声音的特征分类在新能源汽车主动发声设计中的应用

Application of Sound Feature Classification in Active Sound Design of New Energy Vehicles

包括电动车在内的新能源车日益普及,由于缺乏了传统发动机噪声的掩蔽效应,其他噪声在车内凸显,在安静的车内和车外增加车辆主动发声尤为必要。在设计主动发声音效的过程中,为了对大量的音效进行分类,以便有更加清晰的自定义选择音效过程,利用梅尔频率倒普系数(MFCC)作为特征值,以支持向量机(SVM)作为分类器,对大量音效进行分析并对其分类。研究表明,该方法对音效分类有较好的效果,对后续新能源汽车音效设计有非常好的借鉴意义。

New energy vehicles,including electric vehicles,are becoming increasingly popular.Due to the lack of masking effects of traditional engine noise,other noises are prominent in the vehicle,and it is particularly necessary to increase the active sound in internal and external of the quiet vehicles.In the process of designing active sound effects,in order to classify a large number of sound effects for a clearer customized sound selection process,the Mel-Frequency Cepstral Coefficients(MFCC)is used as the feature value,and the Support Vector Machine(SVM)is used as a classifier to analyze and classify a large number of sound effects.The research shows that this method has a good effect on the classification of sound effects,and it is a good reference for the subsequent design of sound effects for new energy vehicles.