Is there a reason to have yt-dlp convert extracted audio to .wav instead of just leaving the extracted audio in its original .m4a(a tiny fraction of YouTube content) or .opus(the vast majority of YouTube content) formats? I mostly use yt-dlp for batch downloading entire channels for offline listening on my Senseplayer, and other than having to rename .opus files to .aac, have yet to have a problem with playing downloaded files since I added the option to extract audio to my
~/.config/yt-dlp/config
file, and I would expect mpv to be less picky about file extensions. Converting to .wav just seems like a waste of disc space(perhaps no an issue for individual files you plan to listen and delete, but adds up if batch downloading from a newly discovered creator with hundreds of videos in their backlog), and conversion to an obsolete lossy format for archival just seems bizarre.
And incase anyone finds it useful, here's the contents of my ~/.config/yt-dlp/config:
-o "%(title)s.%(ext)s" -x --cookies-from-browser firefox
the -o and what follows prompts yt-dlp to save downloads with the title of the video and the file extension, the default behavior gums up the filenames with the gibberish Google uses to give each video a unique ID. the -x tells yt-dlp to extract audio without doing further conversion or saving the original video file, which both greatly reduces disk space requirements when downloading whole channels and saves the hassle of manually extracting audio afterward, and the bit about cookies cuts down on downloads failing due to Google's anti-bot measures. And having all of this in the config file means I can just invoke
yt-dlp [url]
at the command line.
Though, if anyone wants to suggest other useful things to put in my config file, I would love to hear them.