Role in the mix
The lead vocal is the center of the mix: everything else is built to accompany it. In pop it must be intelligible, consistent, close and big all at once: every word can be understood at any volume, the dynamics are controlled, and the space (reverb and delay) makes it big without pushing it away. Spectrum: fundamentals ~80–350 Hz (male) and ~160–700 Hz (female), body 150–400 Hz, intelligibility and presence 1–5 kHz, sibilance 5–10 kHz, air 10–16 kHz. iZotope describes the "professional" vocal sound as a sequence of steps: editing, pitch, EQ, compression, de-essing, saturation, space, automation.
Recording and preparation
- Mic: large-diaphragm condenser (U87, C414, TLM 103, SM7B dynamic for aggressive vocals or bad rooms), pop filter, 15–25 cm. Dry room.
- Comping of the best takes; editing of breaths (lower them by 6–12 dB, don't remove them all), clicks, plosives (HP or a fade on the "p"), noises.
- Clip gain: level the phrases before compression (in FL: audio clip volume or automation) so the compressor works evenly.
- Pitch: Melodyne, Auto-Tune, in FL Pitcher (real-time correction, also as a harmonizer) and NewTone (note-by-note editing). In modern pop the tuning is precise but natural (retune 20–50 ms, humanize).
Gain staging
Peaks −10/−6 dBFS after clip gain; input to analog-modeled compressors around −18 dBFS RMS.
Corrective EQ
- HP 70–100 Hz (male) and 100–120 Hz (female), 12–18 dB/oct.
- Mud/proximity 150–300 Hz: −2/−4 dB, Q 1–1.5.
- Boxiness/nasality 400–800 Hz and ~1 kHz: −1/−3 dB.
- Harshness 2–5 kHz: dynamic EQ or a resonance suppressor (Soothe2) on resonances that appear only on the loud parts.
- Plosives: dynamic HP or editing.
Creative EQ
- Body +1/+2 dB at 150–250 Hz (thin voices).
- Presence +1/+3 dB at 2–4 kHz (after compression).
- Air shelf +2/+5 dB at 10–12 kHz (Pultec, Maag EQ4 Air, soft Pro-Q shelf), after the de-esser.
Compression
- Serial (the standard): a FET (1176/CLA-76: ratio 4:1, medium-fast attack, fast release, 3–6 dB) for the peaks + an opto (LA-2A/CLA-2A: 2–4 dB) for musical leveling. Chris Lord-Alge uses a 1176 at 4:1 with a fast release on the vocal and then rides the faders.
- VCA (SSL, dbx 160) for more precise, modern control.
- Parallel: send to a very aggressive compressor (FET all-buttons, 10–15 dB), blended at 10–20% for density.
- A light final limiter for residual peaks.
- In FL: Fruity Limiter (COMP) and Fruity Compressor; alternatives Pro-C 2 (Vocal), CLA-76, CLA-2A, UAD 1176/LA-2A, Distressor.
De-esser
Band 5–9 kHz (male 4–7 kHz, female 6–9 kHz), 3–6 dB on sibilants; better two light de-essers (before and after compression) than one heavy one. In FL: Maximus with the high band compressing, or Fruity Limiter on a Frequency Splitter; alternatives FabFilter Pro-DS, Waves DeEsser, Sibilance.
Transient shaper
Rare. Useful on unclear consonants in rap and rhythmic pop.
Saturation
- Light tape/tube (Distructor, Saturn 2, Decapitator, UAD/Waves J37) for density and presence without EQ boosts.
- Distorted parallel for aggressive vocals (pop-rock choruses).
- Light exciter (Fruity Stereo Enhancer is not an exciter: use Saturn 2 or Ozone Exciter) for air.
Delay (the main space in pop)
- Slap 80–140 ms, feedback 0–10%, filtered (HP 300 Hz, LP 4 kHz): closeness and thickness (CLA uses a tape slap).
- 1/4 or 1/8 tempo-synced, feedback 15–30%, return filtered and ducked by the vocal itself (3–8 dB): tail in the gaps.
- Throw: high delay automated on the last word of the phrase (CLA: automated "long delay spills").
- Ping-pong dotted 1/8 in choruses for width.
- In FL: Fruity Delay 3 (with Fruity Limiter in sidechain for ducking); alternatives EchoBoy, H-Delay, Timeless 3, Valhalla Delay.
Reverb
- Plate 1.2–2.2 s, pre-delay 20–80 ms (the word stays up front), HP 200–300 Hz, LP 8–10 kHz, possibly a de-esser on the return.
- Long hall only in ballads and epic choruses, at a low level.
- Short room for cohesion.
- Reverb ducked by the vocal in modern dry productions.
Modulation
- Micro-pitch / doubler (MicroShift, Waves Doubler, Pitcher with detune ±5–10 cents and delay 10–20 ms) to widen choruses.
- Chorus and flanger as effects on specific parts.
Stereo and panning
Lead vocal in the center, mono, with width provided by stereo delay/reverb and doubles (see backing vocals). Channel correlation +1.
Sidechain
- The vocal is the mix's most important source of sidechain: dynamic EQ 1–5 kHz on guitars, synths, pads and piano (Trackspacer, Pro-Q, Neutron unmask).
- Ducking of delay and reverb returns.
Automation (essential)
- Vocal ride: volume automation word by word or phrase by phrase (Waves Vocal Rider as a base, then by hand). CLA: it is easier to ride the fader than to EQ.
- Delay throw sends, bigger reverb in choruses.
- +0.5/+1 dB in choruses; presence EQ slightly different between intimate verse and chorus.
Differences by genre
| Genre | Approach |
|---|---|
| Contemporary pop | highly controlled, 12 kHz air, ducked delays, plate |
| Ballad | more dynamic, hall, softer compression |
| Dance-pop / EDM | heavily compressed, pumping returns, wide choruses |
| K-pop | bright vocal, heavily edited, doubles and stacks |
| Indie / alt-pop | more saturation, effects, sometimes further back |
| Singer-songwriter / acoustic | natural, room, little compression |
Common mistakes
- Compression without clip gain: the compressor pumps on the loud phrases.
- Air at 12 kHz before the de-esser: explosive sibilants.
- Reverb without pre-delay and without HP: distant, smeared vocal.
- Vocal too loud or too low: check on multiple systems and at low volume.
- Removing all the breaths: unnatural vocal.
Starting settings
| Parameter | Value | Why |
|---|---|---|
| Clip gain | Phrases within ±3 dB before compression | Even compression. |
| Channel peak | −10/−6 dBFS | Headroom and correct level for plugins. |
| High-pass | 70–120 Hz | Noise and proximity. |
| FET | 4:1, medium-fast attack, fast release, 3–6 dB | Peaks and character. |
| Opto | 2–4 dB | Musical leveling. |
| Parallel | 10–15 dB GR, mix 10–20% | Density. |
| De-esser | 5–9 kHz, 3–6 dB (also in two stages) | Controlled sibilants. |
| Slap | 80–140 ms, filtered | Closeness and thickness. |
| Tempo-synced delay | 1/4 or 1/8, feedback 15–30%, ducking 3–8 dB | Tail in the gaps. |
| Plate | 1.2–2.2 s, pre-delay 20–80 ms, HP 200–300 Hz | Vocal up front and big. |
| Pitch correction | Retune 20–50 ms, humanize | Natural precision. |
| Vocal ride | ±1–4 dB per word/phrase | Constant intelligibility. |
| Air | Shelf +2/+5 dB at 10–12 kHz | Modern brightness. |
Frequencies to watch
| Range | What to do |
|---|---|
| below 70–120 Hz | High-pass (male 70–100 Hz, female 100–120 Hz), 12–18 dB/oct. |
| 150–300 Hz | Proximity and mud: −2/−4 dB; body +1/+2 dB only on thin voices. |
| 400–800 Hz | Boxiness and nasality: −1/−3 dB. |
| ~1 kHz | 'Telephone' nasality: small cut if needed. |
| 2–4 kHz | Presence and intelligibility: +1/+3 dB after compression. |
| 2–5 kHz | Harshness: dynamic EQ or Soothe2 on loud-passage resonances. |
| 5–9 kHz | Sibilance: de-esser 3–6 dB (male 4–7 kHz, female 6–9 kHz). |
| 10–16 kHz | Air: shelf +2/+5 dB after the de-esser. |
Example chain
| Step | Tool | Settings | Why |
|---|---|---|---|
| Editing and clip gain | Playlist / Edison | Breaths −6/−12 dB, plosives, leveled phrases | Clean foundation. |
| Pitch | Pitcher / NewTone / Melodyne / Auto-Tune | Retune 20–50 ms | Precision. |
| Corrective EQ | Fruity Parametric EQ 2 / Pro-Q 3–4 | HP 90 Hz; −3 dB at 250 Hz; dynamic at 3 kHz | Cleanup. |
| De-esser 1 | Maximus / Pro-DS | 6–8 kHz, 2–3 dB | Sibilants before compression. |
| FET compression | Fruity Limiter COMP / CLA-76 | 4:1, 4 dB | Peaks. |
| Opto compression | Fruity Compressor / CLA-2A | 3 dB | Leveling. |
| Saturation | Distructor / Saturn 2 / Decapitator | Light, mix 20–30% | Density. |
| Creative EQ + de-esser 2 | Fruity Parametric EQ 2 / Pultec / Pro-DS | +2 dB at 3 kHz, shelf +3 dB at 12 kHz; de-ess 2 dB | Presence and air. |
| Space sends | Fruity Delay 3 (+ Fruity Limiter SC) / LuxeVerb / EchoBoy / Valhalla Plate | Slap + ducked 1/4 + plate 1.8 s | Bigness without distance. |
| Automation | Automation clip / Vocal Rider | Ride per phrase and throws | Intelligibility. |
How to check the result
- Vocal short-term LUFS stable within ±2 dB across phrases (after ride and compression).
- Sibilants (5–9 kHz) no more than 3–4 dB louder than the vowels in the spectrogram.
- Vocal channel correlation +1; stereo returns without negative correlation.
- In mono and at low volume every word stays intelligible.
- Vocal 1–5 kHz band dominant over guitars and synths in the sung passages.
- Reverb/delay returns filtered and ducked: the words don't overlap with the tails.
- Comparison with a reference track: vocal level relative to the snare similar.