说明：  进行增强语音频域带噪语音的生成，特征提取，帧扩展，归一化等
(The generation, feature extraction, frame expansion, normalization and other tion_DATA_Preparation of speech with noise in enhanced audio domain were conducted)