它主要被用来解决声音分离、分割、声音事件和声音源的刻划及表征、听觉场景分析和音乐分析等问题。
It aims to resolve problems such as sound segregation, segmentation, sound event and source characterization, auditory scene analysis and music analysis, etc.
并将听觉相关图中每个频带的频率分量幅值进行特征编码作为表征当前频带特性的特征向量。
Then the every frequency amplitude of auditory correlogram was coded as a feature vector expressing present frequency band characteristic.
应用推荐