Mixing the vocal in context

The principle

The vocal is almost always the most important element of a song: listeners need to understand the words and feel the emotion. Its mix is decided in context: a vocal that is perfect in solo can disappear into the track (or sound pasted on top). The dedicated vocal mixing guide describes the complete chain; here we look at how to make it sit in the mix.

1. Level: the most important decision

Set the vocal level with the whole track playing, at the volume people will listen at; then check at low volume: every word must be intelligible.
By genre: rap and modern pop = vocal well forward; rock = vocal more “inside” the band; R&B = forward but smooth; EDM = vocal forward in the breaks and more integrated in the drops.
After compression, vocal rides (automation) for the words that get lost (see “Mix Automation”).

2. Dynamics: consistency without flattening

Serial compression: a fast compressor (FET type like the 1176, attack 1–10 ms, release 50–100 ms, 3–6 dB of reduction) for the peaks, then an optical one (LA-2A, 2–4 dB) for evenness.
Clip gain before the chain for big differences between phrases.
De-esser at 5–9 kHz (male voices lower, female voices higher).
Parallel for density in the choruses.

3. EQ in context

HP at 70–120 Hz (higher for female voices and for backing vocals).
Mud at 200–400 Hz: cut 2–4 dB if the vocal sounds “boxy”, but only if the music is covering it in that zone.
Nasality at 800 Hz–1.2 kHz.
Presence at 2–5 kHz: a small boost and the matching cut on the instruments (slotting), or better, a dynamic EQ on the instruments driven by the vocal.
Air at 10–14 kHz with a shelf, after the de-esser.
Make decisions by listening to the vocal while the rest plays: the same vocal in a mix with bright guitars needs less presence than you think in solo.

4. Space

Reverb pre-delay (60–120 ms) to keep it close even with reverb.
Slapback (80–140 ms, mono, filtered) for thickness without tails.
Tempo-synced delay (1/4, dotted 1/8) with ducking from the vocal itself: the repeats are heard in the pauses.
Reverb ducking: a sidechain compressor on the return, driven by the dry vocal.
Backing vocals and doubles farther back and wider than the lead vocal.

5. Making room for the vocal in the music

Arrangement: no melodic instruments in the same register during the phrases (call and response).
Sidechain dynamic EQ: −2/−4 dB at 1–4 kHz on guitars, piano, synths, and pads when the vocal sings.
Light volume sidechain (1–2 dB) of the music bus from the vocal in dense genres.
Stereo: vocal in the center, competing instruments moved to the sides.

6. Saturation and character

Light saturation (tape, tube) for harmonics and presence at lower volume; parallel distortion in aggressive genres; character effects (telephone: BP 300 Hz–3 kHz) at specific points.

Example in A minor

Pop at 100 BPM, female vocal E4–A5 over Am – F – C – G: HP 100 Hz; 1176 4 dB GR; LA-2A 3 dB; de-esser at 7 kHz; −2 dB at 300 Hz; +2 dB at 3 kHz; +2 dB shelf at 12 kHz. Plate 1.6 s with 90 ms pre-delay and ducking; dotted 1/8 delay at −22 dB with HP 400 Hz and LP 4 kHz. On the MUSIC bus, a dynamic EQ at −3 dB at 2.5 kHz sidechained from the vocal. In the chorus: doubles panned ±70%, −8 dB below the lead.

Checks

Listen at low volume and in mono: the vocal must stay clear.
Compare with references at equal loudness: relative vocal/music level.
Listen on a phone: sibilance and presence change a lot.

Common mistakes

EQing in solo with excessive boosts at 3–5 kHz (harsh in context).
Too much reverb without pre-delay: distant vocal.
Over-compressed vocal: loses emotion and sounds smaller.
Raising the fader instead of making room in the music.

Exercise

Take a mix where the vocal “doesn't sit”: solve it first with a change to the music (dynamic EQ, arrangement) and only then on the vocal. Compare the result with the one you get by simply raising the vocal.

Links

“Frequency Masking”, “Slotting and Interlocking”, “Mix Automation”, “Depth and Perspective”, “Compression”, “Space in the Mix: Reverb and Delay”.