Putting different audio tracks into a playlist has the problem of different loudness. So, the practice is to "Normalize" the tracks so they play at the same volume.
Does normalizing the tracks damage the audio quality of the music? If so, is there a way to batch level the tracks?
When normalization is done correctly there will be no alternation in the audio quality of the music.
Tidal uses "Album Normalization" which means that the loudest track on the album dictates how much the volume is lowered or raised for all the rest of the songs to keep the loudness levels intact on a song-to-song basis. So if the loudest song on the album has a LUFS level of -10 dB, all the songs on the album will be lowered in level by 4 dB to reach the normalization level of -14 LUFS plus a headroom of 1dB under digital zero. Usually, the level is lowered, but if it is a quiet album, a positive gain can be applied but just so that the loudest track reaches -14 LUFS or until the loudest peak reaches -1 dB, so there will never be any clipping or compression going on.
They use "Album Normalization" even when someone is listening to a playlist, as they have concluded in testing with many users that if a certain level works between different songs on an album, the same level will also usually work well between different songs from different albums put together in a playlist.
Spotify on the other hand uses three different normalization levels. The one they call "loud" does affect the audio quality as a limiter kicks in which will clip the peaks of the audio tracks so they reach the normalization level.
I'm not sure, but I think Apple uses a similar normalization as Tidal does, so I think they only use positive gain until the peak of the track reaches a certain level with headroom under digital zero. If a track has lots of dynamics, it will still be played more quietly than other tracks, which is the case with Tidal as well.
