共查询到19条相似文献,搜索用时 109 毫秒
1.
2.
3.
基于压缩感知观测序列倒谱距离的语音端点检测算法 总被引:2,自引:0,他引:2
本文基于语音信号在离散余弦基上的近似稀疏性,采用稀疏随机观测矩阵和线性规划重构算法对语音信号进行压缩感知与重构。研究了语音信号的压缩感知观测序列特性,根据语音帧和非语音帧压缩感知观测序列频谱幅度分布分散且差异较大的特性,提出基于压缩感知观测序列倒谱距离的语音端点检测算法,并对4dB-20dB下的带噪语音进行端点检测仿真实验。仿真结果显示,基于压缩感知观测序列倒谱距离的语音端点检测算法与奈奎斯特采样下语音的倒谱距离端点检测算法一样具有良好的抗噪性能,但由于采用压缩采样,减少了端点检测算法的运算数据量。 相似文献
4.
5.
6.
为了提高低信噪比下语音端点检测的准确性,提出一种基于经验模态分解与功率谱熵的语音端点检测方法。对带噪语音信号进行经验模态分解获得一系列语音本征模函数,选取功率谱熵作为语音端点检测的特征,并计算特定阶本征模函数的功率谱熵实现语音的端点检测。通过EMD分解可以有效地消除白噪声的影响,仿真结果表明,在低噪比情况下结合经验模态分解和功率谱熵的方法能够有效实现语音端点检测。 相似文献
7.
8.
9.
10.
基于子带二次谱熵的语音端点检测 总被引:1,自引:0,他引:1
为了提高在强噪声环境下语音端点检测的准确度,提出基于子带二次谱熵的端点检测算法.该算法把子带二次谱熵作为端点检测新的特征参数,首先计算每帧语音信号的二次谱,再多子带分析,计算二次谱熵;引入顺序统计滤波对二次谱熵平滑处理;将有限状态机判别方法与子带二次谱熵相合,形成新的语音/噪声判别算法,有效地解决单门限法易出现的两类误判.实验表明:与传统的两种方法相比,提出的端点检测算法具有准确性高、抗噪性强等优点. 相似文献
11.
12.
In this paper, we present a new approach to high-resolution spectral characterization of the unknown number of spectral line components embedded in colored noise. The addressed method resolves the spectral analysis problem via intelligent fusing the two spectrum estimation paradigms: (i) the parametric line spectral estimation that employs the modified regularized Prony (MORP) method for multi-harmonic signal characterization and (ii) nonparametric spectral estimation. Two nonparametric high-resolution spectral estimation methods are proposed to be fused with the MORP: the minimum variance (MV) and maximum entropy (ME) techniques. Via aggregation of the developed model-based MORP and model-free MV/ME techniques into the fused MORP-MV/MORP-ME resulting method a substantial improvement of the spectral characterization performances is gained when those are applied to characterization/analysis of the composed distributed scenes that contain noised closely spaced spectral lines to be localized with high resolution and accuracy. The simulation results are presented to illustrate the performance enhancement gained with the proposed fused MORP-MV/MORP-ME method. 相似文献
13.
14.
高脉冲噪声坏境中双门限法语音端点检测研究 总被引:1,自引:0,他引:1
语音端点检测是对有效语音段的识别关键技术,准确的端点检测使语音信号的后续处理计算量减少,有效地节约资源。现在多数语音端点检测技术例如能频值、谱熵、小波能量熵变换等都能准确检测出有效的语音段。文中介绍了一种双门限端点检测法,即利用短时平均过零率和短时平均能量法进行双门限检测,再设置一个最短时间门限,有效地在高脉冲噪声环境中准确识别汉语发音。通过与其他方法对比实验,文中双门限技术在短时高脉冲噪声环境下能有效提高语音识别率。仿真结果表明,端点检测正确率达93%。 相似文献
15.
16.
Use of sample entropy approach to study heart rate variability in obstructive sleep apnea syndrome 总被引:6,自引:0,他引:6
Sample entropy, a nonlinear signal processing approach, was used as a measure of signal complexity to evaluate the cyclic behavior of heart rate variability (HRV) in obstructive sleep apnea syndrome (OSAS). In a group of 10 normal and 25 OSA subjects, the sample entropy measure showed that normal subjects have significantly more complex HRV pattern than the OSA subjects (p < 0.005). When compared with spectral analysis in a minute-by-minute classification, sample entropy had an accuracy of 70.3% (69.5% sensitivity, 70.8% specificity) while the spectral analysis had an accuracy of 70.4% (71.3% sensitivity, 69.9% specificity). The combination of the two methods improved the accuracy to 72.9% (72.2% sensitivity, 73.3% specificity). The sample entropy approach does not show major improvement over the existing methods. In fact, its accuracy in detecting sleep apnea is relatively low in the well classified data of the physionet. Its main achievement however, is the simplicity of computation. Sample entropy and other nonlinear methods might be useful tools to detect apnea episodes during sleep. 相似文献
17.
提出了一种基于最小统计及短时对数谱幅度最小均方误差估计单通道语音增强算法,并在此基础上对其进行了修正。噪声功率谱估计不需要VAD进行有无语音的检测,并在每帧数据都进行更新,可跟踪变电平噪声。该算法在MMSE—LsA准则下得到谱增益函数,并考虑到纯净声音信号频谱特性,对增益函数进行了修正。实验结果表明,该算法可有效去除噪声,在消除音乐噪声的同时对语音信号产生很小的失真,并易于实时处理。 相似文献
18.
19.
针对在没有对称语音库的情况下,该文提出了一种基于混合线性变换的语声转换算法,在最大似然估计准则下,使用EM迭代算法计算变换函数的参量。为了减小线性加权对语音谱包络的平滑作用,使用线性调频Z变换来调节语音信号的LPC系数。客观评测和主观感受的实验结果都表明,基于混合线性变换的语声转换算法也可以取得与传统语声转换技术相当的转换效果,解除了传统语声转换技术需要对称语音库的要求。 相似文献