对不同特征进行融合，得到更好的语音特征，识别率高。

2400bps MELP语音编解码源程序。-2400 MELP voice codec source.

不错的FIR滤波器源代码！-good FIR filter source code!

利用HMM的语音识别代码，里面包含向前向后算法，重估baum算法，viterbi算法等，用C++编写的

用于语音识别的离散隐马尔可夫模型程序（非特定人，多个观测序列），已调试通过。-used discrete speech recognition Hidden Markov Model (non-specific, a number of observation sequence), has been through testing.

语音技术（VOIP）中常用的编码解码器（codess）源码-voice technology (VoIP), commonly used codec (codess) FOSS

这是城市坐标的程序. 注释:这是本人自己编写的.不足之处请各位大哥大姐多多指教,另外请注意版权问题.-This the city coordinates procedures. Note : This is my own prepared. Deficiencies Department asked Big Brother Big Sister to the exhibitions, and to copyright issues.

赤木正人教授在论文“Development of an F0 control model based on F0 dynamic characteristics for singing-voice synthesis”中提出的f0 overshoot模型的matlab实现-Professor Masato Akagi paper "Development of an F0 control model based on F0 dynamic characteristics for singing-voice synthesis" presented f0 overshoot model matlab implementation

OPNET bank_net-scalar

Speech Recognition system using HMM...includes Viterbi , Forwarding and Backtracking Algorithm

语音识别知识讲稿，了解语音识别基本知识和基本方法的捷径-Speech Recognition knowledge speech, speech recognition to understand the basic knowledge and basic methods of shortcuts