Alcohol or late night delirium if no one else hears it, too.
One person on multiple nights is a repeated measure from a sample size of one. Not particularly conclusive.
OP, I’m not saying the phenomenon you describe isn’t possible; I’m saying it’s a sort of thing that’s extremely easy for people to assume whether real or not.
There is some evidence out there to suggest even very small (real) changes to the noise floor might be enough to trigger a “different/better/worse” perception from some folks. But add alcohol to that and I stay skeptical at most.