Acoustic feedback cancellation is an important task in audio processing systems, aiming to mitigate the effects of feedback loops on system stability and sound quality. State-of-the-art methods rely on adaptive filtering algorithms and face challenges in balancing between rapid convergence and low steady-state error. In this work, we introduce a novel approach inspired by traditional adaptive filtering and deep learning techniques to achieve a significantly faster convergence and lower steady-state errors at the same time. Our proposed system, termed Deep Feedback Cancellation (DFC), leverages deep neural networks to predict the impulse response of the feedback path directly. Hence, it replaces traditional gradient based adaptive estimation of the impulse responses. Experimental evaluation, in a hearing aid setting, conducted on real-world data demonstrates the superiority of DFC over traditional methods. Specifically, in a practically very important situation, where the feedback path undergoes rapid changes, the proposed DFC achieves increased convergence rate by a factor of 30, while decreasing the steady-state error by 2 dB. Generally, DFC leads to very significant improvements over traditional methods. These improvements are confirmed by objective evaluations and subjective listening tests. Our findings suggest that DFC presents a promising alternative for acoustic feedback cancellation in hearing aid applications.

State-of-the-art feedback control systems make use of adaptive filters to model and compensate for the acoustic feedback paths, which vary in different acoustic situations. In a hearing aid application, the acoustic feedback paths are user dependent and vary over time. Although there are repeatable patterns in these variations, it remains to be seen that modern feedback control systems actively learn from and then use these patterns to improve the adaptive filter performance. In this letter, we present a new idea of learning and storing acoustic feedback paths as hearing aid users wear their hearing aids, and then to apply these learned and stored acoustic feedback paths to control and support the adaptive filter updates, thereby to achieve significantly improved feedback control performance.

Adaptive filters have been widely used for feedback cancellation in audio systems including hearing aids. In addition to adaptive filters, the frequency shifting has often been used to obtain an unbiased estimation of the adaptive filters, by decorrelating the incoming and outgoing signals of the audio system. However, it has been shown in a recent study, that although the frequency shifting technique is effective in decorrelating the signals, its use in addition to the adaptive filter would introduce an additional and signal dependent periodically time-varying bias also referred to as the residual bias, especially for tonal signals such as music. In this work, we make use of that knowledge and propose a method to detect different acoustic situations, based on the level of residual bias. We further discuss on some control actions to be taken in acoustic feedback cancellation systems upon these detections. Finally, we demonstrate the behavior of the residual bias and the detection of different acoustic feedback situations in an example computer simulation.

Acoustic feedback is one of the major problems in hearing aid applications. During a fitting session of a modern hearing aid, typically a feedback path prediction or an in situ measurement of feedback path is used as part of the gain and earpiece prescription to minimize the risk of feedback problems. It is well known that there are a lot of variations in feedback paths across users due to ear differences, however, there is limited knowledge from published studies to actually quantify these variations, especially for the earpiece types as domes and micro molds which became popular since the introduction of receiver-in-the-ear (RITE) style hearing aids. In this empirical study, we measured feedback paths on different users wearing a RITE style hearing aid fitted with different domes and micro molds. Our results confirmed that there are large variations across users/ears, and magnitude differences in measured feedback paths can be more than 60 dB for otherwise identical hearing aid and earpiece type.

Compared to monaural hearing aids (HAs), binaural hearing aid systems, in which there is a communication link between the two devices, have improved noise reduction capabilities and the ability to preserve binaural spatial information. However, the limited HA battery lifetime puts constraints on the amount of information that can be shared between the two devices. In other words, the rate of transmission between the devices is an important constraint that needs to be considered, while preserving the spatial information. In this article, a linearly constrained noise reduction problem is proposed, which jointly finds the optimal rate allocation and the optimal estimation (beamforming) weights across all sensors and frequencies, while preserving the binaural spatial cues of point sources. The proposed method considers a rate constraint together with linear constraints to preserve the binaural spatial cues of point sources. Minimizing the mean square error on the estimated target speech at the left and the right side beamformers, the optimal weights are found to be rate-constrained linearly constrained minimum variance (LCMV) filters, and the optimal rates are found to be the solutions to a set of reverse water filling problems. The performance of the proposed method is evaluated using the averaged binaural signal-to-noise ratio (SNR), the interaural level difference (ILD) error and the interaural time difference (ITD) error. The results show that the proposed method outperforms spatially correct noise reduction approaches that use naive/random rate allocation strategies.

Wireless acoustic sensor networks (WASNs) can be used for centralized multi-microphone noise reduction, where the processing is done in a fusion center (FC). To perform the noise reduction, the data needs to be transmitted to the FC. Considering the limited battery life of the devices in a WASN, the total data rate at which the FC can communicate with the different network devices should be constrained. In this article, we propose a rate-constrained multi-microphone noise reduction algorithm, which jointly finds the best rate allocation and estimation weights for the microphones across all frequencies. The optimal linear estimators are found to be the quantized Wiener filters, and the rates are the solutions to a filter-dependent reverse water-filling problem. The performance of the proposed framework is evaluated using simulations in terms of mean square error and predicted speech intelligibility. The results show that the proposed method is very close in performance to that of the existing optimal method based on discrete optimization. However, the proposed approach can do this at a much lower complexity, while the existing optimal reference method needs a non-tractable exhaustive search to find the best rate allocation across microphones.

The short-time Fourier transform (STFT) and the inverse short-time Fourier transform are often used in signal analysis, modification, and synthesis. In this work, we focus on the effect of overlapping STFT subbands in relation to signal modification and synthesis. We illustrate that the subband overlapping can impose a negative mixing effect on STFT domain signal processing such as bandlimited frequency shifting. We propose a subband filtering method, by using subband low-pass and high-pass filters in the affected frequency region, to reduce the mixing effect locally without changing the general STFT processing. We show in simulation experiments that our subband filtering method is efficient; the bandwidth of the affected frequency region is typically reduced by a factor of 2-3, and there is significantly less signal distortion in the affected frequency region.

While the majority of binaural beamformers aim to minimize the output noise power while (approximately) preserving the binaural cues of the sources using constraints, we propose in this paper to minimize the binaural-cue distortions of the sources in the acoustic scene, such that the output noise power is below a predefined threshold. This new problem formulation is a convex QCQP problem, which leads to an efficient trade-off between noise reduction, binaural-cue preservation and complexity. In particular, the proposed beamformer provides a better trade-off between noise reduction and binaural-cue preservation (in terms of interaural level and phase differences) compared to the well-known binaural minimum variance distortionless response-η beamformer.

Binaural hearing aids (HAs) can potentially perform advanced noise reduction algorithms, leading to an improvement over monaural/bilateral HAs. Due to the limited transmission capacities between the HAs and given knowledge of the complete joint noisy signal statistics, the optimal rate-constrained beamforming strategy is known from the literature. However, as these joint statistics are unknown in practice, sub-optimal strategies have been presented. In this paper, we present a unified framework to study the performance of these existing optimal and sub-optimal rate-constrained beamforming methods for binaural HAs. Moreover, we propose to use an asymmetric sequential coding scheme to estimate the joint statistics between the microphones in the two HAs. We show that under certain assumptions, this leads to sub-optimal performance in one HA but allows to obtain the truly optimal performance in the second HA. Based on the mean square error distortion measure, we evaluate the performance improvement between monaural beamforming (no communication) and the proposed scheme, as well as the optimal and the existing sub-optimal strategies in terms of the information bit-rate. The results show that the proposed method outperforms existing practical approaches in most scenarios, especially at middle rates and high rates, without having the prior knowledge of the joint statistics.

Adaptive filters have been widely used for feedback cancellation in audio systems including hearing aids. However, the ability to cancel feedback in dynamic feedback situations is still a big challenge; the adaptive filters need to ensure a small enough steady-state error to facilitate sufficient amplification in hearing aids, hence their convergence rates are typically insufficient to handle fast feedback path changes. Recently, we proposed a novel method by using spectro-temporal modulation (STM) in the time-frequency regions where the adaptive filters have insufficient convergence rate. Applying STM prevents feedback to occur and replaces traditional loud/annoying feedback whistling sounds with soft/non-intrusive STM processed sounds. In this work, we introduce an extension to make the STM processed sound even less audible. Furthermore, we present novel evaluation results regarding feedback cancellation and sound quality from listening experiments, which confirm that without degrading sound quality in static feedback situations we significantly improve feedback cancellation performance upon fast feedback path changes.

Frequency shifting has been used for acoustic feedback control in audio reinforcement systems since the 1950s. It can be used as a stand-alone system and/or it can be combined with an acoustic feedback cancellation system using adaptive filters. A spectral shifting of the loudspeaker signal in an audio system breaks the acoustic feedback loop, and it decorrelates the reference and error signals to the adaptive filter in the cancellation system, which is useful for getting an unbiased adaptive filter estimation. In this work, we analyze the decorrelation effect from the frequency shifting in an acoustic feedback cancellation system. We show that the influence from the frequency shifting, on the correlation function between the reference and error signals, can be divided into two parts: a fast time-varying part and a slowly time-varying part. Especially the slowly time-varying part of the correlation function leads to a periodically time-varying bias in the adaptive filter estimation, which limits the feedback cancellation performance. We propose a solution to obtain an unbiased estimation by removing the slowly time-varying part in the adaptive filter estimation and we verify it by simulations.

Audio applications often suffer from the acoustic feedback problem which leads to significant sound quality degradation and howling in the worst case. Many solutions exist to minimize the effect of feedback. However, it is not trivial to evaluate these solutions. In this work, we focus on intrusive howling detection methods by comparing a test signal to a known reference signal without howling. Traditional howling detection methods are less reliable when acoustic feedback control systems make use of some decorrelation techniques, such as frequency shifting and/or probe noise injection. In this paper, we propose two new simple detection methods which are robust against these processing strategies.

This work focuses on an adaptive beamformer in a hearing aid application using a generalized sidelobe canceller structure (GSC). In this application, the constraint and blocking matrices in the GSC structure are specifically designed using an estimate of the transfer functions between the target source and the microphones to ensure optimal beamformer performance. We show that, in practice, the GSC always-unintentionally-attenuates the target sound in a special but realistic situation where all signals, including the target and noise signals, originate from the look direction reflected by the look vector. This happens because, in practice, the blocking matrix in the GSC structure is non-ideal. We introduce a simple modification to the GSC structure, which solves the problem of undesired target signal attenuation in situations where all signals originate from the look direction. Furthermore, this modification can also prevent desired signals, originating from positions spatially close to the look direction, to be removed. We also show that the solution has no impact on other acoustic situations.

In a previous study, the performance of an acoustic feedback/echo cancellation system was analyzed using a power transfer function method. Whereas the analysis result provides very accurate performance predictions in open-loop acoustic echo cancellation systems, it is less accurate in closed-loop acoustic feedback cancellation systems if there is a strong correlation between the loudspeaker signal and the signals entering the microphones. This work extends the performance analysis to include the effects of the nonzero correlation on the adaptive filters. Simulation results verify that this extension provides much more accurate performance predictions in closed-loop acoustic feedback cancellation systems.

A probe noise signal can be used in an acoustic feedback cancellation system to prevent biased adaptive estimation of acoustic feedback paths. However, practical experiences and simulation results indicate that whenever a low-level and inaudible probe noise signal is used, the convergence rate of the adaptive estimation is significantly decreased when keeping the steady-state error unchanged. The goal of this work is to derive analytic expressions for the system behavior such as convergence rate and steady-state error for a multiple-microphone and single-loudspeaker audio system, where the acoustic feedback cancellation is carried out using a probe noise signal. The derived results show how different system parameters and signal properties affect the cancellation performance, and the results explain theoretically the decreased convergence rate. Understanding this is important for making further improvements in the existing probe noise approach.

Adaptive filters are widely used in acoustic feedback cancellation systems and have evolved to be state-of-the-art. One major challenge remaining is that the adaptive filter estimates are biased due to the nonzero correlation between the loudspeaker signals and the signals entering the audio system. In many cases, this bias problem causes the cancellation system to fail. The traditional probe noise approach, where a noise signal is added to the loudspeaker signal can, in theory, prevent the bias. However, in practice, the probe noise level must often be so high that the noise is clearly audible and annoying; this makes the traditional probe noise approach less useful in practical applications. In this work, we explain theoretically the decreased convergence rate when using low-level probe noise in the traditional approach, before we propose and study analytically two new probe noise approaches utilizing a combination of specifically designed probe noise signals and probe noise enhancement. Despite using low-level and inaudible probe noise signals, both approaches significantly improve the convergence behavior of the cancellation system compared to the traditional probe noise approach. This makes the proposed approaches much more attractive in practical applications. We demonstrate this through a simulation experiment with audio signals in a hearing aid acoustic feedback cancellation system, where the convergence rate is improved by as much as a factor of 10.

In this paper, we analyze a general multiple-microphone and single-loudspeaker system, where an adaptive algorithm is used to cancel acoustic feedback/echo and a beamformer processes the feedback/echo canceled signals. This system can be viewed as part of a typical hearing aid system and/or a traditional acoustic echo cancellation system. We introduce and derive an approximation of a useful frequency domain measure - the power transfer function - and show how to predict the system stability bound, convergence rate and the steady-state behavior across time and frequency. Furthermore, we show how the derived expressions can be used to determine e.g. the step size parameter in the adaptive algorithms to achieve a desired system property e.g. convergence rate at a specific frequency.