speech-signal-processing

目前2.4kbps的混合激励线性预测 (MELP) 语音编码方法已经被确定为美国新的联邦语音编码标准.本文提出了一种改进的MELP语音编码方法,利用滤波器相似度和基于LPC系数分类的矢量量化技术,可以把MELP的码率降到1.7kbps以下,仍有较好的合成语音质量.(An Improved MELP Speech Coder)

语音语谱图分析程序,提供了数字样本数据,值得参考.matlab程序.(language voice spectrum analysis procedures, and providing a digital sample data, a good reference. Matlab procedures.)

基于倒谱短时部分反映了语音的声道特性，先用汉明窗取一帧语音，然后经变换得到语音倒谱，将倒谱短时部分取出，进行正交反变换后将得到声道的对数谱，即得到语音频谱的包络。将频谱包络和频谱画在一张图上，有很好的对比效果。获取的包络效果十分好。(Based Cepstral partly reflects the short channel characteristics of the speech, first take a Hamming window with a frame of speech, and speech cepstrum obtained by converting the cepstrum short segment out inverse orthogonal transform to obtain channel will of the spectrum, i.e. to obtain the envelope of the speech spectrum. The spectral envelope and spectral painted on a chart, there is a good contrast. Get the envelope effect is very good.)

语音识别程序。有需要的抓紧啦。用MATLAB实现的。(Speech recognition procedures. You need to have the grasp. Achieved using MATLAB.)

3、语音特征提取与分类 首先, 待识别语音转化为电信号后输入识别系统, 经过预处理后用数学方法提取语音特征信号, 提取出的语音特征信号可以看成该段语音的模式。(3, the voice feature extraction and classification First, to be recognized voice into electrical signals input recognition system, after pretreatment with a mathematical method to extract speech characteristic signal extracted speech characteristic signal can be seen as the voice segment model.)

这是基于隐马尔可夫模型的连续语音识别代码，不同于dtw,这是个完整的工程，我把需要的voicebox也放进去了，一共用到9个子函数，包括模板的训练算法何识别算法，是完全可以使用的。(This is based on the hidden Markov model of continuous voice recognition code, different the dtw, which is a complete project, I need to put the voicebox into a total of nine sub-functions, including the template training algorithm to identify the algorithm, Is fully available.)

使用TD-PSOLA（td-psola、psola）算法，做了一个语音合成，实现语音声调的变调，时长的变化等功能。参考praat源码，提取出其合成方法，效果不错。需要提供pitchtier、durationtier、pointprocess等准换信息文本（可由praat产生）供readfile函数读入(Use TD-PSOLA (TD-PSOLA, PSOLA algorithm, made a speech synthesis, speech tone tone sandhi, long-term changes in the function. Reference Praat source code, extract its synthetic method, the effect of the same.)

基于espeak的一个文本转语音的代码,可以把文本转成WAV文件(basic espaeck to translate txt to wav file)

mfcc特征提取法 以及svm训练 可以使用(MFCC feature extraction method and SVM training can be used)

本文介绍MVDR算法实现及其改进 用于多麦克风语音增强(The minimum variance distortionless response (MVDR), originally developed by Capon for frequency-wavenumber analysis, is a very well established method in array process-ing. It is also used in spectral estimation. The aim of this paper is to show how the MVDR method can be used to es-timate the magnitude squared coherence (MSC) function, which is very useful in so many applications but so few methods exist to estimate it)