lms_dfe

实验示例是基于语音中的mfcc，语音倒谱特征来进行聚类，先利用训练样本来计算训练样本聚类中心（用到了lbg算法），之后再进行分类。 注意：使用代码时需要自己更改文件路径。(This example is based on the MFCC in speech and the feature of speech Cepstrum to cluster. First, the training sample is used to calculate the training sample clustering center (using the LBG algorithm), then the classification is then carried out.)

一个完整的基于Matlab的DTW模型算法及高效算法程序，能快速识别数字0-9,运行testdtw即可。(An excellent MATLAB program for the algorithm as a DTW model. It can recognise the number of 0-9. Try it, just run testdtw.)

I:CSDN技术中心 进程管理器---一个SDK的例子(I : CSDN Technology Center process management-- for example an SDK)

VAD去噪算法，可以有效的过滤静音和无效的语音，同时消除了一些背景噪声。(VAD denoising algorithm, can effectively filter invalid mute and voice, while eliminating some background noise.)

C语言fft和ifdct的函数，可以直接使用(C language functions fft and ifdct can be directly used)

说明：  做毕业设计编写的基于谱熵端点检测的程序，欢迎分享(Designed to prepare graduates to do spectral entropy-based endpoint detection process, please share )

说明：  基于自适应子带频谱熵的稳健性语音端点检测(Based on adaptive sub-band spectral entropy of the robustness of Speech Endpoint Detection)

HMM语音识别,包含主要算法,进行语音识别(HMM speech recognition, contains the main algorithm for speech recognition)

基于最小均方误差的助听器语音去噪算法, 用于以最大程度消除助听器接受的语音杂音.(Speech denoising algorithm based on the minimum mean square error of a hearing aid for the hearing aid acceptable voice murmur eliminated to the greatest extent.)

gmm模型下的语音情感识别系统，GMM只是一个数学模型，只是对数据形态的拟和，但是和你所看到的数据分布存在出入也是正常的，因为用EM估计GMM的那些参数时，一般假设我们所得到的数据是不完备的（也就是说假设我们看到的数据分布不是真正的分布，它在运算时把那部分丢失或者叫隐藏的数据“补”上了）(gmm model speech emotion recognition system, GMM is a mathematical model, but fitting the data form, but you can see there is access to the data distribution is normal, because those with the EM estimated GMM parameters, the general assumption we get the data is incomplete (that is, assuming we see the distribution of the data distribution is not real, it is part of the operation at the time that is lost or hidden data is called " fill" on the))