PyHa
PyHa copied to clipboard
TweetyNet/Microfaune sometimes have differing file lengths after output
Are they off by a lot? Do you have a screenshot?
My best guess as to why this is happening is that for the calculations when we go from the processed local score array outputs of Microfaune and Tweetynet respectively, a ratio between the length of the clip and the length of the time-axis of the spectrogram is used as an approximation. Since the input spectrograms of Microfaune and Tweetynet are not the same, they will be a bit different.