And for the curious: this is my work laptop with Windows installed. Most days I just M-x eshell and cd to the correct directory where I run “mpg123 *.mp3” (this is an old mpg123 version 1.24.0)

When I feel more up to it, I open wsl (which is where postgres and dictd are running) and use cmus. There, however, I often run into an issue. I think the problem is that when cmus dies for some reason, the audio bridge (?) isn’t freed so from then on there is no more sound from wsl. And I’m back to mpg123 from the command line.