Frequency-Band Noise Gate for Stationary Noise Removal
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing noise gate devices remove audio signals with signal levels below the threshold, including utterance voices, leading to degradation of audio quality.
Innovation Solution
A noise gate that estimates a noise spectrum based on the input audio signal's frequency spectrum, computes a signal level ratio, and adjusts the signal level in frequency bands below a predetermined threshold, while maintaining levels above the threshold to preserve utterance voices, and considers estimation errors by setting the threshold value greater than 1.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If a noise gate removes audio signals below a threshold value, then stationary noise is removed, but utterance voices with low signal levels are also removed degrading audio quality
Solution Approach 1:
The patent segments the audio signal processing by separating stationary noise removal from transient noise removal. It divides the frequency spectrum into multiple bands and applies different processing strategies to each band based on the signal level ratio, allowing selective noise removal without affecting voice quality.
Solution Approach 2:
The patent applies local quality by treating different frequency bands differently. It computes signal level ratios for each frequency band and applies noise removal only where appropriate (when ratio < threshold) while preserving bands containing voice signals. The threshold value > 1 provides local adaptation to estimation errors in specific frequency regions.
2Device complexity
If a noise gate uses a fixed threshold value, then simple noise removal is achieved, but utterance voices with signal levels near the threshold are degraded
Solution Approach 1:
The patent introduces dynamics by making the threshold adaptive rather than fixed. The threshold value is set greater than 1 to dynamically compensate for estimation errors in the stationary noise spectrum. This dynamic adjustment allows the system to adapt to varying noise conditions while preserving voice signals that would otherwise be incorrectly identified as noise.
3Measurement precision
If the threshold value is set to 1 for accurate noise removal, then stationary noise is effectively removed, but estimation errors cause voice signals to be incorrectly removed
Solution Approach 1:
The patent applies beforehand cushioning by setting the threshold value greater than 1 to compensate for potential estimation errors before they cause problems. This pre-adjusted threshold creates a safety margin that prevents voice signals from being incorrectly removed due to noise spectrum estimation inaccuracies, while still maintaining effective noise removal capability.
Data Source
Figure 1~2
Figure 3(A)~3(C)
Figure 4
AI summary
A purpose of the invention is to provide a noise gate that can output an audio signal in which only a stationary noise is removed, without degrading an utterance voice of a speaking person. A sound collection device 1 includes an FFT processing unit 11, the noise gate 12, and an IFFT processing unit 13. The sound collection device 1 transforms a collected audio signal NE'T into a frequency spectrum NE'N by using the FFT processing unit 11. The noise gate 12 estimates a noise spectrum N'N of a stationary noise based on the frequency spectrum NE'N of the audio signal. The noise gate 12 decreases a signal level (a gain) of the audio signal in a case where a signal level ratio of the frequency spectrum NE'N of the audio signal to the noise spectrum N'N is less than a threshold value, and outputs the audio signal. The sound collection device 1 outputs an audio signal CO'T which is generated in such a manner that the IFFT processing unit 13 inversely transforms a frequency spectrum CO'N after removing the stationary noise N'N.