Learn Music and Audio Production | iZotope Tips and Tutorials
8 Essential Tips for Editing Vocals
We often talk about editing vocals, but we haven’t written a step-by-step guide for how to edit vocals in a musical context. So, that’s exactly what we’re going to do today: we’re giving you eight simple yet effective tips for vocal editing.
1. Pick the best comp
Comping might fall into your editing purview, though mostly it falls to the producer. However, a lot of times you are the producer. So, it behooves us to ask, how do we go about comping?
The easiest way, I find, is to identify the best overall take and fill in the lesser bits with other takes. This means listening to every take in full before proceeding, separating the good regions from the bad. When doing this, I like to color-code the results—red regions for the best, green for usable, orange for passable, and purple for all out horrible. The take with the most red? That’s your master track. Fill in the green and orange bits as need be.
You could also listen to each take with a pen and paper before hand, making a graph for verses and choruses and checking off the best ones as you go. This may work better for you, if you need to step away from the computer to make everything work.
2. Learn where to make the cut
Sometimes you’ll have to cut between two takes in the middle of a phrase. When you do this, look for a hard consonant—a plosive (P) or dental consonant (T, D) are both good candidates. Fricatives (F or TH) are slightly less secure, but you can find a good place therein.
Instances of sibilance sometimes don’t match in intensity between takes, so you have to be careful when selecting those. Do not cut on a vowel if you can help it—it’s hard to get a good result.
Sometimes, over a sustained vocal, you can slowly crossfade from one take to another and fool the listener, but it’s not wise to seek these instances out. Wherever you cut, do make sure that the intention and tonality of the following region matches its antecedent.
As you’ll have to make cuts between takes and regions sometimes, it’s best to keep this guide in mind for the inevitable comping session.
3. Position phrases for the best timing (if need be)
This is where the utility of editing meets the creativity of music. Moving a word in a phrase forward (i.e., later in time) can create a more relaxed atmosphere, a kind of “leaning back” effect that’s often called “behind the beat.” Placing words earlier in time could create a feeling of rushing, which may be in order.
Develop an innate musical feel for what works by listening to great songs and identifying what the vocalist is doing. For instance, if you need a reference on what vocals sound like when they’re very relaxed, check out D’Angelo’s Voodoo or Black Messiah; no one slinks behind the beat (which ironically means later in time) like D’Angelo. For the other side—for anticipatory phrasing—many famous tunes by Eminem fit that bill. He’ll often anticipate a beat to create urgency.
When you tend to your vocals, keep the song in mind, and you’ll know if you can help things along by positioning phrases or words. Does the tune feel more like a D’Angelo song—and is the singer rushing? Well, now you know what to do.
4. Use fades judiciously
A.B.F.—Always Be Fading.
Seriously, you want to make sure that all your vocal regions have fades at their onset, fades at their endings, and crossfades stitching them in between. Why? Because not doing so sometimes causes audio hiccups—clicks and pops you don’t want in the session.
Depending on how CPU-heavy the session is, and which software you’re using, you could be especially susceptible to audio problems when you’re applying heaps of processing without fades.
5. Take care of forensics
Forensics, here, means any bit of repair work necessary to clean up the track. It could include de-noising, de-reverberating, de-breathing, de-clicking, removing mouth noises, removing clipping distortion, EQ-matching out-of-whack phrases, and even ambiance-matching, should something sound drastically out of place.
This article references a previous version of RX. Learn about RX 10 and its powerful new features like Adaptive Dynamic Mode in RX De-hum, improved Spectral Recovery, the new Repair Assistant, and more.
For these purposes, no better tool comes to mind than RX, and many tutorials exist both on iZotope for implementing these corrective processes.
6. Take care of pitchiness
Even with the best comping—even if the style of music is raw, even if the singer is exemplary—you may run across a bum note. The good news is that by following the steps before this one, you’ll minimize your need for pitch-correcting the singer. You’ll only have to grab and affect one or two phrases, or, conversely, you’ll be using pitch-correction as a noticeable effect because of the song’s genre.
Either way now’s the time to implement the process, using something like Melodyne 5 essential, which is included with Nectar 3, or other pitch-correction software you’re already using, such as RX’s various algorithms, your DAW’s innate correction (Logic has passable pitch-correction), and others.
7. Clip-gain for evenness or for intentionality
Before hitting a compressor, before hitting a leveler, before hitting an EQ or any other mixing process, try this time-consuming trick to save you trouble going down the line: Clip-gain the phrases, either for congruity or for intentionality. If someone is very quiet on a line that needs to be quiet, preserve it. But if the line isn’t supposed to be as quiet as it is, you’ll recognize this fact, so go ahead and juice its region by the appropriate number of dB.
You can, if you like, use the automatic clip-gaining in Nectar 3, Auto Level Mode, which is quite intelligent. However, I’ll always advocate for fine-tuning the level of clip gaining for consistency by ear, as knowing how to do so benefits you on unfamiliar DAWs, in unfamiliar studios, and more. Also, if you’re clip-gaining for intentionality, an automatic real-time leveler might not help you there.
The takeaway: even if you rely on excellent technology like Nectar 3 for automatic leveling, learn how to manually clip-gain, too.
When clip-gaining, use metrical tools to help you get in the ballpark. A LUFS meter giving you momentary readings can be useful here, as can Insight 2's Intelligibility Meter, provided you’ve got the Relay module on the vocals and the rest of the song feeding Insight so that the meter can make the necessary calculations for your reading.
8. Work dry, without effects
In general, you shouldn’t have effects like delay or echo on while you’re editing vocals, as they can obfuscate the imperfections or timing issues you need to fix. If the muse is urging you to work on the mix before the editing, hey, that happens sometimes. Just be sure that when you do come back to edit a vocal, you do so with the dry, unprocessed vocal.
Editing vocals does more than clean up the proceedings. As you’ve seen from tips like timing and pitch correction—not to mention comping—editing vocals can take a track from a lifeless demo to a full, lively mix. So follow these tips if you want a lively vocal track. Indeed, follow me if you want (your vocal track) to live!