Adaptive Howl Detection and Frequency-Selective Suppression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional wearable devices with integrated microphones and speakers suffer from howling speech signals due to audio feedback loops, which degrade sound quality and user experience, and existing adaptive feedback cancellers fail to effectively suppress howl in dynamic environments.
Innovation Solution
An audio system employs adaptive notch filters to detect howl by monitoring signal flatness and tonality, using linear prediction to identify specific frequency ranges, and applies gain reduction to suppress howl while minimizing impact on other frequencies.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If adaptive feedback cancellers are used to mitigate howl, then howl suppression is achieved in steady state, but the system fails to react fast enough to path changes and system variations
Solution Approach 1:
The patent applies dynamics by transitioning from steady-state adaptive feedback cancellation to a dynamic detection and suppression system that continuously monitors for howl conditions and rapidly responds to path changes. The system dynamically adjusts its behavior based on real-time detection of feedback loops, enabling fast response to changing acoustic paths while maintaining effective howl suppression.
Solution Approach 2:
The patent enhances feedback mechanisms by implementing a detection system that monitors acoustic feedback paths and triggers suppression actions when howl conditions are detected. This feedback loop detects the presence of harmful feedback and activates targeted suppression, creating a responsive system that adapts to real-time acoustic conditions rather than relying solely on slow adaptive cancellation.
2Reliability
If gain is reduced for all frequencies to suppress howl, then howl is eliminated, but sound quality deteriorates
Solution Approach 1:
The patent applies local quality by implementing frequency-selective suppression that targets only the specific frequencies where howl occurs rather than reducing gain across all frequencies. The system identifies the problematic frequency components through detection and applies suppression locally to those frequencies while preserving the quality of other audio content, thereby eliminating howl without degrading overall sound quality.
Solution Approach 2:
The patent segments the audio spectrum into different frequency ranges and applies suppression selectively to the segments containing howl frequencies. By dividing the frequency domain and targeting specific segments for gain reduction, the system can eliminate howl while maintaining the integrity of other frequency components, avoiding the need to reduce all frequencies uniformly.
Data Source
AI summary
A system suppresses howl in a device including microphones and speakers, for example, an artificial reality headset. A speaker of the device presents audio content. The audio content presented by the speaker is received by a microphone of the device thereby creating a howl in certain situations. The system detects the presence of the howl in a region of the audio content using an adaptive notch filter. The system suppresses the howl by reducing gain of one or more frequencies of the audio content. The system may detect presence of the howl by monitoring flatness of the signal. The system may detect presence of the howl based on tonality detection based on linear prediction.