Comments

Former user wrote on 9/29/2023, 9:45 PM

The AI versions are all subscription payware that I've seen. Here's a non-AI command line python variant

https://github.com/WyattBlue/auto-editor

and this is an out of date command line windows version (no python needed)

https://github.com/petermg/auto-editor/releases

There was something I didn't like about the windows version, only used it a couple of times, have not used python version.

I use Resolve Studios's built in silence detector now. it's related to automatic subtitles, so is detecting actual voices rather than measuring changes in loudness the way 'audio-editor' does. Magix/Vegas themselves could add this function as part of their automatic subtitles, problem is it's a subscription only feature. Both methods have a potential to fail, Resolve may not be able to detect a voice if low enough or there is loud background noise, and 'audio-editor' is only detecting changes in audio levels afaik.

Also I use the beta versions of capcut(free), and it had an AI silence removal option but is now gone, it may be available in the paid version, but it's subscription software so would not be economical if you only had one use for it.

 

Robert Johnston wrote on 9/29/2023, 10:24 PM

@andy-0 Are you concerned with only the audio portion and don't care about slicing up video? Sound Forge Audio Studio 17 has that, assuming you purchased Vegas Pro 21 Suite, then you already have it.

Intel Core i7 10700K CPU @ 3.80GHz (to 4.65GHz), NVIDIA GeForce RTX 2060 SUPER 8GBytes. Memory 32 GBytes DDR4. Also Intel UHD Graphics 630. Mainboard: Dell Inc. PCI-Express 3.0 (8.0 GT/s) Comet Lake. Bench CPU Multi Thread: 5500.5 per CPU-Z.

Vegas Pro 21.0 (Build 108) with Mocha Vegas

Windows 11 not pro

jetdv wrote on 9/30/2023, 7:30 AM

Part 1 of a six part series:

Please go through all six parts and you can see the techniques used here to find the "quiet" areas. You can also adjust the parameters to better fit what is actual "silent" in your case - or add a slight buffer before and/or after silence is detected. It's all straight code so you can adjust it to work however you like.