Adaptive Noise Suppression Gain for Speech Quality Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing noise suppressors in communication systems often apply excessive or insufficient noise reduction, leading to artifacts like musical noise, and struggle to maintain optimal signal quality due to fixed gain settings.
Innovation Solution
A noise suppression system that dynamically adjusts the minimum overall gain based on time-varying signal-to-noise ratio improvements, using a post-filtering analyzer and minimum gain adapter to calculate and control the gain on a frame-by-frame basis, ensuring effective noise reduction while preserving voice quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If fixed gain settings are used in noise suppressors, then device complexity is reduced, but noise suppression performance deteriorates due to inability to adapt to varying signal conditions
Solution Approach 1:
The patent implements dynamic gain adjustment by continuously adapting the gain parameter based on real-time signal-to-noise ratio measurements and voice activity detection. The system transitions from fixed gain settings to time-varying gain control, allowing the noise suppressor to respond to changing acoustic conditions while maintaining manageable complexity through algorithmic adaptation.
Solution Approach 2:
The patent changes the gain parameter dynamically based on measured signal characteristics. By monitoring the signal-to-noise ratio and voice activity status, the system adjusts the gain parameter to optimize noise suppression performance for different operating conditions, resolving the contradiction between adaptability and complexity.
2Measurement precision
If high-level noise suppression is applied, then signal-to-noise ratio is improved, but speech quality deteriorates due to excessive attenuation and distortion
Solution Approach 1:
The patent applies different gain levels to different time segments based on local signal characteristics. By detecting voice activity and signal-to-noise ratio in specific time windows, the system applies aggressive suppression only when noise dominates and minimal suppression when speech is present, preserving speech quality while improving overall signal-to-noise ratio.
Solution Approach 2:
The patent uses partial action by applying noise suppression selectively rather than continuously. The system applies strong suppression only when noise is detected and speech is absent, while using gentle or no suppression when speech is present, thus achieving partial suppression that optimizes the trade-off between signal-to-noise ratio improvement and speech quality preservation.
3Object-affected harmful factors
If spectral weighting is used for noise reduction, then noise suppression is achieved, but artifacts such as musical noise are introduced in the output signal
Solution Approach 1:
The patent implements periodic re-evaluation of signal characteristics through frame-by-frame analysis. By continuously measuring signal-to-noise ratio and voice activity in successive time frames, the system periodically updates its suppression strategy, preventing the accumulation of artifacts and allowing recovery of natural signal characteristics between suppression events.
Data Source
AI summary
Methods and corresponding systems for suppressing noise in an input signal include setting a minimum overall gain in a noise reduction processor for processing a first frame of data associated with the input signal. In response to a new minimum overall gain being set, the minimum overall gain in the noise reduction processor is replaced with the new minimum overall gain, and a second frame of data associated with the input signal is processed to suppress noise using the new minimum overall gain. The new minimum overall gain can be a function of the input signal or an output signal of the noise reduction processor. The new minimum overall gain can correspond to a difference between an estimated signal-to-noise ratio (SNR) improvement that is calculated using time-domain data and a target SNR improvement.