资源列表
VoiceActivityDetection
- 在本文中,主要讲了在语音识别和语音合成之前我们所要做的主要工作，包括去噪,预加重,端点检测,特征参数提取等技术.-In this article, the main speaker in the speech recognition and speech synthesis to be done before we have major work, including de-noising, pre-emphasis, endpoint detection, feature extraction
huazheshibie
- matlab写的话者识别的程序，界面简单大方，程序绝对好用，能学习人的声音并且识别，进行话者识别和分析-speaker recognition program matlab write, the interface is simple and elegant, absolutely easy to use program, learning human voice and recognition, speaker identification and analysis
应用模式识别技术导论——人脸识别与语音识别
- 应用模式识别技术导论——人脸识别与语音识别-Introduction pattern recognition technology -- face recognition and voice recognition
ICAcrosstalkerror
- 我自己编的串音误差计算函数，对公式进行了一定化简，比本网站其他同类计算函数运算速度要快。用于盲源分离的质量评价，另外还包含ICA中有关性能评价指标的文章。-My own series of crosstalk error calculation function, to the formula for a certain simplification, other than this site similar calculation function operation faster. Qual
HTK.rar
- HTK开发工具包,是编译好的可执行程序，下载后可以直接使用,HTK_tool
sy3
- （1）录制语音文件 利用GoldWave或NGwave录音软件录制单字发音语音文件，以wav音频格式保存，共录制两组，一组用于计算参考模板，另一组作为测试模板用于语音识别。 录制语音时，为了能够在主程序中循环读入语音文件，文件名采用数字顺序命名。记录语音文件以及所对应的发音。记于表3中。 另外，录制语音时，注意调整合适的Mic音量，避免语音波形幅度过小，难于同噪声区分开。同时也要避免波形幅度过大，造成波形失真。 （2）调试端点检测程序 结合流程图读懂Matlab程序，先单独调
HTKresearch
- 基于HTK的维吾尔语连续数字语音识别研究.nh-htk research about recognising
yuyinshibie
- 语音信号的时频特性 语音信号的产生模型 语音信号的短时分析技术 语音信号线性预测分析 语音识别-Time-frequency characteristics of speech signal generation model of speech signal short-time speech signal analysis of linear prediction analysis of speech signals speech recognition
sanhuikaifa
- 三汇语音卡程序员手册-3 Department of Voice Card Programmer manual
PicthDetection
- 基音检测，通过检测最大值以及零值的位置来实现，同时以图的形式展现.-Pitch detection, by calculating the position of maxium and zero, and then showing the results as figures.
10-2---Morphological-Image-Processing
- morphological operation in image processing
YUYIN
- 语音开发资料包，讲解语音采集、存储、回放等功能要求-Functional requirements of the speech development information package to explain the speech acquisition, storage, playback, etc.