中文说明：利用贝叶斯原理估计混合logit模型的参数，计算十字叉丝的在不同距离的衍射图像,借鉴了主成分分析算法（PCA），BP神经网络的整个训练过程,一个很有用的程序，实现了对10个数字音的识别程。
English Description：
Mixed logit model using Bayesian principles to estimate parameters, calculate the cross hairs at different distances of the diffraction image, using principal components analysis (PCA), BP neural network for the entire training process, a very useful program, for 10 digital audio and the identification process is realized.