双通道回声抵消系统中改进算法的定点化实现

本文研究了双通道回声抵消系统的三个核心模块:频域多延时自适应滤波器算法(MDF),双端说话检测算法(DTD),以及残留回声抑制算法(RES)。针对频域算法的特点提出了改进的双端说话检测算法和基于维纳滤波与谱减法的残留回声抑制算法,然后以上述三种算法为核心模块实现双通道回声抵消系统,并对系统定点化以便在定点DSP处理器上实时实现,分析并解决了定点化的精度问题所带来的影响。     

应用于助听器反馈抑制的信噪比自适应滤波算法

针对基于自适应滤波器的助听器反馈抑制系统,该文提出了一种基于信噪比的归一化最小均方误差算法,采用最小值统计法估计误差信号的噪声分量,从而计算出误差信号的信噪比来计算自适应滤波系数的更新步长。当误差信号信噪比越高,语声占主要成分,信号的相关性越强,此时将滤波器的更新步长控制在较小值,减小滤波器的失调量;当信噪比越低时,噪声占主要成分,信号的相关性相对较弱,更新步长取较大值,加快滤波器的收敛速度。在仿真实验中,该文提出的基于信噪比的归一化最小均方误差算法相较于传统算法在平均稳态失调量和稳态失调范围上分别低1 dB和2 dB,其最大稳态增益提高了4 dB,同时具有更快的稳态收敛速度,验证了该文提出算法的有效性。     

一种改进的快速递推干扰抵消多用户检测算法

由于传统的MIMO-OFDM系统干扰消除多用户检测算法对错误传播和干扰消除顺序考虑的较少,针对这一缺点,提出一种采用逆向排列顺序的快速递推干扰消除多用户检测算法。该算法采用逆向检测顺序,选择用户数据相关程度最大的数据进行最后检测,得到信噪比最优的串行干扰消除用户排列,避免了错误传播;采用递推的方法实现最小均方误差(MMSE)检测器与排序的求取,避免了传统检测算法中的直接矩阵求逆运算,降低了算法的计算量。理论分析和实验结果表明,该算法在有效降低复杂度的同时,具有更好的系统容量性能和误码性能。     

数字助听器中广义旁瓣抵消器结构的汉语语音处理技术

自适应波束形成算法是数字助听器的核心算法之一。针对自适应波束形成算法中不可避免存在的语音泄漏,本文先对传统GSC结构自适应波束形成算法进行理论研究,并提出一种汉语处理技术,补偿泄漏的语音。这种汉语处理技术利用汉语语音特有的基音频率信息,调整语音幅度谱包络,提高谱包络与基频曲线形状的相似度以提高语音的可懂度。针对泄漏的语音在高频清辅音段有较大损失的特点,在频域上对清辅音进行放大,在不改变共振峰结构的情况下,提高清辅音的能量,同时降低语音间隔段GSC算法泄漏的噪声能量,提高对语音的辨别。仿真实验结果表明,这种汉语语音处理能够补偿自适应波束形成算法造成的语音泄漏,提高语音的可懂度。     

仿人耳听觉的助听器双耳声源定位算法

为提高复杂场景下的听障患者的语言理解度,本文提出一种仿人耳听觉的助听器双耳声源定位算法。算法首先借鉴耳蜗分频特性和听觉掩蔽特性,将声音信号进行多通道分解,并提取人耳敏感频带的信号进行双耳时间差(Interaural Time Difference,ITD)估计;然后基于人耳哈斯效应,提取有效的ITD信息;最后采用头相关模型,将ITD转化为声源方向信息。同时,为了改善混响和多干扰声场景下的声源定位能力,本文提出一种多通道的加权联合策略。仿真和场景测试实验表明,算法的抗干扰性强,定位精度高。而且,在7名受试者的理解度测试中,同现有的助听器增强算法相比,结合定位算法的语音增强算法达到3~5 dB的性能改善。     

仿人耳听觉的助听器双耳声源定位算法

为提高复杂场景下的听障患者的语言理解度,本文提出一种仿人耳听觉的助听器双耳声源定位算法。算法首先借鉴耳蜗分频特性和听觉掩蔽特性,将声音信号进行多通道分解,并提取人耳敏感频带的信号进行双耳时间差(Interaural Time Difference,ITD)估计;然后基于人耳哈斯效应,提取有效的ITD信息;最后采用头相关模型,将ITD转化为声源方向信息。同时,为了改善混响和多干扰声场景下的声源定位能力,本文提出一种多通道的加权联合策略。仿真和场景测试实验表明,算法的抗干扰性强,定位精度高。而且,在7名受试者的理解度测试中,同现有的助听器增强算法相比,结合定位算法的语音增强算法达到3~5dB的性能改善。     

噪声抵消法估计和抑制声呐部位主要自噪声

本文采用修正格型过滤自适应噪声抵消方法，成功地抑制艇艄声呐部位自噪声中的螺旋桨噪声成分等，抑制效果达４０ｄＢ以上。所采用的经改进后的ＬＳＬ算法收敛速度快，精度高，具有宽带抑噪性能。利用上述噪声抵消系统（软件）分析、判别了各噪声源对艇艏声呐部位声场的贡献，在确定主要噪声源方面，获得比较满意的结果。     

球谐域自适应混响抵消与声源定位算法

提出了一种基于球谐域的自适应混响抵消与声源定位算法,该方法通过去混响处理改善语音质量,并提高球谐域定位算法在混响环境下的定位性能。推导了基于多通道线性预测的自适应混响抵消算法在球谐域的表达式,针对刚球模型提出分阶处理的去混响方法,并对去混响后的信号进行波达方向估计。采用32元球阵的仿真结果表明,相比于球谐域不分阶去混响方法,该方法最大可减少约2/3的运算量,同时语音PESQ得分及SRMR均显著提高。利用实验数据对算法性能进行测试,实验结果验证了该方法在实际声学环境中去混响和声源定位的有效性。     

基于模式搜索的自适应干扰抵消器算法的研究

为了减小同址干扰对接收机性能的影响,设计了一种基于正交矢量合成的自适应干扰抵消器;根据其中控制单元提取出的数据的特点,提出了将模式搜索算法(PSA)作为控制器算法,并对其进行了改进;利用实际测量的数据进行了仿真分析,结果表明,相比于PSA算法、模拟退火算法、遗传算法,改进后的PSA算法具有更快的收敛速度,同时收敛精度相差无几;最终的实现结果也证明了该方法可满足性能要求。     

门控卷积神经网络多通路声学回声消除算法*

多通路声重放系统能够增强听者的现实感与空间感,但在免提通信条件下,其不可避免会受到噪声和回声干扰,严重影响通信质量。针对上述问题,本文提出了一种基于门控卷积循环神经网络的多通路声学回声消除和噪声抑制方法。该方法以传声器接收信号和重放声道的压缩复数谱为网络输入,以近端语音的压缩复数谱为网络的输出目标,直接从传声器拾取信号中恢复近端纯净语音,无需对声重放信号进行去相关处理,解决了传统自适应滤波方法中存在的非唯一解问题,同时保证了多通路声重放质量。仿真和真实声学环境实验均表明本文所提出的方法可显著消除多通路声重放系统的噪声和回声,在语音质量和回声返回衰减增益方面均优于传统算法。     

Constrained adaptation for feedback cancellation in hearing aids.

In feedback cancellation in hearing aids, an adaptive filter is used to model the feedback path. The output of the adaptive filter is subtracted from the microphone signal to cancel the acoustic and mechanical feedback picked up by the microphone, thus allowing more gain in the hearing aid. In general, the feedback-cancellation filter adapts on the hearing-aid input signal, and signal cancellation and coloration artifacts can occur for a narrow-band input. In this paper, two procedures for LMS adaptation with a constraint on the magnitude of the adaptive weight vector are derived. The constraints greatly reduce the probability that the adaptive filter will cancel a narrow-band input. Simulation results are used to demonstrate the efficacy of the constrained adaptation.     

An efficient robust sound classification algorithm for hearing aids

An efficient robust sound classification algorithm based on hidden Markov models is presented. The system would enable a hearing aid to automatically change its behavior for differing listening environments according to the user's preferences. This work attempts to distinguish between three listening environment categories: speech in traffic noise, speech in babble, and clean speech, regardless of the signal-to-noise ratio. The classifier uses only the modulation characteristics of the signal. The classifier ignores the absolute sound pressure level and the absolute spectrum shape, resulting in an algorithm that is robust against irrelevant acoustic variations. The measured classification hit rate was 96.7%-99.5% when the classifier was tested with sounds representing one of the three environment categories included in the classifier. False-alarm rates were 0.2%-1.7% in these tests. The algorithm is robust and efficient and consumes a small amount of instructions and memory. It is fully possible to implement the classifier in a DSP-based hearing instrument.     

Active cancellation of occlusion: an electronic vent for hearing aids and hearing protectors

The occlusion effect is commonly described as an unnatural and mostly annoying quality of the voice of a person wearing hearing aids or hearing protectors. As a result, it is often reported by hearing aid users as a deterrent to wearing hearing aids. This paper presents an investigation into active occlusion cancellation. Measured transducer responses combined with models of an active feedback scheme are first examined in order to predict the effectiveness of occlusion reduction. The simulations predict 18 dB of occlusion reduction in completely blocked ear canals. Simulations incorporating a 1 mm vent (providing passive occlusion reduction) predict a combined active and passive occlusion reduction of 20 dB. A prototype occlusion canceling system was constructed. Averaged across 12 listeners with normal hearing, it provided 15 dB of occlusion reduction. Ten of the subjects reported a more natural own voice quality and an appreciable increase in comfort with the cancellation active, and 11 out of the 12 preferred the active system over the passive system.     

Adaptive feedback cancellation in hearing aids with clipping in the feedback path

Adaptive linear filtering algorithms are commonly used to cancel feedback in hearing aids. The use of these algorithms is based on the assumption that the feedback path is linear, so nonlinearities in the feedback path may affect performance. This study investigated the effect on feedback canceller performance of clipping of the feedback signal arriving at the microphone, as well as the benefit of applying identical clipping to the cancellation signal so that the cancellation path modeled the nonlinearity of the feedback path. Feedback signal clipping limited the amount of added stable gain that the feedback canceller could provide, and caused misadjustment in response to high-level inputs, by biasing adaptive filter coefficients toward lower magnitudes. Cancellation signal clipping mitigated these negative effects, permitting higher amounts of added stable gain and less misadjustment in response to high-level inputs, but the benefit was reduced in the presence of the highest-level inputs.     

A robust adaptive hybrid feedback cancellation scheme for hearing aids in the presence of outliers

It has been demonstrated that the Filtered-x Wilcoxon LMS (FxWLMS) based adaptive filter mitigates the effect of the outliers acquired by the microphone signal of hearing aids by minimizing the Wilcoxon norm and hence shows better cancellation performance than the existing Filtered-x LMS (FxLMS) algorithm. The prediction error method based adaptive feedback canceller (PEMAFC) reduces the bias present in the estimate of the feedback path due to the continuous adaptive filtering (CAF). However, the impulse response of the measured feedback path is close to zero for the first many samples due to the delay introduced by ADC converters and then contains few significant values, which results in slow convergence rate when an adaptive filter is used to model the same. To overcome this limitation, we propose a proportionate normalized WLMS (PNWLMS) algorithm based PEMAFC (P-PNWLMS) for feedback cancellation in hearing aid in the presence of outliers. Further, with an objective to improve the convergence rate and performance accuracy simultaneously, this paper proposes a novel convex PNWLMS (CPNWLMS) algorithm which incorporates convex combination of PNWLMS and WLMS algorithms. The weight update equations are derived for PEMAFC trained by PNWLMS (P-PNWLMS) and CPNWLMS (P-CPNWLMS) algorithms respectively. The results of the simulation study show improved performance of the proposed CPNWLMS based adaptive filter over its component filters.     

Using a signal cancellation technique to assess adaptive directivity of hearing aids

The directivity of an adaptive directional microphone hearing aid (DMHA) cannot be assessed by the method that calls for presenting a "probe" signal from a single loudspeaker to the DMHA that moves to different angles. This method is invalid because the probe signal itself changes the polar pattern. This paper proposes a method for assessing the adaptive DMHA using a "jammer" signal, presented from a second loudspeaker rotating with the DMHA, that simulates a noise source and freezes the polar pattern. Measurement at each angle is obtained by two sequential recordings from the DMHA, one using an input of a probe and a jammer, and the other with an input of the same probe and a phase-inverted jammer. After canceling out the jammer, the remaining response to the probe signal can be used to assess the directivity. In this paper, the new method is evaluated by comparing responses from five adaptive DMHAs to different jammer intensities and locations. This method was shown to be an accurate and reliable way to assess the directivity of the adaptive DMHA in a high-intensity-jammer condition.     

一种用于立体声声学回波消除的新型鲁棒梯度法格梯形自适应滤波算法

立体声声学回波消除的梯度法格梯形算法对脉冲干扰不具鲁棒性。针对这个问题,引入了鲁棒M-估计方法,并从理论上予以了论证。由此产生的双通道M-估计梯度法格梯形算法具有显著提升的抗脉冲干扰鲁棒稳定性,从而也具有了更高的实际应用价值。     

汉语降频助听算法的实验分析与改进

降频助听算法是改善听障患者声音辨识能力的最安全有效的方法. 本文以主观测试实验为手段, 通过分析当前算法的声音识别能力的不足, 提出一种自适应慢放降频算法. 算法结合慢放算法和频移算法的优点, 并能根据信号的频谱结构, 自适应调整慢放因子, 降低时域不同步性. 并且, 通过分析含噪信号和噪声信号的频谱关系, 提出一种噪声下的慢放因子评估方法. 实验结果显示, 同其他降频算法相比, 该算法可以提高15%到20%的识别率. 在对听障患者的测试中, 同传统的助听设备相比, 平均识别率也获得显著改善.