add tool to extract loudest segments from panda zoo recordings

[kgarwoodsdzwa]

We are collecting passive panda acoustic data from 3 recorders at the zoo. We need a tool that tosses out silent portions of the audio so that we're not sifting through hours of data with nothing. Would be ideal if the filename could extrapolate the timestamp from the was it came from (because the wav recordings contain the timestamp) so that we don't need to continuously reference the original sound file. the segments will need a metadata file to go with them.

[kgarwoodsdzwa]

kinda working, was able to notice the pattern of eating times for recorder 1, 7-730 am seems to be when one of them is in there eating, its faint though, and i think the rms calculator is actually picking up a different sound in that time to choose to save the clip, so might want to use it just to determine the feeding times and then directly label the audio from those times. in this case, 7-730am seems to be when a panda is heard eating on SMM1

[kgarwoodsdzwa]

to be fair, i think the eating is triggering the rms sometimes