Musical synchronicity ::

I’ve been transferring, restoring, and remastering some ancient recordings. There have been a number of reminders of how this process is just postponing entropy: tapes with dropouts, bands that never quite played a piece as tightly as I might have liked, having little idea where the sequencer files for the pieces might be, sequences using synthesizers I don’t have any more... For that matter, all the sequences are in Studio Vision, and there could come a day when I don’t have a way to open them.

I can live with that; what’s finished is finished, recorded as well as it will likely ever need to be. And I can always make more.

Still, it’s nice to have a couple of pleasant surprises along the way, both aesthetic and technological. “That’s beautiful, I can’t believe I wrote that.” One DAT recording had a nasty digital dropout near the end. Fortunately I’d recorded a previous mix, which was quite different—but at the spot in question it was sufficiently similar to make an almost seamless digital splice.

Fitting the old vocal track (bottom) to the new backing tracks

The kicker was a piece I wrote in 1979. In 1981 I recorded it to cassette in one pass, with electric piano on the left channel and the vocal on the right channel. In 2000 I did a version with MIDI and digital audio tracks in Studio Vision. On the 1981 recording, I’d transposed the song down a fourth—much better for my vocal range. The vocal performance was decent (well, about as well as I ever sing at least!); but the poor old Fender Rhodes was not sounding its best and the piano playing was not smooth. On the 2000 recording, the instrument tracks are fine (and easily edited should I get the urge to improve them), but it was in the original higher key, just too high for my voice.

I wondered how much work it would be to take the 2000 arrangement, transpose the MIDI instruments down a fourth, and drop in the 1981 vocal. It turns out, the tempos are darn near close enough (and the vocal phrases few and sparse enough) that I think it’s going to work—haven’t done any digital time-stretching yet...

Fixing a plosive

I also got to try some techniques that came up on the Opcode Users list this week. Some consonants (p’s, g’s, d’s, etc.), when sung into a mic with a large proximity effect, come out way too loud. Doing a steep low-shelving cutoff of everything below around 150 Hz (the lowest fundamental in this piece) helped a lot, but some of those consonants still come out as little explosions, which is why they’re called “plosives,” I guess. Some careful use of volume automation does the trick without having to get into the mess of destructively editing the track.