automatic editing using audio levels

[spudwa@gmail.com]

I have some skydiving video recorded using a GoPro and I would like to automatically edit based on audio volume.
The scenario is.

1. Switch GoPro on in plane 20-30 seconds before jumping -> quiet, delete this section
2. Exit plane and skydive -> loud, save this section
3. Open parachute, land and switch GoPro off -> quiet, delete this section

Anyone know of software to do this? I think the correct term is a "noise gate"

Thanks

[smrpix]

It's not possible, and I can't imagine any way to make it possible that isn't more difficult than simply manually trimming the heads and tails of your clip.

[spudwa@gmail.com]

Thanks for a quick and definite answer. Guess I'll just have to write something myself

[pandy]

You can use ffmpeg in 2 steps, first perform audio level analysis and export this to log file later based on log file create ffmpeg trim sequence to extract useful fragments.
Probably need more advanced scripting tool (python?) but perhaps possible under plain shell.

[jagabo]

You can do it in AviSynth too. But for one (or a few) video it's much easier to do it by hand in an editor.

[spudwa@gmail.com]

The reason I want to do it automatically is that there will be 100-150 clips a day to trim. I should of mentioned earlier that I need a Windows and Mac solution

[smrpix]

Sure, avisynth and ffmpeg methods can be used, but it takes just a few seconds after linking to the clip to manually identify the head and tail visually and mark the trims. Then GoPro's tools can batch export.

You would spend as much time confidence-checking the scripted outputs for false positives as you would just doing it manually in the first place.