Adaptive SNR Estimator for Multi-band Audio Noise Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional noise reduction systems face challenges in adapting to varying noise environments and system parameters, leading to suboptimal sound quality and intelligibility, particularly due to the interaction between speech and noise estimators and gain functions, which can introduce biases and require complex parameter tuning.
Innovation Solution
A multi-band noise reduction system that employs an adaptive signal-to-noise ratio estimator with time-varying low-pass filters, where the low-pass cut-off frequency is adjusted based on signal-to-noise ratio estimates, allowing for effective suppression of noise fluctuations and rapid response to changes in signal conditions, and utilizes monotonic functions for compressive and expansive mapping to optimize gain calculations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional noise reduction systems use fixed parameter estimation, then the system structure is simple, but the sound quality and intelligibility are suboptimal in varying noise environments
Solution Approach 1:
The patent implements dynamic adaptation by making the low-pass filter cut-off frequency time-varying and signal-dependent. The cut-off frequency automatically adjusts based on the estimated signal-to-noise ratio, allowing the system to adapt to changing noise environments without manual intervention or complex parameter tuning.
Solution Approach 2:
The patent changes the parameter of the low-pass filter (cut-off frequency) from a fixed value to a dynamically adjustable parameter. This parameter change enables the system to optimize its performance for different noise conditions while maintaining a relatively simple overall structure.
2Manufacturing precision
If time-varying low-pass filters with adaptable cut-off frequency are used, then sound quality and intelligibility are enhanced, but computational resources increase
Solution Approach 1:
The patent uses parameter changes to optimize the balance between sound quality and computational resources. By dynamically adjusting the low-pass filter cut-off frequency based on signal-to-noise ratio estimates, the system achieves high sound quality only when necessary, rather than using fixed high-computation parameters throughout.
Solution Approach 2:
The patent applies local quality by making the filtering characteristics specific to each frequency band and time-dependent on the local signal-to-noise ratio conditions. This allows computational resources to be focused on frequency bands and time periods where noise reduction is most needed, rather than uniformly processing all signals at high computational cost.
3Productivity
If monotonic functions are used for compressive and expansive mapping, then gain calculations are optimized, but the system complexity increases
Solution Approach 1:
The patent introduces monotonic compressive and expansive functions as intermediary transformations in the signal processing chain. These functions act as mediators that optimize the mapping between different signal representations, improving processing efficiency while maintaining a clear and manageable system architecture.
Data Source
Figure 1
Figure 2
Figure 3A~3B
AI summary
The present invention relates to a multi-band noise reduction system for digital audio signals producing a noise reduced digital audio output signal from a digital audio signal. The digital audio signal comprises a target signal and a noise signal, i.e. a noisy digital audio signal. The multi-band noise reduction system operates on a plurality of sub-band signals derived from the digital audio signal and comprises a second or adaptive signal-to-noise ratio estimator which is configured for filtering a plurality of first signal-to-noise ratio estimates of the plurality of sub-band signals with respective time-varying low-pass filters to produce respective second signal-to- noise ratio estimates of the plurality of sub-band signals. A low-pass cut-off frequency of each of the time-varying low-pass filters is adaptable in accordance with a first signal-to-noise ratio estimate determined by a first signal-to-noise ratio estimator and/or the second signal-to-noise ratio estimate of the sub-band signal.