我们尝试把这些端点侦测的方 法与兩种强健式语音特征參數撷取法结合，即频谱消去法(spectral subtraction)与静 音对數能量正规化法(silence log-energy normalization, SLEN)等，发现皆有相当程 度的提高辨識效率。
基于24个网页-相关网页
常用的算法有谱减（Spectral Subtraction，SS）法［55-571、非线性谱减（Non—linear SpectralSubtraction，NSS）...
基于8个网页-相关网页
spectral subtraction method 光谱差减法
power spectral subtraction 功率谱减法
Multi-Band Spectral Subtraction 多带谱减法
Long Term Log Spectral Subtraction 长时对数能量减谱法
nonlinear spectral subtraction 法 ; 谱减法
Method of spectral subtraction is used to be compared with the former mothod at last.
最后,用谱减法去噪和前者做了比较。
参考来源 - 基于小波变换的多尺度多阈值语音信号去噪The main innovation ideas of the dissertation are listed as follows.1. An endpoint detection algorithm that combines expanded spectral subtraction with the SAP (speech absence probability) dynamic threshold is proposed based on traditional methods.
论文的主要研究内容如下:1.提出了一种基于扩展谱相减的语音增强算法,使得对背景噪声的估计相对传统方法更加精确。
参考来源 - 说话人识别中语音特征参数提取方法的研究The simulation result shows that the new method is not only able to remove the strong noises effectively,but also can hold the main details wel1,which is more effective method of speech signal denoising than method of spectral subtraction.
仿真实验结果表明,本方法能有效去除信号中的噪声和较好保留语音细节,与谱减法去噪相比,能达到更佳的语音信号去噪效果。
参考来源 - 基于小波变换的多阈值法语音去噪研究·2,447,543篇论文数据，部分数据来源于NoteExpress
The application range of the spectral subtraction is expanded by the new noise estimation method.
算法提高了谱减法的适用范围，还在一般谱相减方法的基础上提出了改进的谱相减算法。
In this paper, a speech enhancement approach using minimum estimate and spectral subtraction is proposed.
提出了一个基于最小统计及谱减法的语音增强方法。
The comparison of threes methods, spectral subtraction, LOGSTSA-MMSE and subspace approach, is the major work in researching the noise reduction algorithm.
在去噪算法上本文重点比较了谱相减法、LOGSTSA-MMSE与子空间方法。
应用推荐