Adobe has a tool that’s free to use in beta. It enhances audio significantly by removing background noise. It part of their podcast tools platform, but I forget what it’s called.

There is another project that does that same thing that can run locally called “mayavoz”, but it’s not as good as Adobe and a bit difficult to get running as the sample code on their GitHub, is missing some text that saves the file after it’s processed.

I forget exactly what text is missing, but the creator of the app shared it with me on Reddit. I guess he never updated his GitHub.