NVIDIA ShadowPlay/Share : Extract audio, remove mic-buzzing, fix AV-desync
This video will show you how to extract the audio from videos recorded via ShadowPlay / Share, how to fix their AV-desync and how to remove the buzzing from your mic.
I hope this helps you, going further down the description you'll find timestamps, links, a written solution in short and a template.
Timestamps :
0:42 - 2:56 extracting audio
command is : ffmpeg -i inputfile -map 0:a:0 -c:a copy audiooutput1 -map 0:a:1 -c:a copy audiooutput2
ATTENTION
Don't use wav/mp3 as filetype like I did, the encoded audio will be shorter than the original. To prevent that I edited this command ( -c:a copy copies the audiocodec ). ShadowPlay uses aac as audiocodec.
3:32 - 3:54 removing buzzing in Premiere Pro
right click the audio, select audiochannels and use the left stream for the right channel
3:54 - 4:08 removing buzzing in Vegas Pro
right click the audio, select channels and left only
4:08 - 4:22 removing buzzing in Audacity
right click the left drop down, split stereo tracks, mute the bad one, make the good one mono
4:28 - end proof the buzzing-removal works
inbetween these are things like disclaimers.
Please keep in mind that you also have the choice to roll back to an older version of GFE, which will most likely fix the mic-buzzing.
Links :
ffmpeg :
https://www.ffmpeg.org/download.html
K-Lite Codec Pack :
https://www.codecguide.com/download_k...
( I don't know which version's enough why I'd suggest downloading the full version )
Template for the batch-script :
https://pastebin.com/0re23DY6
Solutions explained in short for people who know what to do :
To extract the audio, map the audiostreams from the mp4 into multiple files using ffmpeg. ( ffmpeg is not required, it's just what I use )
To remove the mic-buzzing copy the left channel onto the right ( and switch the audio to mono ).
To fix the AV-desync encode your audio and use it. You might need to encode your video aswell, because encoding only the audio doesn't work every time. I'm using ffmpeg for this too.
If that is NOT working, you can encode your audio and video into the same file and then extract the audiolayers from the new file afterwards.
You can do this by repeating the -map command for every stream ( 0:v:0, 0:a:0 and 0:a:1 ) and specifying only one output file after these three commands.
Watch in 1440p for better YouTube encoding quality .}
And sry 4 bad England