2400 bps MELP语音编解码器 定点运算算法

《FUNDAMENTALS OF SPEECH RECOGNITION（语音识别基本原理）》（英文），大家可以好好学习下-"FUNDAMENTALS OF SPEECH RECOGNITION (voice consensus other basic tenets) "(English), you can learn where

可以使用WIN CE的录放音程式码,可以储存录音档，由embedded visual c++ sp4 所开发-WIN CE can use the playback audio code, audio files can be stored by the embedded visual c++ Sp4 developed

intel的ipp性能库的示例代码，在多核平台下可以充分利用多核的性能实现语音编码-intel performance library of ipp sample code, in multi-core platform can take full advantage of multi-core performance to achieve speech coding

为了实现高速语音特征参数的提取，在分析了美尔频率倒谱特征参数提取算法的基础上，提出了算法的硬件 设计方案，介绍了各模块的设计原理。该方案增加了语音激活检测功能，可对语音信号中的噪音帧进行检测，提高了特征参 数的可靠性。-In order to achieve high-speed voice characteristic parameter extraction, in the analysis of Mel frequency cepstral feature extraction algorithm is proposed based on the algorithm the hardware Design, describes the design principles of each module. The program increased the Voice Activity Detection feature, which allows voice signals to detect the noise frame to improve the characteristic parameters The number of reliability.

利用HMM的孤立字(词)语音识别程序,谢谢大家关注-A program of Using HMM s insulation word to recognise speech signal. Thank you for attention

我懂得还很少，但是我所传的是我自己编写的关于用遗传算法实现语音识别的东西，欢迎批评指正-I know very little, but I have of myself prepared on Genetic Algorithms Speech Recognition things, welcome criticism correction

特征分析对于声音信号的处理非常有用，希望这个能对大家有用-Feature Analysis for the audio signal processing is very useful, hope this will be useful to everyone

c++实现HMM，向前向后算法，Viterbi算法，Baum-Welch算法。其中包括用c++定义的HMM数据结构。全部是cpp和h的文件-c achieve HMM, forward backward algorithm, Viterbi algorithm, Baum-Welch algorithm. C including the use of the HMM definition data structure. Cpp all the documents and h

将IBM TTS 和微软的SAPI封装在ACTIVEX中供其他语言调用,例子指写了用WEB网页如何调用的方式-The IBM TTS and Microsoft SAPI packaged in ACTIVEX call in for other languages, for example, refers to the use of WEB pages to write how to call the way