资源列表
record_audio
- 用C语言实现录音，可以设定采样频率和量化位数-Recording with C language, you can set the sampling frequency and quantization bits
FUNDAMETALSOFSPEECHRECOGNITION
- 英语语音识别参考文档。 英语语音识别参考文档。-English speech recognition reference documents. English speech recognition reference documentation.
mfctest
- 基于MFC的窗口语音识别界面控制程序欢迎大家下载-MFC-based speech recognition interface, the window control program are welcome to download
mfcc
- 用MATLAB分别编写了通过MFCC,EDP,LPC处理后的语音信号-Sound signal is treated by MFCC,EDP,LPC and show the final result.
sphinxbase-0.1.tar
- sphinxbase是卡内基梅隆大学著名的sphinx语音识别工程的公用库，里面实现了语音的AD处理，端点检测，去噪等功能。-sphinxbase Carnegie Mellon University is famous sphinx voice recognition works in the common the inside of the voice of the AD, endpoint detection, denoising function.
julius-4.3.1.tar
- Julius 是一种高性能，两通大词汇量连续语音识别（LVCSR）语音相关的研究和开发的解码器软件。基于字的N-gram和上下文相关的HMM模型，它可以进行几乎实时实时解码目前大多数电脑在60K字听写任务。完全纳入，如树的N-gram词汇，保，跨词的上下文依赖处理，包围梁搜索，高斯修剪，高斯的选择，除了搜索效率等各大搜索技术，它也是模块化小心从模型结构独立，如共享状态triphones的和并列混合模型与任意数量的混合物，州或手机，支持各种HMM的类型。采用标准格式，以配合HTK的，债务工具中央结
Flow
- Optical Flow is a powerful method to a number of applications in Digital Image Processing
HTKbook
- HTKBOOK 语音识别必看，介绍语音模型建立，语言模型建立-HTKBOOK must-see speech recognition, speech model introduced, the language model
PythonHTK
- Document for Python programming in HTK
基于矢量量化（VQ）的说话人识别实验
- 基于矢量量化（VQ）的说话人识别实验MATLAB源程序(Speaker Recognition Experiment MATLAB Source Program Based on Vector Quantization (VQ))
基于动态时间规整（DTW）的孤立字语音识别实验
- 基于动态时间规整（DTW）的孤立字语音识别MATLAB程序(Isolated word speech recognition MATLAB program based on dynamic time warping (DTW))
CZTtransformer
- CZT运算是在z域以任意螺线采样，源码采用512点做CZT运算求出频域最大值-CZT operation is in the Z domain to arbitrary spiral sampling, the source code using 512 points to do CZT operations to calculate the maximum frequency