What is a cue mix in recording?

A cue mix (also called a foldback or headphone mix) is the blend of audio a performer hears in their headphones while tracking: the backing tracks plus their own live signal. It is separate from the mix listeners will eventually hear — its only job is to make the performer comfortable, in tune, and in time, which directly improves the quality of the takes you capture.

Should I use direct monitoring or software monitoring?

Use direct monitoring whenever the dry signal is usable — vocals, acoustic guitar, DI bass. It routes the input straight to the headphones in hardware with no perceptible delay. Use software monitoring only when the performer needs to hear processing, like an amp sim on electric guitar or a reverb in the singer's ears, and drop the buffer size as low as your computer handles cleanly. Never leave both on at once, or you'll hear a doubled, slap-echoed signal.

Why do I hear an echo on my voice while recording?

You are almost certainly monitoring the same signal twice: once through the interface's direct monitoring and again through the DAW's input monitoring, which arrives late after passing through the audio buffer. Turn one of them off. If there's no doubling but the single signal still sounds delayed, your buffer size is too high for software monitoring — lower it or switch to direct monitoring.

How loud should a headphone mix be for recording vocals?

Comfortable and moderate, not exciting. Loud cue mixes cause ear fatigue within a session, and fatigue shows up as pitch drift. Sharpness in particular usually means the singer can't hear enough of their own voice against the track and is pushing to compensate — the fix is to turn the backing track down rather than raising the voice or the overall level. If the singer keeps asking for everything louder, pull the whole mix down and rebalance instead.

Should singers record with one headphone off?

It's a legitimate, common technique: one earcup slid half off lets the singer hear their voice naturally in the room, which helps many people sing in tune. The tradeoff is bleed — the open earcup leaks click and backing track into the microphone. If your singer prefers it, use closed-back headphones, keep the cue level modest, and check quiet sections of the recorded take for audible click bleed.

Do singers need reverb in their headphones?

Most singers perform better with a little reverb in the cue mix — it smooths the dry, exposed sound of a close mic and helps them judge pitch against the track. The practical question is whether your setup can deliver it. With hardware direct monitoring the signal is split inside the interface before it reaches the computer, so a DAW plugin can never be heard in the cans. You need either an interface whose DSP monitor mixer offers a monitor reverb, software monitoring at a low buffer size, or an outboard reverb in the cue path. If none of those apply, skip it — a dry cue mix beats a delayed one. Where reverb is possible, keep it subtle; mainstream DAWs record the input before track inserts, so it stays monitor-only rather than being printed.

Can I use in-ear monitors instead of headphones for recording?

Yes, and the cue mix itself doesn't change — you build an in-ear mix exactly the way you build a headphone mix, with the same routing and the same direct-versus-software decision. The advantage is bleed: sealed in-ears put far less sound into the air beside the microphone than over-ear cups do, so click bleed on quiet passages largely disappears. The tradeoffs are comfort and the sealed-off feeling some singers dislike. Loose earbuds that don't seal properly give you leakage without isolation.

// Recording

How to Set Up a Headphone Mix for Recording Better Takes

A HomeRecordingPro guide · Updated 2026-08-19 · 9 min read

Quick answerSend the performer a dedicated blend of the backing tracks plus their own live signal, monitored with near-zero delay. Use your interface's direct monitoring for dry sources (or a low buffer size when they need to hear effects), sit their voice or instrument slightly above the track, and keep the overall level moderate to avoid ear fatigue. Reverb in the singer's ears helps, but it only reaches them if your interface has a DSP monitor mixer or you are monitoring through the DAW — hardware direct monitoring splits the signal before any plugin can touch it.

Contents

The Short Answer: What a Good Headphone Mix Does
Cue Mix Explained: Three Things the Performer Needs to Hear
Direct Monitoring vs. Software Monitoring
Step by Step: Setting Up the Cue Mix
The Vocal-Specific Tweaks
Common Cue Mix Problems and Fast Fixes
A Two-Minute Pre-Take Checklist

The Short Answer: What a Good Headphone Mix Does

To set up a headphone mix for recording, send the performer a dedicated blend of the backing tracks plus their own live signal, monitored with as little delay as possible. Use your interface's direct monitoring (or a low buffer size) so they hear themselves in real time, balance their voice or instrument slightly above the track, and keep the overall level moderate. Reverb in the performer's ears helps many singers, but only if your monitoring path can actually deliver it — more on that below. Doing this well is one of the cheapest performance upgrades in home recording.

A headphone mix — engineers call it a cue mix, a foldback mix, or just "the cans" — is what the performer hears while tracking. It is separate from what you'll eventually mix for listeners. Its only purpose is to make the person performing feel comfortable, in tune, and in time.

That matters more than most gear decisions. A singer who can't hear enough of themselves against the track will push to compensate, which usually means straining and drifting sharp. A guitarist hearing a smeared, delayed version of their own playing will drag behind the beat. Fix the cue mix and the takes improve immediately, before you've touched a single plugin.

Cue Mix Explained: Three Things the Performer Needs to Hear

Every usable cue mix answers three questions for the performer:

Pitch reference — something harmonically solid to tune to. Usually keys, guitars, or a pad. Not just drums.
Time reference — drums, a click, or a percussive element that makes the groove obvious.
Themselves — their live vocal or instrument, loud enough to control pitch and dynamics, quiet enough that they still lean into the performance.

If any of the three is missing or buried, the take suffers in a predictable way: no pitch reference means drifting intonation, no time reference means sloppy timing, and not enough of "me" means shouting, straining, or tentative delivery.

A useful starting balance: backing track at a comfortable level, the performer's own signal sitting slightly on top of it, and everything at a moderate overall volume. Louder is not better — sustained loud headphone levels cause ear fatigue within a single session, and fatigue shows up as pitch problems long before the singer notices it.

Direct Monitoring vs. Software Monitoring

This is the fork in the road, and the source of most "I can hear myself but it's delayed" complaints. It also decides what you can and can't put in the performer's ears, so settle it before you build the blend.

Direct monitoring routes the input signal straight from the interface's input to its headphone output, in hardware, before the signal ever reaches the computer. The delay is effectively imperceptible. Most interfaces offer this as a button, a switch, or an input/playback blend knob.

Software monitoring sends the input into the computer, through your DAW, and back out. You hear the signal with any plugins on the track — but you also hear it after a round trip through the audio buffer, which adds latency. At small buffer sizes on a healthy system this can be acceptable; at large buffer sizes it's an obvious, take-wrecking echo.

	Direct monitoring	Software monitoring
Latency	Effectively none	Depends on buffer size and interface
Hear plugins/effects?	No (dry signal only)	Yes
CPU load matters?	No	Yes — pops and dropouts at low buffers
Best for	Vocals, acoustic instruments, most tracking	Amp sims, virtual instruments, effect-dependent parts

The rule of thumb: use direct monitoring whenever the dry sound is usable. Vocals, acoustic guitar, a bass DI the player is happy to hear clean — all fine dry. Switch to software monitoring only when the performer genuinely needs to hear processing to play the part, the classic case being electric guitar through an amp sim. In that case, drop your buffer size as low as your computer tolerates (many systems handle 64 or 128 samples while tracking) and freeze or disable heavy plugins elsewhere in the session. If you're still hearing a delay, work through the causes systematically — our guide to fixing recording latency covers buffer size, sample rate, and driver settings step by step.

One thing to avoid at all costs: hearing both at once. If direct monitoring is on and the DAW track's input monitoring is enabled, the performer hears the dry signal immediately plus a delayed software copy — a short slap echo that makes clean singing nearly impossible. When you use direct monitoring, turn input monitoring off in the DAW (or mute the track's live input).

Step by Step: Setting Up the Cue Mix

1. Get the input level right first

Before touching the headphone blend, set your preamp gain properly — a healthy signal peaking well below clipping, with room to spare. Cue mix balance can't fix a distorted input, and a too-quiet input tempts you into noisy gain-boosting later. If you're unsure where levels should sit, gain staging is the place to start; the cue mix is built on top of it.

2. Choose your monitoring path

Dry signal usable? Enable direct monitoring on the interface and disable input monitoring in the DAW. Need effects — an amp sim, or reverb in the singer's ears — to perform? Use software monitoring at a low buffer size instead. Decide once per session, not per take.

3. Build the backing blend

Play the song and set a rough balance the performer will work against:

Bring up drums/click and the main harmonic instrument first.
Keep the existing lead vocal (if the singer is doubling or harmonizing) present but not dominant.
Pull down anything dense or distracting — big reverbs, busy pads, competing melodies.
Some singers pitch better with less low end in the cans; try dipping the bass if intonation wanders.

4. Blend in the performer

Now bring in their live signal until it sits just above the track. On interfaces with a mix/blend knob, that knob is the whole job: turn toward "input" for more of them, toward "playback" for more track. Ask the performer directly — "more of you, or more of the band?" — and adjust in small moves.

5. Set the overall volume, then leave it alone

Comfortable, not thrilling. If the singer keeps asking for "everything louder," turn the whole mix down first and rebalance — the request usually means fatigue, not preference.

The Vocal-Specific Tweaks

Vocals are the most cue-mix-sensitive thing you'll record, and a few adjustments earn their keep every session.

Reverb in the cans — only if your routing can carry it

A short, subtle reverb in the performer's ears relaxes almost every singer and helps them hear their own pitch against the track. The catch is routing, and most articles skip it.

A reverb plugin only reaches the headphones if the live signal is passing through the DAW — that is, if you're software monitoring. With hardware direct monitoring, the signal is split inside the interface before the computer ever sees it, so no plugin can touch it. On a basic interface with a single mix/blend knob and no onboard mixer, there is simply no way to put reverb in the direct path. If your checklist says "direct monitoring on, input monitoring off," a DAW reverb is not going to arrive.

That leaves three honest options:

An interface with a DSP monitor mixer. Some interfaces build the cue mix inside the hardware via a control application, and some of those include a monitor-only reverb. That's the ideal route — reverb in the cans, latency still effectively zero. Check your interface's manual rather than assuming it's there; plenty of entry-level units have no monitor mixer at all.
Software monitoring at a low buffer. Route the live vocal through the DAW, put a reverb on the track or a send, and accept whatever latency your buffer costs. Fine on a healthy system at small buffers, unusable if you're already fighting delay.
A hardware reverb in the cue path. Uncommon in home setups now, but an outboard unit or a small mixer with an effects send does it with no latency at all.

Keeping it monitor-only is the easy part: mainstream DAWs record the input before track inserts, so a reverb sitting on the vocal track is heard but not printed. Confirm it once on a short test take rather than trusting it blind. And if none of the three routes apply to your setup, skip the reverb — a dry cue mix with the voice sitting properly on top of the track beats a delayed wet one every time. It's a comfort feature, not a requirement.

The one-ear-off trick

Many singers slide one earcup half off so one ear hears the room naturally. It genuinely helps intonation for a lot of people — but it lets the headphone signal leak into the microphone. If your singer does this, keep the click and overall cue level modest, and pan the click away from the open side. Closed-back headphones are the right tool here; open-back designs leak far too much for tracking.

In-ear monitors: the low-bleed option

If bleed is your main problem, in-ears are worth trying. Recording an in-ear mix works exactly like recording with headphones — the routing, the balance, and the direct-versus-software decision are all identical, and only the thing on the end of the cable changes. What changes in the room is leakage: a sealed in-ear monitor puts far less sound into the air next to the microphone than an over-ear cup does, so click bleed on quiet intros largely stops being an issue. That makes them handy for close-miked vocals and quiet acoustic sources.

The tradeoffs are comfort and feel. Good isolation can feel claustrophobic, and some singers dislike hearing so much of their own voice conducted through their own head. The one-ear trick still works — pull one bud out. Loose-fitting earbuds that don't seal give you the worst of both worlds: leakage and no isolation.

Two more small wins

Watch for click bleed. During quiet intros and stops, a loud click ticking away in leaky headphones ends up faintly audible on the vocal track. Solo the recorded take and listen to the gaps before you call it done. More on the full vocal chain — room, mic choice, and technique — in recording vocals at home.

Tuning-sensitive singers may want the pitch reference louder. Piano or acoustic guitar up, atmospheric layers down. Simple, and it works.

Common Cue Mix Problems and Fast Fixes

"There's an echo on my voice." Double-monitoring — turn one of the two paths off, as above.
"I can't hear myself" / "I keep going sharp." These are the same problem. The singer's own signal is buried under the track, so they push harder to hear themselves, and pushing drags the pitch up. Turn the track down rather than the voice up: same balance, less fatigue, less bleed. If the whole mix is already loud, pull everything down and rebuild the balance from there.
"It sounds delayed." You're software monitoring at too high a buffer. Lower it, or switch to direct monitoring.
Crackles and pops while monitoring. Buffer too low for the session's CPU load. Freeze heavy tracks or raise the buffer and switch to direct monitoring.
Only one headphone output, two performers. A simple headphone splitter or a small headphone amp solves it; just know everyone shares one mix. Two headphone outputs carrying genuinely independent mixes is a step-up feature — most entry-level interfaces have a single jack, and some units with two outputs simply mirror the same signal. If separate cue mixes matter to you, check the spec sheet before you buy — worth weighing when choosing your first audio interface.

A Two-Minute Pre-Take Checklist

Input gain set, peaks with headroom to spare.
Direct monitoring on, DAW input monitoring off (or the reverse — never both).
Backing blend: pitch reference and time reference clearly audible.
Performer's signal just above the track.
Overall level moderate; performer confirms it's comfortable.
Cue reverb only if your monitoring path can carry it — a DSP monitor mixer, or software monitoring at a low buffer.
Record ten seconds, play it back, listen for click bleed and distortion.

Do this once at the top of the session and takes two through twenty get better for free. The cue mix is invisible on the final record — but you'll hear its fingerprints on every performance you keep.