If my music is very dynamic, should I run it through a compressor after recording?

Question

I am finishing the recording of a song which starts out with gentle piano, builds up to a fairly "loud" bridge with several more instruments, then gradually goes back to the same arrangement as the intro.

Today I was listening to it in my car, and I noticed that the intro and outro are almost inaudible when there is any kind of background noise. This made me wonder: should I compress the song after having finished recording, so that the more quiet parts will gain some volume? With the more "crowded" arrangement in the middle section, I think the song's dynamics would still be quite good despite the lack in evident volume difference.

If this is not the best idea, what would a better solution be? Is there some trick that people use to solve this problem?

Answer 1

You can compress it, but I don't think you should.

If you are still in the mixing phase, you should be able to increase the amplitude of the piano and any other elements that are too low. Make the quieter parts louder by raising their gain in their respective tracks.

If you can't do this, if you only have access to the mixdown of the track, then try to go back to the mixing phase. It's worth it.

If for some reason you can't do this in the mix, if you can only work with the mixdown, then you can use gain automation to increase the amplitude of the parts that are too quiet.

The difference between the quiet and loud parts is too big, you would need a lot of compression to even things up. It'll be very easy to mess up the dynamics if you don't know what you are doing. If you still want or need to compress, make sure that you like the new dynamics that the peaks will have, make sure you like what you hear.

In that case, maybe it's worth to try parallel compression, using several compressors in parallel. This technique is not harsh with the peaks.

Unless I'm missing something, there shouldn't be a need for compression on this situation, though. You might need compression to handle the dynamics of each element separately (as usual), but there is no reason to correct the quiet/loud issue with compression.

Answer 2

In this particular case, maybe you could just set the gentle piano a bit higher in the mix. Playing style and tone colour define 'soft' and 'loud' almost as much as dB level does. But you needn't be frightened of trying a bit of automated compression. With a moderate ratio and a slow release setting you may get a good result, without turning the louder sections into a squashed-dynamics 'wall of sound'.