Sep 6, 2018 · 1 min read
Vivek, just read your article, really enjoyed the depth of your insights on matrix representation of music. I wish I knew about pychorus earlier; I used msaf — also librosa based — but the cluster segmentation wasn’t particularly great. The labels of musical clusters provided by the software was off most of the time so I had to manually create labels for my own generated music to emulate. I’m really curious about the time lag matrix. Could you please help me better understand the concept? And how to code it?
