老师布置的一个project，里面有题目和源程序，能够识别十个数字。

its an lpc code working for speech recognition

G.729aa 编码器和解码器代码 G.729A 是 G.729 的降低的复杂性标准之一。G.729 是国际电信联盟-T 语音压缩算法，基于共轭结构化代数代码激动的线性预测 (CS-ACELP）。G.729A 压缩 8 kHz 线性音频信号和对他们进行传输在 8 Kbps.输入进行编码为 16 位 PCM。输入码流是分割成 10ms年帧和处理。预测分析做使用合成的筛选器的顺序 10。

用VC实现的语音库的建立的程序. 首先建立单音节语音库,然后将所有的单音节语音文件合成一个大的.wav文件,并且制定索引机制,便于快捷查找-VC realize voice with the establishment of library procedures. First of all, the establishment of monosyllabic speech database, and then all of the monosyllabic voice synthesis of a large document. Wav files, and to develop indexing mechanisms to facilitate the quick search

静态Huffman的小程序，帮助理解Huffman的原理，不可多得。-small procedure to help understand the principle Huffman indeed.

2400 bps MELP语音编解码定点算法-2400 bps MELP speech codec fixed-point algorithm

Bark和ERB滤波器组实现的Matlab源程序-Bark and ERB filter banks realize the Matlab source

实现了语音回放功能，很不错的-Realized the voice playback, it is good

美国进口最新语音识别芯片,识别率90％,在身份识别,玩具,控制方面应用广泛-imports to the United States the latest speech recognition chips, 90% recognition rate, the identification, toys, extensive application of control

C语言开发工具包 （包括制表键宽度的变化，图形图像的处理，字符串的处理等等）-C-language development tools (including tabulation bond width changes, graphic image processing, string handling, etc.)

关于情感识别方面的论文，主要讲了基于hmm和ann的识别方法-On the emotion identification papers, the main speaker hmm-based identification methods and ann