Black Friday//Up to 28% off plugins & packs//Essentials $249//All-Access first year $99//Ends DEC 1//
MixingVocalsDelay

How To Duck A Vocal Delay So The Words Stay Clear

A vocal delay that repeats over the next line turns words into mush. Timing, ducking, filtering and throws keep the echo in the gaps where it belongs.

All articles
By Nolvra

The Echo Is Fighting The Next Word

A vocal delay goes wrong in one specific way. The singer finishes a word, the repeat of that word arrives just as the next one starts, and the two land on top of each other. On their own both are fine. Together the consonants smear, the listener loses the lyric, and the usual reaction is to turn the delay down until it stops being audible at all. That throws away the reason the delay was there. The better fix is to keep the echo at full level and move it to the moments where nothing else is being said.

Why Turning It Down Does Not Work

A delay return has two jobs that pull against each other. Between phrases it should be loud enough to be heard as a part, a tail that carries the last word into the space before the next line. While the vocal is singing it should be close to silent, because anything that repeats a syllable over a new syllable competes with the one thing the listener is following.

A fixed send level can only pick one of those. Set it for the gaps and it masks the words. Set it for the words and the gaps go dry. Every technique below is a way of letting the level change over time instead: the echo gets out of the way while the voice is present and comes back up the moment the voice stops.

Four Ways To Keep The Repeats In The Gaps

Time The Repeats To Land In The Rests

Before reaching for any ducking, listen to where the repeats fall. A delay synced to a note value that matches the phrasing, a dotted eighth or a quarter for most rap and pop verses, will often put the first repeat into a natural rest. Set the feedback low enough that the second and third repeats are gone before the next line. Three repeats is plenty for a vocal that is still moving; a long tail belongs to the end of a section.

Duck The Return From The Dry Vocal

Put a compressor after the delay on the return and feed its sidechain from the dry vocal. When the singer is singing the compressor pulls the echo down; when they stop it releases and the tail comes back. Start with a fast attack so the first consonant is protected, and set the release by ear against the gap length: too short and the echo pumps up between syllables, too long and the tail never recovers before the next word.

Filter And De-Ess The Return, Not The Vocal

Repeats of sibilants and low body are the parts that clutter fastest. High pass the return so it does not thicken the low mids, low pass it so the repeats sit behind the dry voice, and de-ess before the delay so an S is not echoed three times. Inside the feedback loop a filter darkens each repeat a little more than the last, which reads as distance. The same reasoning is behind keeping effects on a return in reverb send versus insert.

Throw Instead Of Leaving It On

For the most exposed lines, skip the constant delay and automate a throw: open the send for the last word of a phrase only, and close it again before the next line. The echo becomes an event the arrangement can use rather than a wash under everything.

Where YONDER Fits

Building the ducking chain by hand works in every DAW and costs a compressor, a sidechain route and some time with the release knob. YONDER folds that job into the delay itself. Its Vocal Gap mode holds the repeats while the voice is singing and opens them between words, at phrase ends, or both, so the timing decision is made on the voice rather than on a release setting. It is off in the Default preset and you switch it on when you want the echo to wait.

The rest of the chain is there too: ducking from an external sidechain if another source should push the echo down, a de-esser, and filters that can sit inside the feedback loop. Throw and Kill send or cut the input on demand for the moments when a throw is the right move. Where the delay goes in your vocal chain order still matters, and a return is still the safer default than an insert.

Checks Before You Print The Delay

  • Solo nothing. Judge the echo with the beat playing, because a delay that is clear on its own can still mask in the mix.
  • Read the lyric along with the playback. If any word gets harder to follow when the delay is up, the ducking is too slow.
  • Count the repeats you can hear before the next line. More than three under a busy verse is usually clutter.
  • Check the tail at section ends separately. That is where the echo is allowed to be loud and long.
  • Listen in mono once. Ping pong repeats that look wide can collapse onto the centre and sit right on the dry voice.

Let The Echo Wait Its Turn

YONDER's Vocal Gap holds the repeats while the voice sings and lets them out between words, at phrase ends, or both, across twelve delay modes.

See YONDER

FAQ

How do you stop a vocal delay from muddying the lyrics?

Lower the echo while the voice is singing and let it return in the gaps. A compressor on the delay return, keyed from the dry vocal, does this in any DAW; filtering and de-essing the return removes the parts of each repeat that clutter fastest.

What delay time works best on vocals?

A synced note value that matches the phrasing, often a dotted eighth or a quarter, puts the first repeat into a natural rest. Keep the feedback low enough that the repeats are gone before the next line starts.

Should vocal delay be on a send or an insert?

On a send in most mixes, so the dry vocal stays untouched and the return can be filtered, ducked and automated on its own.

Guides

You may also like

Our latest news

More guides