Direct Sound Extraction via Amplitude Spectrum Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for extracting direct and reverberant sounds from acoustic signals require significant processing load and high-powered devices, making them costly and impractical for widespread use.
Innovation Solution
A direct sound extraction device that performs Fourier transforms, low-pass filtering, and amplitude limiting to isolate the direct sound from reverberant sound, using a preset normalized cutoff frequency to adjust extraction time and include stationary components, thereby simplifying the extraction process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional techniques (pseudo-whitening, multi-step linear prediction, rear reverberation prediction) are used to reduce reverberant sound, then the reverberant sound can be reduced and direct sound clarity improved, but the processing load becomes extremely high requiring high-powered devices
Solution Approach 1:
The patent extracts only the essential characteristic needed for reverberation reduction - the low-frequency components of the amplitude spectrum - and processes only this extracted information. By taking out the relevant low-frequency part and discarding high-frequency components, the system achieves reverberation reduction without requiring complex full-spectrum processing, thus reducing processing load while maintaining effectiveness
Solution Approach 2:
The patent replaces expensive, complex processing algorithms with a simple, computationally inexpensive low-pass filtering approach. The low-pass filter unit provides a cheap computational alternative to sophisticated algorithms like pseudo-whitening and multi-step linear prediction, achieving acceptable reverberation reduction with minimal processing resources
2Measurement precision
If sophisticated signal processing algorithms are implemented to extract direct sound, then extraction accuracy improves, but the computational resources and device cost increase significantly
Solution Approach 1:
The patent replaces complex mechanical/computational signal processing systems with a simpler spectral filtering approach. Instead of using computationally intensive time-domain algorithms, the system transforms to frequency domain and applies simple low-pass filtering to the amplitude spectrum, substituting complex mechanical processing with a more efficient spectral approach
Solution Approach 2:
The patent changes the processing parameter from time-domain signal manipulation to frequency-domain amplitude spectrum manipulation. By applying low-pass filtering in the frequency domain rather than complex time-domain algorithms, the system achieves direct sound extraction with reduced computational power requirements while maintaining extraction accuracy
Data Source
AI summary
A direct sound extraction device includes: a spectrum transform unit that transforms an input signal, which includes a reverberant sound in a direct sound and on which a Fourier transform process has been performed, to a first amplitude spectrum signal Lfa; a low-pass filter unit (4) that performs a low-pass filtering process on the first amplitude spectrum signal Lfa for each frequency to generate a second amplitude spectrum signal Lfa1; a first subtraction unit (18) that calculates a third amplitude spectrum signal by subtracting the second amplitude spectrum signal Lfa1 from the first amplitude spectrum signal Lfa; and an inverse Fourier transform unit that generates a direct sound signal Lfd from a frequency spectrum signal calculated based on a phase spectrum signal and the third amplitude spectrum signal.