It still requires a lot of manual work, such as cutting the tracks, and making sure there are no "skips" (which somehow can happen). Skips could theoretically be removed using a consensus algorithm (using multiple recordings).
I just wish someone would develop a fully automated workflow for converting playlists to audio files.
I'm not familiar with the 'parec' command. Are you able to filter out other audio streams (e.g. Slack alerts or accidentally opening a page with an auto-play video)? I was under the impression that one of the big advantages of Pulse Audio was its ability to separate multiple streams.
Interestingly enough, I had similar scripts about 15 years ago pre-Napster. The earliest mp3 sharing sites tended to push full-albums instead of breaking things up by track. I had a lot of fun using some of the earlier mp3 tools to break up and tag tracks. I still have a lot those mp3s on various HDs, and I know it because my splits weren't perfect for certain tracks that don't have 2 second gaps.
In theory PulseAudio should be able to separate the streams, or at least turn off audio for selected applications. However, this was never really a problem anyway because I usually ran the command at night :)
Yes, audio processing in the early days was fun, though it's easier now because of better tools and especially bigger harddrives and faster CPUs :)
I just wish someone would develop a fully automated workflow for converting playlists to audio files.