Frequency-Domain Howling Suppression Using Multi-Microphone Maxima
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional loudspeaker systems experience delays in howling detection, leading to uncomfortable sounds due to the beating phenomenon, which conventional techniques address only after howling occurs, violating the requirement for low-delay processing.
Innovation Solution
A howling suppression apparatus that integrates maximum or minimum values of frequency-domain signals from multiple microphones to determine howling occurrence and perform suppression processing, preventing delays in detection and improving howling prevention effects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional howling detection methods are used (thresholding on power for each frequency), then howling can be detected, but detection is delayed and requires long observation periods
Solution Approach 1:
The patent transitions from analyzing single-channel frequency power to analyzing the phase relationships between multiple microphones in the frequency domain. By examining phase differences across multiple spatial dimensions (multiple microphones), the system can detect howling components more rapidly without requiring long temporal observation periods, thus resolving the contradiction between detection accuracy and detection speed.
Solution Approach 2:
The patent performs preliminary integration of frequency-domain signals from multiple microphones to obtain integrated values that highlight howling components. This preliminary processing prepares the data in advance, enabling faster and more accurate howling detection without waiting for extended observation periods, thereby reducing detection delay while maintaining precision.
2Reliability
If amplified voice is reproduced with delay, then howling prevention is easier, but the amplified voice sounds unsynchronized with direct sound creating auditory unnaturalness
Solution Approach 1:
The patent implements a feedback mechanism where the system continuously monitors the acoustic environment using multiple microphones, detects howling components in real-time through frequency-domain analysis and integration, and applies suppression processing to the amplified signal. This closed-loop feedback enables effective howling prevention without requiring excessive delay, maintaining synchronization with direct sound and auditory naturalness.
Solution Approach 2:
The patent dynamically adjusts the howling suppression parameters based on real-time analysis of the acoustic environment. By changing the suppression gain and frequency targeting parameters adaptively, the system achieves effective howling prevention while minimizing impact on the natural sound quality and maintaining low latency, thus resolving the contradiction between prevention effectiveness and auditory naturalness.
Data Source
AI summary
A howling suppression apparatus includes: an integration processing part that obtains the maximum value among L values corresponding to n-th frames of L i-th signals, for i=1, 2, . . . , L, L being any integer equal to or greater than 2, the L i-th signals being frequency-domain signals obtained from sound signals collected by multiple microphones; and a howling suppression processing part that performs howling suppression processing on at least any of the L i-th signals using the maximum value.