基于短时自相关及过零率的语音端点检测算法

传统的基于短时能量端点检测算法,在高信噪比环境下可以比较准确地检测出语音端点,但在低信噪比环境下检测效果不理想。文中提出了基于短时自相关最大值与短时过零率之积的改进算法。利用短时自相关最大值可以有效地区分出语音段和噪音段,利用短时过零率可有效地检测出清音信号,将两参数相结合可有效地检测出低信噪比语音信号的端点。实验证明,在低信噪比环境下该改进算法相比短时能量算法减小了检测误差,可以有效地检测出语音端点。     

一种用于光纤周界安防系统的端点检测方法

为了提高光纤周界安防系统中扰动信号端点检测的 性能,提出一种新型端点检测方法。首先通过计算信号 的短时能量和过零率分布分别找出信号中短时能量比和过零率满足一定门限的信号点,通过 对这两信号点 进行一定的加权确定扰动信号的端点位置。与离散小波提取起始点信号相比,本文方法由于 综合了短时能量 和过零率法的优势,能够适应更多的扰动事件类型,通过加权极大提高了端点检测的精度 ,并且运算时 间大幅降低,提高了系统的实时性。实验结果表明,本文算法在10M Hz的采样速度和采集0. 3s数据情况下,93.47%的数据定位误差低于10m,计算时间少于0.3s。     

基于心率和腕动信号的睡眠监测系统

设计一种基于心率和腕动信号的睡眠监测系统,通过传感器采集睡眠过程中的心率和腕动信号,利用串口通信将信号传输到上位机,上位机软件通过睡眠分期算法对采集的信号进行处理分析,达到客观评价睡眠质量的目的。该设计为客观评估睡眠质量提供了实现方案,帮助人们了解自己的睡眠状况。     

基于压电薄膜传感器的睡眠监测枕研究

针对人体睡眠的健康监测问题,分析睡眠时人体头部对枕头压力状态变化,研究出一种基于聚偏氟乙烯(PVDF)压电薄膜传感器的睡眠监测枕。睡眠时采集人体脉搏、呼吸及鼾声生理信号,硬件电路调理、MSP430F149单片机有机处理后经蓝牙上传至LabVIEW程序,结合打鼾无规则振动特征,依据信号幅值、均值同预设值对比的双阈值分析方法,实现睡眠时呼吸暂停识别报警,鼾声自动监测记录及睡眠呼吸暂停综合症前期检查监测等功能。     

基于体震信号的睡眠检测系统的设计与实现

设计并实现了一种基于体震信号的睡眠监测系统。肢体动作、胸腔扩张收缩、心脏泵血都会引起人体和与人体接触物体的震动,采集这些震动,经过压电传感器转换为电信号,通过放大模块等信号处理电路,再传输到后台进行预处理,利用一种基于动作幅度、心率、呼吸而建立的睡眠检测算法得到睡眠质量。同步使用专业睡眠监测仪来监测睡眠,并检验系统的准确性。     

SOM-LSTM递归神经网络语音端点检测系统

语音端点检测是语音信号处理的一个重要环节,传统的语音端点检测算法往往是基于短时能量以及过零率等实现,在低信噪比的环境下,检测算法的准确度较低。因此,提出了一种基于自组织映射(SOM)神经网络和长短时记忆(LSTM)递归神经网络相结合的端点检测算法。该算法通过检测语音信号在每个时间节点上的特征属性利用SOM神经网络进行聚类,并根据每个时间节点的语音状态对聚类结果进行调整,构造能够判别语音状态和噪声状态的样本作为LSTM递归神经网络的输入,利用LSTM递归神经网络实现端点检测的目的。     

实时语音采集系统中语音端点检测和增强方法

针对实时语音采集系统应用到工业行业中所面对的增加有效记录时间和提高噪声环境下的录音回放效果问题,研究了端点检测方法和语音增强方法。端点检测方法基于有限状态机模型,引入通信中AMR-WB标准的激活检测算法作为语音帧或噪声帧判据,端点检测可达到良好的实时性和准确性,并对于低幅度信号增加了过零率判据可进一步降低检测算法对语音的漏报率。语音增强方法以参数可配置的低频带阻滤波结合维纳滤波来增强语音采集系统对多种环境噪声的消除能力。实际使用表明该端点检测方法和语音增强方法对于提高工业领域应用中实时语音采集系统的记录长度和回放效果具有很强的实用性。     

基于ChaffMatrix的光载声音声纹变化识别系统设计

采集光载声音时会受到激光传输信道噪声的干扰,传统声纹变化识别系统对于信道噪声的去噪效果不好,影响最终的声纹识别准确性,因此,设计一种基于ChaffMatrix的光载声音声纹变化识别系统.系统沿用传统系统的硬件设计,主要对于软件方面进行详细研究.首先对一段语音端点进行检测,采用单门限阈值法来检测能量端点,排除噪声优化计算...     

非平稳强噪声环境中的音频信号端点检测系统

为提高音频信号端点识别能力,设计一种非平稳强噪声环境中的音频信号端点检测系统。构建音频信号端点检测硬件单元,利用预处理单元对音频信号进行预加重、分帧以及加窗处理后,端点检测单元在提取处理音频信号的MFCC倒谱距离特征、频带方差特征的基础上,依据动态阈值估计策略确定恰当阈值;通过双特征参数双门限法来实现对音频信号起止点的确定以及语音帧和非语音帧的分离;利用包络确定延时单元,防止噪声段被错误识别为语音段,避免出现拖尾太长问题。实验结果表明,所设计系统可实现非平稳强噪声环境音频信号端点检测,检测误差满足设定要求。     

基于PVDF的可穿戴生理信号监测系统

呼吸、心率、鼾声反映了人体在睡眠时的大量信息，该文以聚偏氟乙烯(PVDF)作为敏感单元进行呼吸、脉搏、鼾声信号的监测。根据所采用的传感器特性分别进行了电荷放大电路、陷波电路及电压放大电路的设计。硬件电路通过聚合物锂电池进行供电，根据聚合物电池的特性分别为电池设计了充电电路、保护电路及放电电路，硬件电路整体集成在一块印制电路板(PCB)上。同时设计了基于Android 设备的APP，以可视化形式实时显示生理信号数据，并对其进行长期储存，便于后期医生进行睡眠呼吸病症的分析诊断。该研究的目的是能准确监测睡眠生理参数，提高被测试者的使用舒适感。     

超脉冲CO2激光悬雍垂腭咽成形术治疗鼾症的近期临床研究

目的：探讨超脉冲CO2激光悬雍垂腭咽成形术治疗鼾症的可行性，技术操作及疗效。方法：采用超脉冲CO2激光悬雍垂腭咽成型术治疗鼾症64例。结果：该组病例近期均获得满意疗效，远期疗效达87.5%。术中出血极少，勿需缝合切口，切割创缘周期组织反应轻，手术时间短，无并发症发生。结论：该术式具有明显的临床应用价值，值得推广应用。     

Nonconstrained Sleep Monitoring System and Algorithms Using Air-Mattress With Balancing Tube Method

We evaluated the performance of a bed-type sensor system using the air-mattress with balancing tube (AMBT) method to noninvasively monitor the signals of heartbeat, respiration, and events of snoring, sleep apnea and body movement of subject on the system. The proposed system consists of multiple cylindrical air cells, two sensor cells and 18 support cells, and the small physiological signals were measured by the changes in pressure difference between the sensor cells, and the dc component was removed by balancing tube that is connecting the sensor cells. Using newly developed AMBT method, heartbeat, respiration, snoring, and body movement signals were clearly measured. For the concept of a home healthcare system, two automatic processing algorithms were developed: one is to estimate the mean heart and respiration rates for every 30 s, and another one is to detect the snoring, sleep apnea, and body movement events from the measured signals. In the beat-to-beat heart rate and breath-by-breath respiration rate analyses, the correlation coefficients of the heart and respiration rates from the proposed AMBT method compared with reference methods, electrocardiogram, and respiration effort signal from piezoelectric belt, were 0.98 ( $p$ ≪ 0.01) and 0.96 ($p$ ≪ 0.01), respectively. Sensitivity and positive predictive value (PPV) of the detection algorithm for snoring event were 93%, 96%, for sleep apnea event were 93%, 88%, and for body movement event were 86%, 100%, respectively. These findings support that ABMT method provides an accurate and reliable means to monitor heartbeat, respiration activities and the sleep events during sleep.      

基于分形维的连续语音实时分割技术

为能准确有效地判断出连续语音中各个音节的起止点,提出了一种改进的分形维计算方法,该算法将插值分形维的步长因子进一步精确到采样频率的数量级上,先求出语音分形维的最小二乘能量轨迹,再差分求其动态特征;在此基础上,设计了连续语音的两级搜索实时分割算法,并进行了基于DSP的硬件系统实验。结果证明,该算法较好地实现了语音段的实时分割和汉语音节切分,鲁棒性好,使得系统在信噪比为0 dB时音节分割准确率仍可保持在一个较高的水平上。最后开发了一个在线汉语语音标注器,借此阐述了论文工作在语音识别方面的应用。     

Snore signal enhancement and activity detection via translation-invariant wavelet transform

Acoustical properties of snores have been widely studied as a potentially cost-effective and reliable alternative to diagnosing obstructive sleep apnea (OSA), with a common recognition that the diagnostic accuracy depends heavily on the snore signal quality and intelligibility. This paper proposes a novel preprocessing system that performs two critical tasks concurrently in a translation-invariant wavelet transform domain. These tasks include enhancement of snore signals via a level-correlation-dependent (LCD) threshold, and identification of snore presence through a snore activity (SA) detector. Various experiments were conducted to warrant the robustness of the system in terms of theoretical statistics quality, signal-to-noise ratio, mean opinion score, and clinical usefulness in detecting OSA. Results indicate that the proposed LCD threshold and SA detector are highly comparable to the existing denoising methodologies using level-dependent threshold and segmentation approaches using short-time energy and zero-crossing rate, yielding the best results in all the experiments. Given the strong initial performance of the proposed preprocessing system for snore signals, continued exploration in this direction could potentially lead to additional improvement in signal integrity, thereby increasing the diagnostic accuracy for OSA.     

基于语谱图的语音端点检测算法

利用语音在语谱图中表现出的不同特征,提出了一种基于语谱图的语音端点检测算法。首先利用基音频率检测的原理在语谱图矩阵中搜索浊音段,然后计算出浊音段的信噪比,再根据信噪比和语谱图矩阵中浊音段的峰值进行完整的端点检测。因多数突发噪声并没有稳定的频率或者频率不在人的基音频率范围内,因此,该算法能够很好地抑制突发噪声的干扰,实验结果表明,在信噪比为10dB以上时该算法能够准确检测出语音的端点位置。     

Speech Endpoint Detection in Noisy Environments Using EMD and Teager Energy Operator

Accurate endpoint detection is a necessary capability for speech recognition.A new energy measure method based on the empirical mode decomposition(EMD)algorithm and Tcager energy operator(TEO)is proposed to locate endpoint intervals of a speech signal embedded in noise.With the EMD,the noise signals can be decomposed into different numbers of sub-signals called intrinsic mode functions(IMFs),which is a zero-mean AM-FM component.Then TEO can be used to extract the desired feature of the modulation energy for IMF components.In order to show the effectiveness of the proposed method,examples are presented to show that the new measure is more effective than traditional measures.The present experimental results show that the measure can be used to improve the performance of endpoint detection algorithms and the accuracy of this algorithm is quite satisfactory and acceptable.     

天线组阵中基于SUMPLE的联合信号检测

针对低信噪比下天线组阵信号的检测问题,本文提出了一种基于SUMPLE合成的联合检测算法。传统信号检测算法在低信噪比下的检测效率很低,并且难以直接应用于多天线组阵系统。本算法综合利用SUMPLE合成时权值收敛特性以及各路信号之间强相关性,对信号存在性进行检测。理论分析和仿真结果表明,该算法具有适用于多天线组阵系统以及检测概率高的优势。并且在接收信号信噪比低至-12dB时,算法仍然有80%以上的检测概率,提高了传统多天线组阵信号检测的性能。      

基于软件无线电的GNSS干扰和多径监测系统设计

针对干扰和多径导致全球导航卫星系统(GNSS)不能精确、安全和可靠地提供位置、速度和时间服务等问题,分析了研制GNSS干扰和多径监测系统的意义,介绍了当前干扰和多径常用的检测方法,设计了一套针对当前四大导航系统(美国GPS、俄罗斯GLONASS、欧盟Galileo和中国BDS)各信号带内以及附近频率的干扰和多径监测系统.系统基于软件无线电(SDR)思想,采用模块化设计,具有良好的功能扩展性和配置灵活性.根据不同干扰的特点,结合实际应用改进了其检测算法:采用能量阈值法检测不同形式的压制式干扰,采用基于捕获多相关峰法判别欺骗式干扰,采取基于最大似然准则的多径估计延迟锁定环(MEDLL)技术估计多径信号的参数.实验结果表明系统能够有效检测各种干扰和多径信号的存在,使用户获取其特征参数,为干扰源的定位、查处、排除、规避以及多径信号的抑制提供决策依据.