Five Sounds With Joe Chiccarelli
Joe Chiccarelli is a GRAMMY-winning producer and engineer with an impressive list of credits ranging from indie rock icons like the White Stripes to pop legends like Elton John. Recently, Joe teamed up with the software experts at IK Multimedia to create a custom Vocal Strip plug-in ($159) that combines all of his favorite mixing tools into one versatile, easy-to-use interface.
We sat down with Joe for a special Five Sounds With interview, where we talk about five records from his career that utilize the same techniques and tones that inspired the new Vocal Strip plug-in. Read on to learn more about his work with the Shins, Alanis Morissette, Café Tacvba and more.
In 2008, I produced an album for the Shins on an API console. I've found that often, if I use a compressor on the insert of a vocal, I have to be a little delicate with it. Sometimes I like to use a lot of compression, but it can take the life out of things if there’s no way to blend between the dry and wet signals on a channel insert. In those cases, I would put the compressor in parallel.
So when I was mixing the Shins, I had a Teletronix LA-2A on the insert of the vocal with just one or two dB of compression. Then I bussed the vocal out to an 1176, where I compressed it heavily—probably 7 to 10 dB of compression with a slow-to-mid attack and a fast-to-mid release. Then I would blend the compressed track in with James Mercer's lead vocal to give it a little punch and help it jump through the track.
The ratio of the channel vocal to the parallel might only be 30% of the parallel—it was just to give the vocal some thickness, power and punch. When you hear the mix, you hear the natural vocal and its dynamics, but behind it is this sort of fattened, tough sound that gives the vocal a lot of weight and importance in the track.
The White Stripes
I also used the same parallel compression technique when working on the White Stripes with Jack White. Jack’s got a really strong, powerful voice and the tunes on the White Stripes mix we did were quite varied. Some were very simple and folky and others were really big, powerful rock tracks, like the title track.
On pretty much all of the tracks, there was a combination of a few different compressors. Normally, I like to use a blue stripe 1176 on Jack’s vocal, but it would have been bussed out to a Fairchild 670 with a very slow release, which kept the vocal present in the mix all the time.
But on many of the songs, the way I use multiple compressors is to help change the tone of the vocal from section to section of the song. In other words, you might want something nice and warm for the verse, but in the chorus, you might need to get above a lot of rock guitars. So you might want to use a compressor that's got more growl for the bridge, which may be heavily produced. You may need something that just gives it a strong forward presence to get above all the production elements.
I would have the vocal sent out to two or three different compressors and just turn them on and off from section to section. So it really gave the vocal a different personality from each section of the song. Choruses would have a little bit more bite to them and the bridge might have a little bit more growl.
I used the same parallel compression technique on the Alanis Morissette records that I did. She's got a beautiful voice, which I recorded with a C12 microphone so it had a lot of nice, open top end to it. There wasn’t much compression that went on while recording because she's very quick in the studio—it's only a couple of takes. To avoid overcompressing while recording, a lot of those cuts were enhanced with parallel compression.
I used an old-school 1176 that was tucked in behind the lead vocal, which probably didn’t have any compression—maybe a Summit TLA-100 or a Retro Instruments 176, just barely moving. The 1176 really helps give the vocal power. You have this really pure, natural vocal that makes up 80% of the tone that you hear in the record and then just tucked behind it is that fattened parallel compressed vocal.
Café Tacvba have made dozens of albums and I’ve been really fortunate to be a part of four or five of them. Ruben, the lead singer, has a great voice where he can be really warm and intimate, but then also he can be really biting and mid-rangey.
I used a lot of tricks to get his vocal sounding right because the band doesn't do a lot of background vocals or doubling of vocals. Getting his vocal over a wall of synths and guitars can be a little tricky. So I would always be using several compressors: usually a Fairchild, a Retro 176 sometimes, or even a Chandler Germanium.
His voice is a bit more mid-rangey and tight when he gets into that upper power range, so I would always be looking for something that would fill his voice out and it would always be things like a Pultec and Fairchild because those to me always had a lot of weight and chest in the bottom. So I'd always be looking to round things out and make the voice as big as possible in the track.
The one thing that I discovered is that when you're working with devices in parallel where they're not featured in the mix that much—in other words, they’re only a part of the sound—you can almost get away with more processing and it won't sound processed.
I realized that I don't like to push a lot of mid-range on the voice—the 3-5 kHz range—because I think it gets harsh. Especially when you're fighting with a lot of guitars that are very mid-rangey. But if I do the EQing and de-essing on the parallel side of it, not on the vocal channel, I can actually get away with more of it and you won't hear it. So I was able to de-ess the parallel compressor or de-ess the sends to the reverb so you don't get those bloody “s’s” in the reverb.
I would be also able to EQ in that 3-5 kHz range that you want to push in the vocal to get the presence. Unfortunately, when you did that to the uncompressed sound, it made it sound harsh. But if I did the EQ after the parallel compressor, I could actually get away with pushing a little bit more of those upper mid-range frequencies into the record that I wanted to get, but couldn't do it on the uncompressed signal.
While in the midst of developing the Vocal Strip plug-in, I was working with this great indie rock band out of Philly called The Districts. Some of the tracks on that record had doubles, triples and lots of harmonies. So, I used a test version of the Vocal Strip plug-in to get the sound I was used to hearing. Normally, I would probably use an 1176, a Fairchild 660 or an old UA 175B—all of those compressors are available within the Vocal Strip plug-in.