资源列表
语音分帧
- 编写的几个子函数对语音信号进行分帧处理。(Several sub functions are written to frame the speech signal.)
sound_to_spectrogram2
- 从大文件下读取子文件夹里的声音，并转成语谱图，并保存在本地电脑上面(sound transfer spectrogram)
dtw
- python版的dtw算法极其改进fastdtw算法实现(The DTW algorithm in the python edition improves the implementation of the fastdtw algorithm extremely)
ABSE
- 熵值越大则每个符号包含的平均信息量越大。有研究发现，在有噪声的语音信号中，语音信号的熵和噪声信号的熵存在着较大的差异，对噪声信号来说在整个频带内分布相对平坦，熵值小，语音信号集中在某些特定频段内，熵值大。因此利用这个差异可以区分噪音段和语音段。(The greater the entropy is, the greater the average information of each symbol is. It is found that, in noisy speech signals, t
75266092lms_noise_cancellation
- 自适应滤波，语音信号处理，变换域lms算法，matlab(Adaptive filter, the speech signal processing, the transform domain LMS algorithm and matlab)
jizhengzhuoye
- 均匀线列阵和均匀平面阵的自然指向性图的画法(just a little program)
duru
- 读入wav语音，实现多语音叠加 对语音进行分帧 左声道右声道分别作图(Reading wav voice to achieve multi speech superposition Speech segmentation Left vocal tract right vocal tract mapping)
LPCC
- 线性预测倒谱系数（Linear Prediction Cepstrum Coefficient,LPCC）是线性预测系数（Linear Prediction Coefficient,LPC）在倒谱域中的表示。该特征是基于语音信号为自回归信号的值设，利用线性预测分析获得倒谱系数。(Linear Prediction Cepstrum Coefficient)
vad
- 可对一段信号进行分帧加窗，利用过零率和短时能量检测语音端点(It can divide the frame into a window and use the zero crossing rate and short time energy to detect speech endpoints.)
beamforming1
- 语音信号处理，阵列为10阵元线阵，宽带和窄带波束形成(Speech signal processing, 10 sensors of microphone linear array,Broadband and narrowband beamforming)
mfcc
- MFCC，Mel频率倒谱系数的缩写。Mel频率是基于人耳听觉特性提出来的，它与Hz频率成非线性对应关系。Mel频率倒谱系数(MFCC)则是利用它们之间的这种关系，计算得到的Hz频谱特征，MFCC已经广泛地应用在语音识别领域。(MFCC, Mel frequency cepstrum coefficient abbreviation. The frequency of Mel is based on the auditory characteristics of human ear. It is
yuchuli1
- 基于python平台的语音信号的预处理和MFCC39维度的特征提取(MFCC based on python)