https://github.com/slhck/ffmpeg-normalize
Haven't used this, I just googled for ffmpeg audio normalization. Not sure if it works for just audio, but the general idea is there.
This looks promising!
They should really enable on-the-fly compression and normalization for every podcast in the player itself. Some podcasts are horrible to listen to not just because they are too quiet but because they are quiet 99% of the time and then way louder 1% of thr time.
ffmpeg can do all of this via various incantations but I don't know how much effort you want to put into your pipeline.
The tools to do it are buried in the ffmpeg incantation here: https://wiki.tonytascioglu.com/scripts/ffmpeg/dynamic_range_compression_dynaudnorm_compand
Interesting, I use audio book shelf, I wonder if I could modify it to use these settings.
That would be cool! If you find a way please let me know!
given your use case the ffmpeg incantation wins out but just for posterity I should add that if you are producing a podcast or other audio medium you can do this in Audacity as well pretty simply
Podcast recommendations, episode discussions, and struggle sessions about which shows need to be cancelled.
Rest In Power, Michael Brooks.