Audio Signal Processing Filter Cascade for Noise Peak Suppression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional audio recording methods often result in mediocre or bad audio quality due to artificial noise at certain audible frequencies, which complicates post-processing and requires substantial efforts to improve audio quality.
Innovation Solution
A computer-implemented method using a simpler geometrically aided model to generate a cascade of filter elements that suppresses identified artificial noise peaks, with adaptive center frequencies adjusted to the actual sound signal, reducing the need for pre-defined profiles and minimizing computational effort.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional denoising processes with pre-defined profiles are used, then noise suppression is achieved, but the method is inflexible and requires many pre-specified profiles for different environments
Solution Approach 1:
The system automatically identifies noise peaks and generates appropriate filter parameters without requiring manual profile selection or pre-specified configurations. The algorithm processes the audio signal itself to determine the necessary denoising parameters, making the system self-adapting to different recording environments.
Solution Approach 2:
The filter parameters (center frequencies, Q factors, gains) are dynamically adjusted based on the actual noise peaks detected in the input signal rather than using fixed pre-defined profiles. This allows the system to adapt to varying recording conditions in real-time.
2Reliability
If iterative approaches or deep learning networks are used to suppress artificial noise, then noise suppression effectiveness is improved, but computational cost increases significantly
Solution Approach 1:
The invention extracts only the essential noise peaks from the audio signal spectrum and focuses computational resources on suppressing these specific peaks using simple filter elements, rather than applying complex iterative processing or deep learning networks to the entire signal.
Solution Approach 2:
The system uses computationally inexpensive filter elements with simple transfer functions that can be quickly applied and discarded, replacing them as needed based on detected noise peaks, rather than relying on expensive deep learning models.
3Manufacturing precision
If manual denoising processing is performed with multiple parameters, then desired quality is achieved, but substantial post-processing effort is required
Solution Approach 1:
The system performs preliminary automatic identification of noise peaks and calculation of optimal filter parameters before applying the denoising filters, eliminating the need for manual parameter adjustment and iterative processing by the user.
Solution Approach 2:
The system uses the smoothed spectrum of the input signal as feedback to automatically determine the center frequencies and Q factors of the filter elements, creating a closed-loop system that adapts to the specific characteristics of each recording without manual intervention.
Data Source
AI summary
The present invention concerns a computer implemented method for processing an audio signal for which an audio signal containing speech is obtained. Then a smoothed difference signal is derived from the obtained audio signal and smoothed spectrum thereof. At least one peak is detected in the obtained difference signal and a peaking filter generated based on the at least one detected peak. The peaking filter comprises at least a Q factor and a gain parameter, whereas the Q factor is based on the geometrically obtained cut-off frequencies around the at least one detected peak; and the gain parameter is derived from a gain interaction matrix with elements based on magnitude responses at cut-off frequencies around the at least one detected peak.