Answer · Vocals
How loud should my reverb return sit against the vocal?
As a starting point, set the reverb return 6 to 10 dB below the dry vocal. At that level the tail is heard as space behind the voice. Near the vocal's own level the voice sounds distant and washed out, and 15 dB or more below it the reverb is effectively inaudible in a full mix.
Set it by ear, check it by mute
With the full mix playing, bring the return up from nothing until you notice the reverb, then back it off slightly. Then mute the return. If the vocal gets drier and nothing else seems to change, the level is right. If you cannot hear the mute, the reverb is doing nothing. If the vocal jumps forward when it is muted, the reverb was too loud.
What moves the right level
- Decay time. A long tail fills the gaps between words and needs to be lower than a short one.
- Pre-delay. 20 to 60 ms between the word and the start of the reverb keeps the vocal in front, which lets the reverb be louder without pushing the voice back.
- Tone. A bright reverb is heard more than a dark one at the same level.
- The arrangement. A sparse verse exposes a reverb that a dense chorus hides, so the same send can be too wet in one and lost in the other. Automate it.
Equalise the return
High-pass the reverb at 200 to 300 Hz and low-pass it at 6 to 8 kHz. Without the low-mids it adds less mud, and without the top it does not emphasise sibilance. A filtered reverb can sit a few decibels higher before it gets in the way.
In Zammoth
The amount is set in two places: the vocal's send knob, which runs from off to 0 dB, and the return's own fader. The engineer measures the loudness of a return the same way it measures a track, so the gap between the vocal and its reverb is a number you can ask for:
Send levels can be automated, for a verse that should be drier than the chorus. See the manual on sends and returns.
Zammoth is a DAW for Apple silicon Macs with an AI audio engineer built in. The beta opens soon.
Join the beta