction performance without requiring a completely complex processing architecture, as it leverages parameter modulation of a relatively simple filter structure
2Measurement precision
If adaptive time constants are used in recursive algorithms for SNR estimation, then speech intelligibility is improved, but computational requirements increase
Solution Approach 1:
The patent employs adaptive time constants in the recursive SNR estimation algorithm, where the time constant is adjusted based on the estimated SNR conditions. This allows the system to achieve higher measurement precision in challenging acoustic environments by increasing the time constant for better smoothing, while in favorable conditions using smaller time constants reduces computational load and energy consumption
3Reliability
If non-linear smoothing is applied to reduce musical noise, then audio quality is improved, but processing time increases
Solution Approach 1:
The patent applies non-linear smoothing in a periodic manner within the recursive filtering framework, where the smoothing operation is performed at each time frame but with adaptive intensity. This allows the system to reduce musical noise artifacts effectively while managing processing time by adjusting the smoothing strength based on current SNR conditions rather than applying maximum smoothing continuously
Data Source
AI summary
A hearing aid includes a) at least one input unit for providing a time-frequency representation Y(k,n) of an electric input signal representing sound consisting of target speech and noise signal components, where k and n are frequency band and time frame indices, respectively, b) a noise reduction system configured to b1) determine an a posteriori signal to noise ratio estimate γ(k,n) of the electric input signal, and to b2) determine an a priori signal to noise signal ratio estimate ζ(k,n) of the electric input signal from the a posteriori signal to noise ratio estimate γ(k,n) based on a recursive algorithm providing non-linear smoothing. The a posteriori signal to noise ratio estimate of said electric input signal is provided as a mixture of first and second different a posteriori signal to noise ratio estimates. The invention may be used in audio processing devices, such as hearing aids, headsets, etc.