基于启发式掩模EMD的音频突变成分检测方法

Audio Mutant Component Detection Method Based on Heuristic EMD with a Masking Signal

多数情况下,音频信号可以视为是由稳态成分和突变成分两种成分组成,稳态成分与突变成分在属性特征方面具有明显的差异,而且突变成分通常承载更重要的信息,是信号处理要分析提取的目标。为有效检测出突变成分并将这两种成分分离,需要完整精确地检测提取出突变成分。为此本文提出一种基于启发式掩模经验模态分解(Empirical Mode Decomposition,EMD)的音频突变成分检测方法,该检测方法中使用启发式掩模EMD分解音频信号并从中提取出各点的瞬时信息作为检测特征,同时本文提出一种窗长自适应更新策略来设定适合突变成分的长度。在IOWA的乐器音频数据集中,该检测方法能够实现以98.68%的检测精确率和87.65%的检测召回率将音频突变成分检测出来。此外该检测方法无需人为干预,并且检测出的突变成分维度一致,便于进行后续的特征提取、分类识别等处理操作。

Under many conditions,audio signals could be consisted of composed of steady components and mutant components these two components and there was a clear difference between steady components and mutant components on features.The mutant components usually carried more important information,which was the target of signal processing to analyze.To detect mutant components and separate these two components effectively,it was necessary to accurately detect and extract the mutant components.Thus,the paper proposed an audio mutant component detection method based on heuristic EMD with a masking signal.In this detection method,heuristic EMD with a masking signal was used to decompose the audio signal and extract the instantaneous information of each point as the detection feature,and meanwhile a window adaptive updating strategy was proposed in this paper to set the suitable lengths of mutant components.In IOWA’s instrument audio dataset,this detection method could detect audio mutant components with a detection accuracy of 98.68%and a detection recall of 87.65%.What’s more,the detection method does not require human intervention and the detected mutant components have the same dimensions,which is convenient to perform subsequent processes such as feature extraction,classification and so on.