该图来自:Masataka Goto 2004 年"A real-time music-scene-description system: predominant-F0 estimation for detecting melody and bass lines in real-world audio signals" 一文中的3.1.1章节。
相关描述如下:
At each level of the binary branches, the audio signal is down-sampled by a decimator that consists of an anti-aliasing filter (an FIR lowpass filter (LPF)) and a 1/2 down-sampler. The cut-off frequency of the LPF in each decimator is 0.45 fs, where fs is the sampling rate at that branch.
很困惑 滤波和降采样率的先后顺序?请大佬们教学一下,谢谢了