I’ve been procrastinating on recording vocals because so far, it’s been nothing but an exercise in frustration. But it took recording someone else singing for me to reach a few epiphanies about how it’s done.
About a week ago, OmarG asked whether he could use my studio to record some parody tracks for a Latino Comedy Project show. I gave him some disclaimers about my ability to record vocals, but he was looking for cheap, and I was free. So we took a stab.
We had a 3-minute karaoke track that had to be cut down to fit some substitute lyrics. I didn’t know if it could be done, let alone whether I could do it. But with some creative splicing and cross-fading, it was done.
Then came the vocals.
I’ve read up on how to record vocals, and I know the general concepts required. But every time I’ve attempted to do it on my own, I get really crappy results. This time, I didn’t have to do both the singing and the knob twiddling — I could just concentrate on the knob twiddling. I also read up on how Cakewalk SONAR works with layers, something I’ve had no use for till that point.
Thus informed, I tackled the recording session and came up with results that really sounded better than either of us expected. In short, I was doing stuff I didn’t know I actually knew how to do.
So this past weekend, I tried it on my own, keeping the setup from OmarG’s session. I recorded a cover of AJICO’s 「波動」 ("Hadou"), using the same techniques.
When I didn’t like something, I focused on that one line and recorded additional takes, till I got something I did like. On one word, I came in early. So I went into the editor and slid that portion of the recording over a bit to be on time. I even recorded backing vocals the same way.
What resulted was the first vocal performance I was actually satisfied with. That’s saying alot since I really don’t like my own singing voice. What does it sound like? Listen.
These results are encouraging. In the past, I would have to break down the mic setup when I was done, and from day to day, the slight changes in mic placement would yield different results, which would annoy me. Now with a room dedicated to studio work, I can record fairly consistent performances, sound-wise.
Now that I understand how layers work in SONAR, I can work a lot more efficiently. On my first tries I would scrap entire takes if something went wrong, even if the rest of the performance sounded good. (How wasteful.)
With every project I do, I’m learning just a little bit more each time. Hell, I have to resist temptation to go back to stuff I did in 2005 and remix and remaster it.