搜索资源列表
noise
- 语音噪声压缩新算法,一种基于vad的语音噪声抑制方法。这种新算法源于vad 技术及谱减法,它先对含噪声信号进 行vad处理,而后是基于感知掩蔽阈值的谱减法运算,降低噪声,获得清晰的增强信号。与其它算法相比,其效果更好,语 音听测试证明了这一点。-New algorithm for speech compression noise, a noise reduction of speech based vad method. This new algorithm derived from v
chapter11_1
- 语音增强之谱相减算法,加窗分帧频域处理,计算信噪比及画出波形对比。-Spectral subtraction speech enhancement algorithm, window-frame frequency domain processing to calculate the signal-to-noise ratio and draw the waveform comparison.
puxiangjian
- 实现语音增强的谱相减算法,实现简单,短小精悍-Enhanced voice spectral subtraction algorithm, simple, short and pithy
soft-decision
- 单通道语音增强,包含所有的经典算法,增强包含稳定环境噪声的语音信号。-speech enhancement algorithms
spectralsubstraction
- 谱减法进行语音增强,其中使用了优化算法减少音乐噪声-Spectral subtraction speech enhancement, which uses optimization algorithms to reduce noise music
omp-column
- 基于字典学习的语音增强中稀疏编码计算稀疏矩阵的一种算法,称作OMP。与一般的OMP不同,本程序针对列向量进行计算,结合给出的总体程序以及KSVD of speech enhancemant.rar文件可以进行字典学习语音增强。-Enhanced sparse coding algorithm to calculate a sparse matrix, called the dictionary-based learning OMP voice. OMP different with the ge
WDRC
- 基于安美森BS250开发的WDRC算法。主要用于语音增强领域。-ON SEN BS300 speech enhancement
MMSE
- 一种基于最小均方误差的噪声估计算法,实现语音增强,算法可以直接使用,已调试通过。-Based on the minimum mean square error of noise estimation algorithm, speech enhancement algorithm can be used directly, have been debug through.
low-complexity-noise-estimation
- 一种低复杂度的噪声谱跟踪算法,用于语音增强和语音识别中,文件夹内有参考文献。-A low noise spectral tracking algorithm complexity, for speech enhancement and speech recognition, there are references folder.
DSBF
- 通过固定波束形成算法进行语音增强的一个仿真程序,这个算法是麦克风阵列语音增强里最基本的,值得研究。-Formed by a fixed beam speech enhancement algorithm, a simulation program, this algorithm is the microphone array speech enhancement in the most basic, is worth studying.
yuyinzengqiang-chengxu
- 含噪语音的语音增强,包换多种算法的去噪效果对比,最后集中在一个操作界面上。-Speech enhancement in noisy speech, noise effect contrast to replacement of several algorithms, finally concentrated in one operating interface.
subspace
- 语音增强中的子空间算法,klt算法以及pklt算法,内附参考文献。- folder contains subspace algorithms: REFERENCES Hu,Y.and Loizou,P.(2003).A generalized subspace approach for enhancing speech corrupted by colored noise.IEEE Trans. on Speech and Audio Processing,11,334-341.
DDalgorithm
- DD算法实现语音增强,噪声谱估计,根据寂静帧和后面能量的对比来估计-DD algorithm speech enhancement, noise spectrum estimation, based on comparative silence frame and rear energy to estimate
mband
- 语音增强的多带谱减算法,函数可直接调用,简单易操作-Speech enhancement of the multi band spectral subtraction algorithm, the function can be called directly
specsub
- 语音增强基本谱减算法,直接调用函数,简单可实现-Speech enhancement of the basic spectral subtraction algorithm, the function can be called directly
logmmse
- 语音增强,对数MMSE算法,函数可直接调用,简单易操作-Speech enhancement, log MMSE algorithm, the function can be called directly, simple and easy to operate
logmmse_SPU
- 语音增强,结合了语音存在不确定度的对数MMSE算法,函数可直接调用-Speech enhancement, combined with the existence of the voice of the uncertainty of the logarithm of the MMSE algorithm, the function can be called directly
sequential_regression
- 自适应语音增强的序贯回归算法,一路为带噪语音信号,一路为噪声信号-Adaptive speech enhancement sequential regression algorithm, all the way to noisy speech signal, all the way to the noise signal
vad
- 关于端点检测的几种方法,语音样本是自己录制的,对传统算法做了一些改进,加入了去噪,去噪之后再进行端点检测,均调通 vad0303:自己设置调整门限为一定值 vad0310:根据能量值和过零率设置门限,自适应门限值 vad0310_2:基于比例因子的门限自调整 vad0310.m加入了噪声,端点检测前都噪声进行了处理 entropy.m:基于自适应子带频谱熵的稳健性语音端点检测 可用于语音增强及端点检测 dbdoor.m:双门限算法,用于语音端点检测。可以通过调整门限值,并
kalmanfilter
- 使用卡尔曼算法对输入的数据进行平滑滤波,减小噪声的影响,可适用于语音增强,运动轨迹平滑等(The Kalman algorithm is used to smooth the input data and reduce the influence of noise. It can be applied to speech enhancement, motion trajectory smoothing and so on.)