Since my previous post, I’ve made some great progress with this project. Last week, I laid down all of the rhythm guitar parts, and I’m pretty happy with them. I mostly tracked electric guitars, with a few absolutely great-sounding acoustic guitar parts as well.
After taking this past weekend off to participate in Ptbo Game Jam 02 (a ton of fun, but why oh why did it have to be the same month as the RPM Challenge?) I put in a solid day on Monday, and finished tracking all of the lead vocal takes for my 10 songs. In my mind, those two are the hard parts: there’s still plenty to do, but most of it’s either going to be sequenced, or just less taxing on me as a performer, so the challenges on the horizon are going to be more about creative decision making rather than hard musicianship.
I have to admit: I find very little joy in tracking vocals. I’m no stranger to singing – I’ve been doing it on stage and in the studio for over 10 years – but while I’m confident enough in front of a microphone I’m, objectively speaking, not a great singer. I long ago made peace with how my recorded voice sounds, but all the same this is the part of the recording process that I find the most taxing, and the time when I feel the most vulnerable, for lack of a better word.
I have a decent idea, from past experience, about how to produce my vocal parts so they sound good, and by the end of this last phase of this project, I got into a pretty solid routine for approaching laying down the takes, so for the rest of this post I’ll go over that in more detail.
First things first: mic technique. As I mentioned in my last post, I borrowed a few great microphones from Leadfoot Studio earlier this month, and put them to great use for tracking vocals. I’m a big fan of using an RE20 on voice, largely because it’s one of the mics in Studio A at Trent Radio so I’m very comfortable with getting the sounds I want out of it. However, I like having options, so I chose to put two mics up: the RE20, and a Shure KSM32. That’s not one I know well, but I like it. I was surprised at how similar the two mics actually sound: I did a few blind listening tests of the same take captured with both of them, and while they sound different, I couldn’t identify which was which, nor did I have a clear preference for one or the other.
One of the challenges with a multi-mic setup, especially for vocals, is preventing phase issues: the sound might not reach both mics at the same time, which can cause the two signals to partially cancel each other out. For that reason, Mike (who has a great practical mind for recording, and has been a source of a lot of wisdom during this project) strongly suggested I only use one mic or the other for any given vocal take.
I kind of listened to him, as I’ll explain more later, but I did track all my performances with both microphones. I took some effort to make sure they were the same distance from my mouth, by placing them directly side by side, and as it turned out the signals were just about exactly out of phase. I flipped the phase on one of the tracks in REAPER, and was good to go.
I’ve had this setup, exactly as pictured, going in my office-turned-project studio for the better part of the last week. It’s still up as I write this, but as I get into tracking fuller arrangements I’m sure it will get shuffled around as I lay down different instruments.
Approach to tracking
Like I said above, I find little joy in tracking vocals. It’s very easy to get caught up in the minutiae of your performances, which is in general exactly what you want to do in a recording session, but I find it quickly gets disheartening when it’s vocal-tracking season. To get my performances dialed in as quickly as possible, I find it helpful to sing along to the guide – this is where the scratch tracks I recorded earlier came in handy. I ran those takes through some pretty aggressive autotune (not actually Auto-Tune™, but ReaTune, a cromulent if somewhat minimal alternative included free In REAPER), and time-stretched them with reckless abandon until I had the pitch and phrasing where I wanted them.
With those ready, tracking the vocals was a matter of singing along to those guides as closely as possible. To an extent, this was iterative: after trying a take or two, in a few places I chose to go back and tweak the timing of the scratch vocal because it didn’t seem quite right. More often than not, this involved undoing some of the shifting I had done on the first pass.
While my main goal was to get a take that was as good as possible, you can do lots with the power of editing that modern DAWs provide, so a secondary goal is to get enough material to work with to put together a good edit. In a real studio setting, of course, you’d hope to have someone in the control room who could give you feedback after a performance, but working on my own that’s not as practical. Rather than stopping after each take to give it a listen, which I did at first, I found it more practical to push through and track multiple takes in one go, and then save the critical listening for the editing phase. I found that 5 takes was a good number – some needed more, and some needed less, but five was a good amount to track in one stretch and didn’t get too far into diminishing returns territory.
Editing it together
I’m a big fan of doubled vocals takes. It’s a technique as old as multitrack recording itself, and a great way to flesh out the sound of a vocal performance. Traditionally, you’d track one vocal take, and then record a double of it matching it as closely as possible – but non-linear digital editing is a wonderful thing, so I can get the same result by creatively editing the 5 or so vocal takes I recorded for each song.
The process goes a little something like this:
- Mute the scratch vocal track – we’re not going to need it any more.
- Ensure the two vocal mic tracks are grouped, and put together an edit of the 5 takes that’s going to be the “main” vocal.
- Rearrange the takes, so that edit is in the top lane. (In REAPER, you do this by saving the edits as a comp, and then choosing “move active comp to top lane” from the Comp menu.
- Duplicate the two tracks, and split the grouping – you want the duplicates grouped together, but not grouped with the originals.
- Solo the two vocal tracks, and start putting together an edit for the “double” vocal. Since you’ve moved the “main” vocal to the top lane, that becomes the forbidden zone, but all the other takes are fair game.
I use the words “main” and “double”, but in reality I don’t intend for one to sit above the other in the final mix. It’s just a reminder, for me, of what I’m listening for when I’m doing the edits: for the “main” track, I want to put together an edit that makes for the strongest performance, that could stand well enough as the only vocal track. For the “double”, I’m looking for an edit that complements that first part: it’s more important to match phrasing and pitch as closely as possible, for example, than to put together a take that’s particularly strong on its own.
Like I mentioned above, I “kind of” listened to the advice Mike gave me, that using just one microphone on vocals is a good choice. While it’s good advice for avoiding phase issues, I found that the tone of both mics together was a bit richer than either on their own. After I had these edits together, I compromised by muting one of the two mics on each track – but a different mic on each. That way, the doubled performance captures the tone of each microphone, without any of the risks accorded to double-micing a voice.
Well, that’s about where things sit now. I’ve got 10 tracks, with vocal takes that I’m happy with and solid-sounding rhythm guitars. There’s still tons to do, but the list is getting smaller by the day – which is a good thing, since there’s just about 6 days left.