字符识别的模式识别

Linux下的CELP语言编码，压缩语音质量极好。支持2K到32K的压缩。实现部分使用C语言编写。可用做Linux下的动态库，有例子。-under Linux CELP language coding, compression excellent speech quality. Support 2 K to 32 K compression. Part of the use of C language. Linux can be done under the DLL, is an example.

用三层小波神经网络实现的与文本无关说话人识别。（识别部分）。输入的是语音特征，输出的是识别结果。训练用的语音特征要事先提取出。-with three wavelet neural network has nothing to do with the text of Speaker Recognition. (Recognition). The admission of voice features, the output is the result of recognition. Training voice features prior to extract.

CCS下调试通过，基于5416DSP及AIC23的G.711编解码算法。-The adoption of CCS under the debugger, based on the 5416DSP and AIC23 codec G.711 algorithm.

语音合成 变速不变调 变调不变速 且每行都有注释 适合学习-The same tune speech synthesis speed transmission and the modulation is not suitable for learning each line has a comment

Peter Howell, Stevie Sackin, and Kazan Glenn Development of a Two-Stage Procedure for the Automatic Recognition of Dysfluencies in the Speech of Children Who Stutter: II. ANN Recognition of Repetitions and Prolongations With Supplied Word Segment Markers J Speech Lang Hear Res, Oct 1997 40: 1085 - 1096 -Peter Howell, Stevie Sackin, and Kazan Glenn Development of a Two-Stage Procedure for the Automatic Recognition of Dysfluencies in the Speech of Children Who Stutter: II. ANN Recognition of Repetitions and Prolongations With Supplied Word Segment Markers J Speech Lang Hear Res, Oct 1997 40: 1085- 1096

AR模形拟合工具，提供AR模型的拟合与模拟-AR model

本课题为给语音识别系统提供语音信号端点检测的功能，对语音信号端点检测进行了探讨和研究,主要是侧重语音区间的端点检测...(附软件源码)-Voice Active Detection (VAD) is discussed in this project. Detection between speech section and speechless section and emphasized.(code available).

语音识别工具，包含源代码-tools of speech recognition ,including source

这是一个xmms的3D音效插件。算法简单，效率很高，如果使用音箱来听效果会更好！-xmms This is a 3D audio plugin. Algorithm is simple, highly efficient, if used to listen to speakers results will be better!

LMS 算法源程序，绘制性能曲线图，误差性能曲线-LMS algorithm source code, rendering the performance curves, the error performance curve