1

i am trying to find the energy of list of frames. And I want to classify them into four criteria. For example "weak, normal, strong, too strong" and make a mixed array like 

normal_frames_array = [
   6, 34, 52, 71, 88, 107, 125, 143, 161, 198, 216, 234, 271, 289,
   352, 397, 416, 434, 461, 479, 498, 516, 552, 571, 589, 644, 662, 716,
   .
   .
   .
   8204, 8222, 8241, 8259, 8277, 8295, 8313, 8332, 8350, 8368, 8386, 8404, 8423, 8441
]

classified_array = [
   [6, "weak"],
   [34, "normal"],
   [52, "strong"],
   [71, "very strong"],
   .
   .
   .
   [8167, "weak"]
]

How can i do this in librosa ?

Zekeriya Akgül
  • 308
  • 4
  • 14
  • First you need to define what *a frame* is. What hop size, what sampling rate etc. Use [rms](https://librosa.github.io/librosa/0.7.2/generated/librosa.feature.rms.html) to compute energy for each frame. If you know you want to classify into 4 classes, you could sort all frame-number/energy pairs and then split into 4/4. Is this a class assignment? – Hendrik Jan 23 '20 at 10:41

0 Answers0