Welcome back. Today we’re doing a super usable drum and bass vocal trick in Ableton Live: delay throws.

A delay throw is when the delay only happens for a moment. Not the whole vocal swimming in repeats, just a little “echo answer” on the last word of a line, a hype ad-lib, or a gap before a fill. It’s one of those techniques that instantly makes a vocal feel more professional in DnB, because you get size and space without losing punch.

We’re going to build this in a clean, repeatable way using mostly stock devices, and the main idea is simple:
keep your vocal clean and upfront, then send only specific words to a dedicated delay return, and make that return behave in the mix with EQ and ducking.

Alright, let’s set the context first, because timing matters a lot in DnB.

Most DnB is around 170 to 175 BPM. Delay throws feel best when they’re synced to the grid, and the usual go-to times are one eighth for tight energy, one quarter for bigger dramatic moments, dotted eighth for that bouncy jungle flavor, and sometimes one sixteenth for quick stutters.

Here’s an easy rule: if your drums are busy, like rolling hats and lots of ghost notes, keep the delay tighter and keep the feedback lower. Tight delay, quick fade. That’s how you keep the groove clean.

Now Step 1: prep the vocal track. The throw works best when your main vocal is controlled first.
On your vocal track, we’re going to do a simple control chain.

First, EQ Eight. High-pass the vocal somewhere around 80 to 120 Hz. The exact number depends on the voice, but the goal is: no rumble, no low-end junk. If it sounds muddy, try a small cut around 200 to 400 Hz.

Next, a compressor. Either Compressor or Glue Compressor is fine. Keep it light: ratio somewhere between 2:1 and 4:1. Aim for about 2 to 5 dB of gain reduction on peaks. We’re not trying to crush it, just keep it stable.

And if your vocal has sharp S sounds, handle that now. If you have a de-esser, great. If not, you can use Multiband Dynamics gently to tame the harsh top. This matters because delay repeats will repeat your sibilance too, and that can get spitty and annoying fast.

Cool. Now we build the fun part: a dedicated delay throw return.

Create a Return Track in Ableton. Name it something like “A - Throw Delay” so you always know what it is.
On that return, add Echo or Delay. Echo is usually more vibey and modern, so we’ll start with Echo.

In Echo, turn Sync on. Set the time to one eighth to start. Set feedback around 25 to 40 percent. You want it to fade out naturally, not loop forever and talk over your next lyric.
Dry/Wet should be 100 percent because this is a return.
If you want stereo width, leave it in stereo mode, but we’ll manage the low end later.
And for now, keep wobble and noise very low or off. You can add character later once the basics are tight.

If you prefer the simpler Delay device, do the same concept: left and right at one eighth, or try making the right side dotted eighth for movement. Feedback around 20 to 35 percent.

Now Step 3, and this is non-negotiable: EQ the delay return.
Put EQ Eight after the delay on the return.

High-pass the delay somewhere around 150 to 250 Hz. If your bass is huge, don’t be shy going even higher, like 250 to 350. This is one of the biggest reasons delay throws stay clean in DnB: you’re protecting the sub and the weight of the track.

Then low-pass the delay around 6 to 10 kHz. This makes the repeats sit behind the lead vocal instead of competing with it. Bright delays can sound cool solo, but in a DnB mix they can steal attention from your snare crack and your vocal consonants.

If the delay still pokes out in an annoying way, try a small notch around 2.5 to 4.5 kHz.
And here’s a nice little coach trick: if you low-pass the delay and it becomes too muffled to understand, do a gentle bell boost around 1 to 2 kHz, like 1 dB, maybe 2 dB. That’s the “mid pocket” that keeps the words readable without making it harsh.

Next, Step 4: make the throw move with the groove using ducking.
This is where the delay feels like it’s part of the rhythm, not just floating on top.

Add a Compressor after the EQ on the return. Turn on Sidechain.

Option one: sidechain the delay to the lead vocal. This is great for intelligibility.
Set the sidechain input to your vocal track. Ratio around 4:1. Attack around 2 to 10 milliseconds. Release around 80 to 180 milliseconds.
Then pull the threshold down until you see about 3 to 8 dB of gain reduction while the vocal is present.
What you’ll hear is: while the vocal is talking, the delay tucks out of the way, and then in the gaps, the delay blooms up. That’s the “professional” behavior.

Option two: sidechain to the kick and snare or a drum bus. This is classic DnB cleanliness.
Set a slightly faster release, like 60 to 120 milliseconds, so it pumps rhythmically with the drums. The delay starts to groove around the transients instead of smearing them.

Now Step 5: create the actual throw. This is the core technique.
We’re going to automate the send amount from the vocal track to Return A.

Go back to your vocal track and find Send A. Hit A to enter automation mode.
Most of the time, the send should be all the way down, basically off. Then, just on the word you want, you pop it up for a moment. A good starting range is anywhere from about minus 12 dB up to minus 3 dB, depending on how obvious you want the throw.

A huge tip here: don’t make the automation a perfect square on and off. That tends to sound like you slapped an effect on.
Instead, draw a quick ramp up into the last syllable, and then a faster drop right after the word ends. It sounds like you “caught” the end of the phrase and tossed it into space.

And here’s another timing tip that fixes a lot of “why does my throw feel late?” moments:
before you change the delay time, try nudging your automation slightly earlier. Like 10 to 30 milliseconds ahead of the transient. That tiny move can turn a draggy throw into a snappy one.

If you’re working with repeating vocal chops, clip envelopes can be easier than drawing automation every time.
Open the vocal clip, go to Envelopes, choose Mixer, then Send A, and draw the throw inside the clip so it repeats consistently.

Now Step 6: widen it in a modern way without wrecking mono.
Put Utility at the end of the return.
Turn on Bass Mono somewhere around 120 to 200 Hz. That keeps the low end centered and avoids phase weirdness.
Then you can increase Width a bit. Try 120 to 160 percent, but keep it subtle. If it starts sounding wide-but-wrong, bring it back down and make sure your low frequencies are staying mono.

Optional Step 7: a hint of reverb after the delay.
Sometimes a delay throw sounds better with a tiny tail, like it’s in a space instead of just repeating.
Add Reverb after your ducking and EQ, or experiment with placement, but keep it tasteful.
On the reverb, keep dry/wet low, like 10 to 20 percent, decay around 0.8 to 1.5 seconds, high cut around 6 to 9 kHz, and low cut around 200 Hz.
In rolling DnB, too much reverb smears transients, so the delay should be the main space, and the reverb is just polish.

Now a quick but important set of coach notes that’ll save you headaches.

First, gain staging on the return. Feedback can accumulate, and suddenly your repeats jump out or clip.
Put a Utility at the very top of the return and set it to minus 6 dB while you build. Once everything’s controlled, adjust your send levels up. This keeps the return predictable.

Second, pre-fader versus post-fader sends.
Right-click the send knob on the vocal and choose Pre/Post.
Post-fader is the default, and it means your throw level follows the vocal fader. That’s great if you’re still mixing the vocal level.
Pre-fader is cool for “ghost throws.” Like, you pull the vocal down or mute it right after the word, but the delay keeps going. That can be super hype for transitions.

Third, if your vocal is really dynamic and your throws feel inconsistent, you can feed the delay a more consistent signal.
Duplicate the vocal track, compress the duplicate harder, then turn that duplicate off to the master so you don’t actually hear it dry. Use it only as the send source into the delay return. Your main vocal stays natural, but your throws trigger consistently.

Now Step 8: timing upgrades. One return is great, but two returns is where arrangement energy starts to pop.
Duplicate your return.

Make Return A your tight throw: one eighth, feedback 25 to 35 percent.
Make Return B your big throw: one quarter, and use lower feedback, like 15 to 25 percent, because the gaps are bigger and you don’t want it to clutter the next line.

Then you can automate Send A for fast sections and Send B for end-of-phrase drama, especially into fills and transitions.

Let’s quickly cover common mistakes so you can avoid the classic beginner traps.

One: leaving the send up too long. That turns throws into a constant wash.
Two: no EQ on the delay return. Low-end repeats will absolutely ruin bass clarity in DnB.
Three: feedback too high. The delay keeps talking over the next lyric and it kills impact.
Four: delays too bright. Sibilance becomes a repeated ice-pick. Low-pass the repeats.
Five: no ducking. Without ducking, delay competes with your vocal and snare, and the whole mix gets crowded.

If you want a darker, heavier vibe, here are a couple fast upgrades.
Add Saturator before the EQ on the return. Drive around 2 to 6 dB, soft clip on. Then low-pass to keep it gritty but controlled.
Or do a “telephone throw” for contrast: put Auto Filter before the delay, band-pass mode, modest resonance, and sweep the cutoff down during the throw. Suddenly the repeats feel like a special effect, not just more vocal in the same space.

And for arrangement, use throws as section markers, not on every line.
Try almost no throws in bars 1 to 4, then small tight throws in 5 to 8, then reset a bit, then bigger throws in 13 to 16 leading into a transition or drop. Restraint makes the throws feel intentional.

Now let’s do a quick 10-minute practice so you actually lock this in.

Load an 8 to 16 bar DnB loop at 174 BPM with drums and bass.
Add a short 2 to 4 bar vocal phrase, even spoken is fine.
Build Return A with Echo using the starter settings.
Automate Send A only on the last word of bar 2, and the last word of bar 4, with bar 4 being a bit bigger.
Add sidechain ducking from the vocal, aiming for around 4 to 6 dB of duck.
Then bounce or resample and listen for two things:
can you still clearly hear the kick and snare transients, and does the throw feel like a moment rather than a constant effect?

If you’ve got time, duplicate the return and make a one quarter version for the last bar before a drop. That’s an instant “lift.”

Let’s recap the whole concept in one sentence:
Delay throws are automated sends to a delay return, used only on key words, shaped with EQ and ducking so they hype the vocal without messing up the DnB groove.

If you tell me what substyle you’re making, like liquid, jump-up, neuro, or jungle, and whether the vocal is sung or spoken, I can suggest a tight little throw palette: exact timings, feedback ranges, and how bright or wide to go for that lane.