Songs and Sonics

“In a lot of ways, everything is sort of judged by the voice.” -Kevin Barnes from Of Montreal

Want to capture better-sounding vocal performances in the studio? (OK, that’s a dumb question. Who doesn’t?)

In this article, I’ll share three tricks I’ve discovered for singing solid lead and background vocal overdubs. These tricks use simple psychoacoustics to coax the best performance out of a vocalist.

Note that I’m not going to talk about vocal technique, microphone placement, or anything like that. I assume you already have a basic understanding of these things. In fact, I’m going to assume all of the following:

You’re working with a singer who can generally sing in tune, and that singer might very well be you.

You’re capable of capturing decent-sounding vocals onto your tape recorder or DAW.

You’re serious enough about recording vocals to own a pop filter.

You have a hardware mixer.

You know what an aux send is and you’re not afraid to use it.

You despise AutoTune and would rather capture a good performance in front of the mic instead of manufacturing one at a later date.

You don’t have to be Stevie Wonder to use these tricks. In fact, these techniques work best with “average” singers.

I’ll be the first to admit I’m no Stevie Wonder. I’ve used these tricks on just about every Simple Carnival song with (I think) pretty good results. Try them on your voice to see if it doesn’t put your vocal tracks in a better light than usual.

This trick is for lead vocals.

I won’t go into the specific implementation of this trick, as it will depend on the gear you own. But this is where you want to go.

First, get a mono mix happening for all of the instruments in the song. That way, when you listen to it through headphones, it’s happening in “the middle of your head” and sounds incredibly bland. Even the reverbs and chorusing effects in the song need to be in mono.

Sing into the mic, but find a way to split the signal. Take one feed and send that into your tape recorder/DAW. Take the other feed and find a way to put big, spacious stereo reverb on it — way more reverb than you’d ever use in the final mix, but not so much that you can’t comprehend what’s going on.

Don’t have any noticeable predelay on the reverb. This rules out using most native DAW software-based reverbs, which typically add some latency to a live signal. So in most setups, the reverb should come from an external hardware unit.

No matter where you get your reverb from, be careful that it’s not adding a subtle chorus-like effect, as that will give the singer slightly incorrect pitch information.

This trick does a couple of things. One, because your instrument mix is in mono, you’re not distracted by mix elements that are panned in the stereo field. This allows you to hear the reverb clearly, because it is very much in stereo.

Two, because you have a long reverb, you can listen to the end of your phrases when you sing and determine how “on” you were. This is important.

I’ve found that if a vocalist’s pitch is a bit off at the beginning or in the middle of a phrase, the listener will usually overlook that. But if the vocalist is flat at the end of a phrase…. ugh. The listener then perceives that phrase as being out of tune. (Disclaimers: This is dependent on the style of music you’re recording, it doesn’t apply if you’re Bob Dylan, etc.)

This trick is for background vocals. Again, I won’t go into the specific implementation, just where you need to end up.

First, get the mono mix happening for all of the instruments on your song, just like Trick #1.

Now record your first background vocal exactly like how you would record a lead vocal using Trick #1.

After you get a performance you’re happy with, take that first background vocal and pan it all the way to the left. (Or right, it doesn’t matter.) Don’t put any reverb on it. Have it up loud enough so you can anticipate your breaths and syllables.

Record your next vocal, using Trick #1 (heavy stereo reverb on the vocal you’re singing).

You can continue stacking duplicate background voices this way. I typically use three voices for each part of a harmony, because three voices are the minimum for masking slight intonation problems.

Why three?

It’s the same reason orchestral arrangers tend to use three or more violins to play a single melodic line. Two well-performed violins playing the same note will often create a flanging sound; three or more and your ear becomes “confused” in a good way. Your ear only perceives a pleasant, natural “chorusing” sound. (Not that two violins/voices aren’t useful — they are! But it’s a different sound. I’ll often record lead vocals with just two voices.)

Don’t take my word for it — try it for yourself and see which your ear prefers for background vocals.

But when it comes time to sing another part of the harmony…

…it helps to keep all the “old” harmony takes in one ear and the “new” harmony takes in the other ear. And lower the volume of the combined earlier harmony part.

This technique, like Trick #1, allows you to clearly hear your pitch because of the reverb. But because you have a previous good vocal tickling one of your ears, it’s easy to be in sync with your previous performance, and it’s easy to tell when your pitch is “on” and blending well.

Some days, everything in the studio happens efforlessly. You’re in a state of flow. You’re in the zone.

On other days, nothing goes right. That’s when it’s time to pull out Trick #3. This trick almost never yields results as good as a day spent in the zone, but it’s better than accomplishing nothing at all.

Use Trick #1 and “Frankenstein” a lead vocal part together. When I say “Frankenstein,” I mean do whatever you have to do to get an acceptable performance: sing one line at a time, punch in individual words, or use — gasp!!! — Autotune.

Once you have your technically OK but uninspired lead vocal part, use Trick #2 and pretend like you’re recording a background vocal part. That is, let the Frankenstein part be your security blanket. Follow along with your breaths and syllables. Once you do Trick #2 enough, nailing a lead vocal like this is easy — it’s like being on autopilot.

When you finally have an acceptable performance, delete the Frankenstein track.

I did a vocal session recently where I used Trick #1 and Trick #2 extensively. I also did some variations to these tricks, like recording a keyboard part to help learn a harmony line. Here is a mp3 illustrating Trick #1 and Trick #2 in action.

This mp3 contains the following excerpts from the session, recorded directly from my headphone feed:

1. The first vocal overdub of the first harmony part, using Trick #1. Note the large amount of reverb to help discern pitch information.

2. The second vocal overdub of the first harmony part, Trick #2. Note how the first take is panned far left and is completely dry.

3. Three vocal overdubs for the first harmony part panned far left and completely dry. A Casio keyboard guide part is being recorded through the vocal mic to help learn the second vocal overdub part.

4. The first vocal overdub of the second harmony part. All three overdubs of the first harmony part are panned far left, are completely dry, and are lowered in volume. (Actually, they’re a little too low — I was recording quickly and didn’t feel like correcting it.) The Casio keyboard guide part is panned far right.

5. The second vocal overdub of the second harmony part. The first harmony overdubs are still panned far left. The first overdub of the second harmony part is panned far right, along with the Casio keyboard guide part.

6. Three harmony parts with three vocal overdubs each panned dead center with no effects, no EQ, and no compression.

And what is the “preamp test for Songs and Sonics?” Stay tuned…!

Four months ago, I announced a remix contest for my song Really Really Weird. The original multitracks were put online, and I encouraged remixers to mangle things to their heart’s content.

The entry ranked the highest by a panel of seven judges and myself would win the grand prize — one of the keyboards I used on the song, a circuit-bent Casio SK-1. Three runner-ups would win a Girls Aliens Food CD and a Girls Aliens Food t-shirt.

Well, something kind of funny happened along the way. Y’see, there were a lot of interesting remixes, and there was at least one thing I liked about each remix. So everybody who submitted a remix — but who didn’t end up in the top four slots — was given a free digital download of the Girls Aliens Food album.

Thanks to the judges (Mathieu Bournazel, Rob Gibson, Pea Hix, Missy Kulik & Raoul De La Cruz, Mike Langlie, and Matt Williams) and to everyone who sent in their mixes. It was fun to hear new interpretations of a song that I was very familiar with; hope everyone else had fun too.

So without further ado — the remixes!

And everyone else… (Don’t let the fact they don’t have a contest ranking next to their name throw you — there’s some good stuff in here!)

Note to podcasters: All of these tracks are released under the Creative Commons Attribution-Noncommercial 3.0 United States License, which means you’re free to use them in your podcast as long as you give proper credit.

Just wanted to let you know that there will be fewer updates on Songs and Sonics for the next month or so.

No, I haven’t lost interest in blogging — in fact, I’m working on some extremely cool things that I can’t wait to tell you about.

What’s going on is, well, I’ve got this little debut album called Girls Aliens Food which will be released on October 1st. My internet time is being (happily) sucked up by getting things in place for the release and interacting with the radio/podcast/press people who will hopefully spread the word about the album.

I’ll be updating The Simple Carnival’s MySpace page, MySpace blog, and web site often, since that’s where my attention will be. Also, the Really Really Weird remix contest is in full swing (you can check out the latest entries here).

But I’m not going to let Songs and Sonics rot. It’s just going to take a short nap. It’ll be back — with a vengeance — when things are a little less busy around the Boller household.

If you want to be updated when something new is posted here, please sign up for the mailing list (look in the right hand column — you might need to scroll down to see it).

In the meantime, thank you for your patience… And don’t forget to order your very own personally autographed copy of Girls Aliens Food! (Sorry — I’m permanently stuck in marketing mode this month…)

Well, it’s taken multiple crazy web server issues and wild goose chases to locate old backup DVDs, but it’s finally here — the Really Really Weird Remix Contest!

Almost everything you need to know about the contest can be found on this page:

Note that there are some new tidbits on that page, like what criteria will be used for judging your remix. And here’s where you can download the raw multitracks.

However… once you download the tracks, there’s some additional information you ought to know about the tracks themselves.

The files that are part of the second mix have the prefix “midsection_”. It’s recommended that, at least until you’re familiar with the tracks, you make two different mix sessions and work with each half separately.

In the Girls Aliens Food version of the song, the main song mix is interrupted at around the 2:57 mark with the midsection mix. When the midsection mix ends, the song cuts back into the main song mix, at the main song mix’s 3:17 mark.

The “door slam” file is tacked on to the ending of the album mix. The door slam is not meant to go at the beginning of the mix, as it would be if you imported all of these tracks into a DAW; it’s meant to be inserted wherever the song ends.

Naturally, you are not expected to stay true to the original mix (in fact, it is encouraged that you don’t) — but this information is meant to help make sense of how the tracks were constructed, should you want to use the original mix structure as a starting point.

Most of the tracks are mono, though a few are in stereo. The tracks in stereo are marked with the word “(stereo)” at the end of the name.

What is the BPM for this song? Good question. The Me and My Arrow EP and Girls Aliens Food album are the first Simple Carnival releases where I used a click track. However, I didn’t want to make things sound too slick — I wanted to keep a connection with the lo-fi sound that was established with the previous EPs. So I varied the tempo of the click track very, very slightly, speeding up or slowing down things in the way a conductor might lead an orchestra.

So what does this mean?

It means your job as a remixer just got a lot trickier. Although these tracks line up with one another in a DAW, they don’t collectively line up with a steady tempo grid. And I have been unable to locate the original Pro Tools session that I used for recording the song, so I can’t tell you the initial BPM. By the time I imported the tracks into Reaper for mixing, I paid no attention to the BPM or grid anyway.

If you have any questions about the files, please post them as a comments on the Really Really Weird Remix Contest page or as a comment at the bottom of this post.

Good luck with your remix!

There’s no interview this week on Songs and Sonics. Instead, I’d like to take this opportunity to announce a Simple Carnival remix contest.

What? A remix contest? Are you serious?

Yep. Here’s the deal. In the very near future, I will be making available the raw multitracks for the Simple Carnival song Really Really Weird:

You can remix, mangle, and twist these tracks to your heart’s content. (They will be released under a Creative Commons Attribution-Noncommercial 3.0 United States License.) Submit your final 192 kbs mp3 file, and whoever makes the mix judged the “best” (whatever that means) by a panel of judges will win the grand prize…

The grand prize consists of a Simple Carnival t-shirt, a Girls Aliens Food CD, and one of the actual instruments I used on Really Really Weird: a circuit-bent Casio SK-1 sampler.

For those who are unfamiliar with the Casio SK-1, it is widely considered to be the holy grail of circuit bent electronic instruments. (And if you don’t know what circuit bending is, here’s an introduction.)

When you flip the switches and buttons that I’ve installed, the SK-1 can mangle and manipulate audio like nobody’s business. Similar modified SK-1s go for upwards of $250 on eBay.

There will be three runner-ups. Each runner-up prize consists of a Simple Carnival t-shirt and a Girls Aliens Food CD.

I’ve recorded nearly fifty songs as The Simple Carnival. And I’ve realized that I’ve only used my SK-1 for about four seconds on one song (Really Really Weird). So I want to donate the SK-1 to someone who will put it to better use than myself.

Also, have I mentioned that I have an album called Girls Aliens Food coming out on October 1st, 2008? This contest is admittedly one way to get the word out. I mean, why not?

A panel of musical judges and I will rank the submissions. My vote counts just as much as the other judges, though I’ll be the tiebreaker if necessary. Our votes will be kept confidential. Don’t send us angry e-mails if you don’t win anything. I would love to give SK-1s to everyone if I could — in fact, that would make a great platform for a presidential candidate. Unfortunately, I have only one instrument to give away, so I ask that everyone who doesn’t win something to please be a good sport. We still love you.

All submissions must be in by midnight EST on November 30, 2008. The winners should be announced (hopefully) a week or two after that. If we end up getting two billion submissions, it might take more time to listen to everything.

They are (in alphabetical order):

Mathieu Bournazel from Purple Submarine Orchestra	Rob Gibson from Mister Fusty	Pea Hix from Optiganally Yours	Mike Langlie from Twink	Matt Williams from The Brigadier

I’m thrilled to have these terrific musicians/songwriters/producers participating, as I listen to their music all the time (and you should too!). Because these folks have generously volunteered their time, it’s always possible that the demands of real life might interfere and take precedence over this contest. Judges may come and go as the contest progresses, but I really hope these guys stay.

Yes, but anyone who was interviewed here at my blog, Songs and Sonics, is not eligible to win. The judges are also not eligible to win. Also, only one mix can be submitted per person. Sorry.

The remixes will be regularly rotated on this special MySpace page:

Bookmark that page, send a friend request, keep checking on it. Subscribe to that page’s blog. Read the comments section. That’s where most of the action should occur. Please keep the comments and the competition friendly; I don’t want to turn this into a backstabbing Survivor-like excursion. It’s only a toy keyboard, folks.

There is also the remote possibility that I’ll release a CD-R/album download with some of the remixes on my Sundrift Records label; by submitting a remix, you will be agreeing to the possible inclusion of your mix on such a release. (Naturally, you will be credited as the remix engineer for your track if that happens.)

Only if you have the legal permission to do so. If you’re not sure, the rule of thumb is, commercial sample libraries: yes. James Brown records: no. Don’t let the judges knock you out of the running because we identified some commercial CD snippet that you shouldn’t be using.

But don’t let that rule inhibit you from going wild with your remix. Record new parts if you want. Put the song in a different key. Change the tempo. Autotune the heck out of my voice, or remove my voice altogether. There is nothing to be gained by staying true to the original mix. We want to reward creativity. Just keep it clean and don’t do anything that the copyright police would disapprove of.

Not at all. There are scratches and sticker residue on this instrument. I bought it secondhand and never meant to give it away or make it look presentable. But it’s a useful/functioning piece of gear, it makes terrifically weird noises, and it was used in a Simple Carnival song. What more could you want?

They’re not online as of this writing. I still have more work to do to get them together. They should be posted next week.

See that little sidebar thing on the right, near the top of the page that says “Mailing List”? Put your e-mail address in there and click the “Send” button. After responding to the confirmation e-mail, you’ll receive a message the moment a new article is posted on Songs and Sonics. I post only about one article a week, and I don’t share your address with anyone. The next article should contain information on where to download the tracks.

So tell your friends, post about it online, let everyone know: Songs and Sonics wants to hear some really really creative stuff!

There are few greater gifts to the creative noise enthusiast than Dollar Tree.

No, I’m not being paid to say that. It’s just something I’ve grown to realize over time.

Today’s project largely consists of parts found at — you guessed it — your local Dollar Tree. I don’t know if these exact parts are available at every Dollar Tree, or even if they’re available at my Dollar Tree. Their inventory seems to change as frequently as I change shirts. So you might need to find some substitute parts. If that’s the case, don’t worry — this isn’t rocket science. It’s just another fun way to play with sound.

First, allow me to introduce this garden hose-looking toy called a whirlie (blatant CD product placement to demonstrate relative whirlie size):

When you hold the end of the whirlie with the wide mouth and spin it fast enough, this unassuming plastic hose generates some otherworldly sustained tones. I used multiple, overdubbed whirlies in this way at the end of “Freetime” (from The Simple Carnival’s Menlo Park EP).

If you spin the whirlie faster, the pitch jumps up a fifth. Spin it even faster, and the pitch will be a fourth above that, making the pitch an octave above where you originally started. Spin it even faster, and the pitch jumps yet again. You can do this until your arm falls off or you go into orbit, whichever happens first.

Whirlies are even capable of playing melodies. If you cut the tubes to different lengths, you’ll get a different set of pitches to play. Of course, this melody-playing technique works best if you have more than two arms. Australian composer Sarah Hopkins does just this, as she writes serious music pieces for… whirlies. You read that right. Check out her web site if you don’t believe me.

But this project isn’t about using whirlies for their naturally pleasing tones. No, this is all about a homemade contraption I made called the Sound Spinner:

The Sound Spinner is a whirlie with a small speaker in the wide mouth end. The speaker is pointed toward the long end of the tube. So when you spin the whirlie, it throws the sound around the room in every direction, creating what’s called a Doppler effect. Have you ever noticed how an ambulance siren’s pitch will change as the vehicle fades into the distance? That’s the Doppler effect.

Pretty much. However, I have a real Leslie speaker, and they sound significantly different, even though they operate on the same principle.

The Sound Spinner costs only a few bucks to build, sounds worse than a real Leslie, and a human can actually “perform” with it. Do not discount the Sound Spinner’s ability to do a synthesizer filtering effect (by covering the open end of the whirlie with your hand) or a Peter Frampton talkback effect (by putting the open end of the whirlie in your mouth and “speaking”). And the Sound Spinner doubles as a variable speed autopanning device, if you spin it in front of a stereo microphone.

And don’t forget the visual impact this thing has! In the right musical situation, playing a Sound Spinner in a live show could leave a lasting impression on the audience… and possibly cause unparalleled fits of laughter.

Still, don’t let the potential embarrassment of using this gadget scare you from building one. It’s perfectly usable in the studio, where nobody can see how you got that weird sound.

What makes the Sound Spinner interesting is what it does to the source material you send through it. I used the Sound Spinner at the end of Really Really Weird (from the Me and My Arrow EP as well as Girls Aliens Food album), and the source material was a bit unusual.

What is this device? The model name has long escaped me, but it’s an environmental noisemaker I found at a thrift store and then circuit bended. It plays annoyingly grating 10-second loops of cricket, bird, and ocean noises, meant to “relax” the unlucky sap who received this as a gift. I installed some glitch-inducing switches and a big knob which controls the pitch/speed (sample rate) of the noise.

Since the end section of Really Really Weird is basically the sound of my studio going ballistic, I recorded some glitchy bird noises out of this device that gradually rise in pitch and speed. I then took that performance and fed it into the Sound Spinner, spinning it in front of a stereo microphone, for an extreme as possible stereo effect. The end result is buried in the mix, so much that it just serves to “tickle the ears” a bit when wearing headphones. (The effect is even more hidden than I originally intended in the final mix, because, as soon as the Sound Spiner becomes audible, the mix is swirling around and well on the way of marching toward your right ear.)

So here’s some mp3s to show what the Sound Spinner does and where it ended up in this particular song. These clips are best listened to with headphones.

The Sound Spinner consists of the following parts (the first five can be found at Dollar Tree):

A whirlie. ($1.00)

An external iPod speaker. ($1.00)

Foam to keep the speaker in place. ($1.00)

Electrical tape. ($1.00)

A piece of cardboard to close off the foam and speaker on the wide mouth end.

A powered computer speaker.

A long, 2-conductor audio cable to go from the computer speaker’s amplifier to the iPod speaker in the whirlie.

The external iPod speaker is one of those Dollar Tree finds; not only do you get two cheap stereo speakers for a buck, but included for no extra cost is a pushbutton switch and a decent 1/8″ stereo molded jack with cable. All of these parts can be utilized in other electronic projects.

What makes this particular model desireable is how the internal speaker fits perfectly within the whirlie’s wide mouth end.

The iPod speaker does not require external power, and it’s not loud enough to sufficiently force meaningful sound waves down the whirlie tube. I ended up using an AC-powered computer speaker as the “amplifier.”

So instead of the powered computer speaker sending a signal to one of its speakers, one speaker is eliminated entirely, and the other speaker’s signal is re-routed to the external iPod speaker.

Note that I don’t recommend working with AC power if you don’t know what you are doing with electronics. Heck, I don’t know even what I’m doing with electronics, and here I am writing an article about it. In my defense, I was scrounging around the house for a cheap amp to sacrifice, and this is what I found. So re-read that disclaimer a few paragraphs back and make sure you know what you’re getting into if you build one of these things.

You’ll want to wedge the iPod speaker in the one end of the whirlie (don’t glue or tape it in) then pack in some foam to keep the speaker in place.

Finally, you’ll need to cut a piece of cardboard and seal the end of the whirlie with electrical tape. The goal is to keep all of the sound from escaping the end that you hold.

And that’s it. Happy noisemaking!