Isolationism

As I’ve mentioned in this space before, I use a software package called Boodler to create environmental audio for entertainment purposes. Although I make use of it on a daily basis I haven’t actually done any soundscape development for a couple months or so; our newborn son took care of a lot of the time, but I needed a bit of a break as well—I had been working on some rather complex stuff to get different Boodler agents to be able to communicate with one another in real-time to be able to perform fancy scheduling tricks (like having crickets suddenly go quiet at the sound of a loud thunderclap).

Right now the only environment I’ve made that could be described as being “complete” is one that generates semi-random thunderstorms that come and go (but I’ll never really finish working on it). I say “semi-random” because while a number of variables are random numbers (pseudo-random, actually, but the distinction is trivial in this case), there is still a lot of structure to the file—after all, it wouldn’t make sense to just plink storm-related sound effects at random; we’d have just put a cat on the keyboard and run a keystroke logger if we thought that’d do the trick. It could also be more random still—different types of rainfall, strength of electrical activity and all that—I just haven’t got around to it yet. It seems strange that making something behave quasi-randomly should be more difficult than having it be precisely the same every time, but that’s the nature of working with computers I guess.

Anyway, this “thunderstorm generator” is employed most frequently (and heavily) overnight at our house, from when we go to bed until we wake—like most people, we tend to sleep more during rainy weather than under clear skies (for whatever reason). My original justification for listening to storms all night long was chiefly for comfort—a sort of aural chicken soup, if you will. A very short while later we discovered that it had other interesting properties, too—chiefly that of masking other environmental noise, such road and air traffic, telephones, people at the door, and so on.

It occurred to me then that environmental audio—especially this particular category of sound effect (which has many of the properties of so-called white noise)—could be potentially useful as a sort of aural camouflage around the house. With my wife and I expecting our first child, it might be helpful if he became accustomed to the sound of rainfall so that we could use it as a smokescreen against potentially disruptive sounds while he slept.

This contradicts the conventional practice of being as quiet around a sleeping baby as possible. It is of course very common for parents to play or sing music to their child to get them to sleep. It’s also not unusual for cribs and playpens (like the one very generously gifted to us by friends Jay and Bevie) to feature their own entertainment devices which play a handful of tunes and a few low-fi looping nature sounds. However, none of these compare to the sound of a full-strength thunderstorm saturating the bedroom (including loud peals of thunder and heavy rainfall), and despite anyone’s best attempts at controlling outside sounds, phones and doorbells will still ring, airplanes and helicopters will still pass overhead, and sirens will still carry.

Amazingly, our little squeaker manages to sleep through it all, seemingly without regard for the playback volume—of course, he’s still only just a little over two months old, so I acknowledge that his noise tolerance may diminish as he gets older.

As I have been heading to bed later than Bex (and much later than my son) in the past couple weeks, I am in the habit of switching on the sound effects remotely before everyone else goes to bed, to provide cover for my eventual entry as I stumble my way to the far side of the bed. Last night I had the misfortune of deciding to head upstairs just as a storm was at its weakest phase, which features only the sounds of the occasional distant rumble and leaves that rustle as though in gusts of wind; it turned out not to matter much as the baby was already awake as he was having difficulty falling (or staying, I’m not sure which) asleep.

Another thought struck me, though. I’ve been using these generated sounds as passive camouflage, using the sound at moment (for better or for worse) to my advantage, or else waiting it out to see if things improve—which could mean a long wait indeed as some storms last as long as 75 minutes, and it’s difficult to tell without careful listening for some time as to whether the storm is inrcreasing or decreasing in intensity (toward its conclusion and near-silence). A much more intelligent approach would be to assert control and leave hooks for direct instructions—such as altering the course of a storm regardless of its current state to, for example, reach a crescendo in a precisely defined amount of time (say, 5 minutes—so the buildup isn’t too sudden, but I don’t have to plan when I go to bed 30 minutes ahead of time, either).

The road is paved with thorns, though: Working on the code for the storms has become considerably more difficult as the various parts grow in size and complexity, requiring refactoring to make them easier to read and modify. To work on anything requires sufficient time, concentration, and silence; I’d have a better chance of winning the lottery than having all three occur simultaneously for the next 18+ years.

My other annoyance is the baby monitor. Specifically, the condenser mic on the base station has such high gain that it picks up every pin-drop in the room so that even the quietest whisper of rain in the bedroom sounds like static on the receiver. I have yet to come up with a cunning plan for that particular problem.