specsubm

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))

最权威和最全面的语音识别的特征提取的 IEEE的期刊论文1980(The most authoritative and comprehensive speech recognition feature extraction of the IEEE Journal Papers 1980)

功率倒置算法的LMS实现，因为自己遇到了点问题，本人一直在找这个程序，没找到，故自己重新检查把问题发现了，相信如果你要实现的话可能会遇到这个问题祝你好运！(Power inversion algorithm LMS implementation, because they encountered a point problem, I ve been looking for this program, did not find it to re-examine the problem yourself discovered, I believe if you want to achieve, then you may encounter this problem Good luck!)

三种降噪，回声和变调的语音信号处理，含UI界面(Three kinds of noise reduction, echo and tone of voice signal processing, including UI interface)

多信号分离MUSIC的经典学习程序 对初学者帮助很大(it is a baisc and useful code of the multiple signal classfication for the beginners.)

LMS多麦克风语音降噪的主程序是lmsspdn.m(Multi-microphone noise reduction LMS voice is the main program lmsspdn.m)

自制DMX512控制台,很不错的一个软件,他对灯光的控制,非常出色.(Homemade DMX512 console, a very good one software, he lighting control, very good.)

simple voice recognition program

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)

the document explains speech recognition using mel frequency cpstrum coeefficient and noise reduction method