yuyingshibie

这是科大讯飞的IOS开发文档，可合成语音，可语音识别(MSC Develop Manual for iOS)

本代码是基于matlab实现的melp算法，编码速率为2.4kb/s(This code is based on the matlab implementation melp algorithms, coding rate of 2.4kb/s)

speaker recognition using MFCC GMM EM

Matlab 语音合成，可进行语音处理与还原(Matlab dealing with voice)

主要是研究光纤通信系统中的克服PMD效应的电域均衡技术，构建电域均衡器的模型并进行仿真实验。找到补偿PMD效应的较理想的电域均衡器结构。(Is to study, fiber-optic communication systems to overcome the effects of PMD electrical domain equalization technology, building a power domain equalizer model and simulation experiments. PMD compensation effect to find a better balance of power domain structure.)

一种关于倒谱的实现研究处理方法的总结以及展望(One kind of research on cepstrum processing method to achieve a summary and outlook)

对语音信号基频的提取对于语音信号的分析有一定作用(Basic frequency of speech Analysis )

HMM隐马尔科夫模型源码，实现模型的训练(Hidden Markov Model HMM source, the realization of model training)

计算语音、音频信号的信噪比程序（包含音频样本文件）(Calculation of voice, audio program signal to noise ratio (including the audio sample file))

用MATLAB来实现谱减法降噪。有源程序和降噪先后的语音信号，有实验的结果图(Use MATLAB to achieve spectral subtraction noise reduction. There has voice source and noise signals, there are experimental results of Figure)