This is a formula editor, the preparation of a miniature non

编译环境：vc++.net2003 名称：语音合成软件 作者：yxifu-Program environment: VC++.net2003 Name: program of speech synthesize Programer: yxifu

用标准C语言实现的基于DTW算法的特定人孤立词语音识别系统，能实现汉语数字0至9个语音发音的识别，具有很好的识别性能。-Using standard C language based on the DTW algorithm to achieve the specific words were isolated speech recognition system, to achieve Chinese figures 0-9 voice pronunciation recognition, with good recognition performance.

改进的gmm（高斯混合模型）算法，是单一高斯几率密度函数的衍生-Improved gmm (GMM) algorithm, a single Gaussian probability density function of the derivative

实现实时声卡声音波动的傅立叶变换后显示输出，附带源代码。-Sound card real-time voice fluctuations show the output after the Fourier transform, with the source code.

AR模形拟合工具，提供AR模型的拟合与模拟-AR model

基于GMM的概率神经网络PNN具有良好的泛化能力，快速的学习能力，易于在线更新,并具有统计学的贝叶斯估计理论基础，已成为一种解决像说话人识别、文字识别、医疗图像识别、卫星云图识别等许多实际困难分类问题的很有效的工具。而且PNN不但具有GMM的大部分优点，还具有许多GMM没有的优点，如强鲁棒性，需要更少的训练语料，可以和其他网络其他理论无缝整合等。-GMM based probabilistic neural network PNN good generalization ability, the ability to learn fast, easy online updates, and with the Bayesian statistical theory based on estimates, and has become a solution as speaker recognition, text recognition, medical image recognition, satellite images and other real recognition when difficulties classification of very effective tool. But GMM PNN is not only the most advantages, but also has many advantages GMM not as strong robustness, require less training corpus, and other networks to other theories, such as seamless integration.

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.

nlms算法及其示例,与大家分享,希望大家可以帮我弄个多输入的NLMS-nlms algorithm and its examples to share with you, I hope everyone can help me multi-input弄个the NLMS

本文分析了用经验模态分解( Empirical Mode Decomposition ,EMD) 方法处理非平稳噪声信号的基本原理 , 并利用 EMD 方法有效地消除了高斯白噪声对瞬态散射回波的干扰. 文中以一人工合成瞬态回波为例 ,对本方法的有 效性作了检验.-This paper analyzes the use of empirical mode decomposition (Empirical Mode Decomposition, EMD) method to deal with non-stationary noise signals in the basic principles and methods of use of EMD effectively eliminates the Gaussian white noise on the transient scattering echo interference. Text in one artificial Synthesis of transient echoes as an example of the effectiveness of this method were tested.

改进的BP神经网络的C++源代码,可用于图象识别-Improved BP neural network C++ Source code, can be used for image recognition