Multi-Microphone Howling Suppression for Low-Delay Audio Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional techniques for suppressing howling in loudspeaker systems are delayed, leading to uncomfortable sounds and ineffective howling prevention, as they require long observation periods to detect howling, which is against the requirement of low delay processing.

Innovation Solution

A howling suppression apparatus and method that utilize integration processing to obtain maximum or minimum values of frequency-domain signals from multiple microphones, determining howling occurrence based on signal differences, and performing suppression processing to prevent delayed detection and improve howling prevention effects.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional howling detection methods using thresholding on power for each frequency are used, then howling can be detected, but detection delay occurs requiring long observation periods

Engineering Contradiction:
Improvehowling detection accuracyVSAvoiddetection delay
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent applies preliminary action by performing integration processing on frequency-domain signals from multiple microphones to obtain maximum or minimum values before howling detection. This preprocessing step enables the system to detect howling occurrence based on signal differences without requiring long observation periods, thus reducing detection delay while maintaining accuracy

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent transitions from single-microphone analysis to multi-microphone integration processing, adding a spatial dimension to the detection system. By obtaining maximum or minimum values across multiple frequency-domain signals and comparing signal differences, the system achieves rapid howling detection without the temporal delay inherent in conventional single-microphone thresholding methods

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Ease of operation

If low-delay processing is implemented to synchronize amplified voice with direct sound, then auditory naturalness is improved, but howling detection becomes more difficult

Engineering Contradiction:
Improveauditory naturalnessVSAvoidhowling detection difficulty
Core Design Contradiction:
Ease of operationVSDifficulty of detecting and measuring

Solution Approach 1:

The patent performs integration processing and obtains maximum or minimum values of frequency-domain signals before howling detection, enabling rapid identification of howling occurrence. This preliminary processing allows the system to implement low-delay howling suppression while maintaining auditory naturalness by synchronizing amplified voice with direct sound

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent segments the audio processing into distinct stages: frequency-domain signal acquisition from multiple microphones, integration processing to obtain maximum or minimum values, howling occurrence determination based on signal differences, and howling suppression. This segmentation enables low-delay processing by identifying and suppressing howling components rapidly without affecting the synchronization of amplified voice with direct sound

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP3755005B1Howling suppression device, method therefor, and program
Publication Date: 2023.06.07 NIPPON TELEGRAPH & TELEPHONE CORP
  • EP3755005B1 patent drawingFigure 1
  • EP3755005B1 patent drawingFigure 2
  • EP3755005B1 patent drawingFigure 3

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.