Linux下的CELP语言编码，压缩语音质量极好。支持2K到32K的压缩。实现部分使用C语言编写。可用做Linux下的动态库，有例子。...

这是小弟在实验室做的说话人身份识别程序，采用MFCC参数提取，并建立GMM模型来进行识别。希望会对大家有帮助。-This is the boy in the laboratory to do the speaker identification procedure for the extraction of MFCC parameters and to establish GMM model to be identified. I hope everyone will help.

是峰值检测的源码，经本人实验过，发现很有用。-Is the source peak detector, the experiment I have found useful.

采用VC＋＋编程和G.729协议，实现语言的实时通信，内含源代码。-using VC++ programming and G.729 agreement, the realization of real-time communication language, containing the source code.

fastICA的matlab实现。快速独立成分分离算法，用于去除图像或声音之间的 相关性。-The FastICA matlab realize. Fast independent component separation algorithm for the removal of images or sound correlation between.

基于MATLAB的语音识别系统程序,包括HMM,DTW,Record三个matlab的M文件.-MATLAB-based speech recognition system procedures, including HMM, DTW, Record three Matlab M documents.

cubica算法通过同时对三阶和四阶累积量进行对角化来实现ICA

静态Huffman的小程序，帮助理解Huffman的原理，不可多得。-small procedure to help understand the principle Huffman indeed.

几篇关于基音检测的论文.可以看看哈.以后还将传上来-Several papers on the pitch detection. Can look at Kazakhstan. Later will mass up

CMU sphinx，是美国卡内基梅隆开发的开源语音识别系统，该程序用C编写，模块结构非常清晰，函数接口明了，非常适合用来进行二次开发。-CMU sphinx, is a Carnegie Mellon developed in open-source speech recognition system, the program is prepared to use C, the module structure is very clear, the function interface, clear, very suitable for secondary development.

windows操作系统下就能使用的SVM-light4.0最新版,很有价值哟-windows operating system will be able to use the SVM- light4.0 latest version, value yo