This class library contains more than 20 classes including feature extraction algorithms (MFCC, LPCC) and modeling techniques (HMM, GMM, DTW, VQ ) for automatic speech recognition and speaker verification

Currently, there are more than 20 classes in this library, including commonly used feature extraction algorithms and modeling techniques for speech recognition and speaker verification. -Currently, there are more than 20 classes in this library.

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

通过计算机对图像进行去除噪声、增强、复原、分割、提取特征等处理的方法和技术。-Computer image noise removal, enhanced recovery, segmentation, feature extraction processing, such methods and techniques.