A single vocal is intimate. A doubled vocal is bigger, thicker and more confident. A full stack - multiple doubles and harmonies layered together - is the sound of modern pop choruses. Knowing how to record and blend them is a core vocal production skill.
Double-tracking
A double is the singer recording the same part again, as closely as possible. Played together, the two takes are never identical - tiny differences in pitch and timing create a natural thickening that no plug-in fully reproduces.
- Tight doubles - matched closely in timing, pitch and phrasing. They thicken the lead without being heard as two voices. Common in pop and rock.
- Loose doubles - more variation, so the listener hears two voices. A warmer, more relaxed sound.
Double-tracking became a studio staple in the 1960s - so popular that engineers invented artificial double tracking (ADT) to save singers the work of recording the part twice.
Recording doubles
- Record doubles after the lead is comped, so the singer is matching a final performance.
- Match the phrasing exactly - same breaths, same word lengths, same endings.
- Watch the consonants. Hard consonants and especially "s" sounds that do not line up are obvious. Many producers ask the singer to soften or drop consonants on doubles, letting the lead carry them.
- Align after recording with your DAW's flex tools or a dedicated alignment plug-in, but leave a little natural variation.
Artificial doubling
When a real double is not possible:
- Duplicate and delay the vocal by 15-30 ms, perhaps with a few cents of pitch shift.
- Micro-pitch shifting - two copies, one a few cents sharp and one a few cents flat, panned left and right, for width.
- Doubling plug-ins that simulate the variations of a real double.
They work, but a real double almost always sounds better.
Building a stack
A vocal stack layers several parts:
- Lead vocal - centre, loudest, carrying the words.
- Lead double(s) - tight, quieter, centre or slightly panned.
- Wide doubles - two more takes panned left and right.
- Octave doubles - an octave above or below, very low in the mix, for power.
- Harmonies - thirds, fifths and other chord tones, each ideally doubled and panned in pairs.
Not every song needs all of it. Verses are often single or tightly doubled; choruses get the full stack. The contrast is part of what makes the chorus feel bigger.
Panning and blending
- Keep the lead in the centre.
- Pan pairs symmetrically - harmony 1 left, its double right, for a balanced image.
- Process the stack together. Route all backing vocals to a bus and compress, EQ and reverb them as one group. This glues them into a single texture.
- Sit the stack behind the lead. The listener should hear the lead words clearly, with the stack as a halo around them.
- High-pass the stack more aggressively than the lead, and consider cutting some upper mids, so it does not compete for presence.
Avoiding mush
Big stacks go wrong in predictable ways:
- Misaligned consonants - fix timing, or remove consonants from the stack.
- Too many "s" sounds - de-ess the stack bus heavily, or edit sibilants out of every part but the lead.
- Too much low-mid - every voice adds energy around 200-500 Hz. Cut it on the bus.
- Too loud - pull the stack down until you just stop noticing it, then bring it up slightly.
Do this now
- Record a lead vocal for a chorus and a tight double. Align the double and listen to the difference.
- Add two wide doubles panned hard left and right.
- Record a harmony a third above and double it; pan the pair symmetrically.
- Route all backing parts to one bus, compress and de-ess it, and balance it behind the lead.