Mixing for Multiple Playback Systems, Done Properly
A symptom-by-symptom guide to mixing for phones, earbuds, cars and monitors: how to spot translation problems, trace the cause, fix them and prevent them.
Tomas Lindqvist
Mixing for multiple playback systems is the practice of checking a mix on several different kinds of speaker, and on headphones, so that it holds up on the systems the audience will actually use rather than only on the ones in the studio. A podcast will be heard on earbuds on a train, a product video on a laptop in an open-plan office, a radio spot in a car, and a training module through the single speaker of a conference-room display. The mix has to survive all of them, and it only gets one chance on each.
The core problem is simple to state: every monitoring system lies in its own way. Studio monitors in a good room exaggerate nothing and hide little, but almost nobody listens that way. Headphones present an unnaturally wide stereo image and an intimate low end. A phone speaker throws away most of the bass and sums everything to one channel. A mix that has only been heard on one system is a mix nobody has tested, and the faults it carries will surface on the first real listener's device.
This guide is written as a troubleshooting manual. It starts with a diagnostic map, sets up a listening method you can trust, then works through the common symptoms one by one: how to recognize each, what usually causes it and how to fix it. It closes with a worked example, a prevention checklist and guidance on when an outside engineer is worth the fee. It is aimed at marketers and in-house teams who approve audio, and at editors and producers who do the mixing themselves.
The Translation Symptom Map: What You Hear, Why, and What to Do
Most translation failures fall into a small number of recognizable patterns. Before reaching for an equalizer, identify which pattern you are dealing with, because the same complaint ("it sounds thin") can have very different causes. Thin on a phone may be a bass-arrangement problem; thin only in mono is almost always a phase problem; thin compared with a reference track may simply be a loudness mismatch that fools the ear. The table below is the short version of the rest of this article.
| Symptom | Where it shows up | Most likely cause | First fix to try |
|---|---|---|---|
| Bass line or kick disappears | Phones, laptops, small speakers | Low-end energy sits only in the fundamental, with no upper harmonics | Add controlled harmonic content in the 100 Hz to 1 kHz range; check on a small single speaker |
| Boomy, muddy low end | Cars, earbuds, consumer headphones | Room or headphone masked a buildup in the low mids or sub bass | Check with a spectrum analyzer and on a second system; cut rather than boost |
| Vocal or instrument gets quieter or hollow | Mono speakers, phones, smart speakers | Phase cancellation between channels or between multi-mic sources | Fold to mono, align or polarity-flip the offending source |
| Reverb or width vanishes | Mono playback | Stereo wideners and wide effects built from out-of-phase content | Reduce width, use mono-compatible widening methods |
| Harsh, tiring to listen to | Earbuds, small speakers, laptops | Excess energy in the upper mids, often exaggerated by loud mixing | Check at moderate level; tame 2 to 5 kHz selectively |
| Dialogue buried under music | TVs, laptops, noisy environments | Music and voice compete in the same range; mix judged on full-range monitors | Carve space for voice; ride the music bed with automation |
| Sounds small next to references | Every comparison | Reference played louder than the mix | Match loudness before judging tone or impact |
| Great last night, wrong this morning | Any system | Ear fatigue and approval at the end of a long session | Leave time between finishing and approving |
Read the table from left to right when a client or colleague reports a problem: note exactly which system they heard it on, match the symptom, test the likely cause, and only then change the mix. Changing things before you know the cause is how a mix that was nearly right becomes one that is wrong everywhere.
Setting Up a Listening Method You Can Trust
Troubleshooting only works if your checks are valid. Two errors invalidate most translation checks before they begin: comparing at mismatched levels and listening too loud. Fix both first, then add the playback systems.
The minimum set of playback systems
The widely taught reference practice is to check on three kinds of system: full-range monitors, headphones and a small speaker. Each covers a different blind spot.
- Monitors in a treated or at least controlled room are your main decision-making tool for balance, depth and tone. Their accuracy depends heavily on placement; our guide to studio monitor placement covers the geometry and boundary effects that most often cause false bass readings.
- Closed or open headphones reveal detail: clicks, edits, noise, distortion and reverb tails that monitors in an imperfect room can smear. They also exaggerate stereo width and isolate you from room problems, which is why they are a check rather than the whole process. Choosing and using them well is covered in headphones and monitoring for editing.
- A small single speaker, such as a compact mono check speaker, a phone, a tablet or a laptop, tells you what survives when the bass is gone and the channels are summed. It is the fastest way to hear whether the voice, melody and rhythm still read.
Beyond those three, add systems that match the delivery. A radio spot deserves a car check. An on-hold message or phone prompt deserves a listen through an actual handset, because the band-limited phone channel is its own world; our guide to IVR and phone audio goes into that channel in detail. Social video deserves a listen on a phone at a typical scrolling volume.
Matching loudness before every comparison
The ear hears louder as better: fuller bass, crisper top, more impact. That bias is strong enough that a comparison between two sources at different levels tells you almost nothing about their tone. Comparisons are invalid unless levels are matched. In practice this means measuring integrated loudness in LUFS, the unit defined by ITU-R BS.1770 and used by broadcast recommendations such as those published by the European Broadcasting Union, and then trimming the louder source down until both read the same. Many reference plugins do this automatically; a gain plugin and a loudness meter do the job just as well.
Choosing a listening level
Mixing loud flatters the mix and tires the ears. Human hearing is less sensitive to low and very high frequencies at quiet levels, as described by the equal-loudness contours standardized in ISO 226, so turning up makes the extremes sound richer than they really are. A mix balanced at high volume tends to sound thin and dull when played back at the moderate levels most listeners use. Pick a moderate, comfortable reference level, mark it on your monitor controller or interface, and return to it. Short checks at lower and higher levels are useful; long sessions at high level are not.
A repeatable check routine
- Set your monitors to the marked reference level.
- Load one or two reference tracks that suit the genre or content type and loudness-match them to your mix.
- Play the same passage, around 30 to 60 seconds with the densest section included, on each system in turn.
- Fold to mono and listen to the same passage again on monitors.
- Write down what you hear on each system before changing anything.
- Make one change at a time and re-check on the system that exposed the problem and on the main monitors.
Written notes matter more than they seem. When you move between four systems, memory is unreliable, and it is easy to fix a problem on one system while quietly creating a new one on another.
Symptom: The Bass Line and Kick Vanish on Phones and Laptops
How to recognize it
On the monitors the low end feels solid. On a phone or laptop, the bass line becomes a vague suggestion or disappears entirely, the kick drum turns into a click or nothing, and the music feels as though it is floating with no foundation. In spoken-word content with a music bed, the bed may seem to drop out whenever the voice stops.
Likely causes
Small speakers physically cannot move enough air to reproduce deep bass. Their useful response usually starts somewhere in the low hundreds of hertz, varying by device. If the bass part lives almost entirely in its fundamental, say a sub-heavy synth or a very clean sine-like bass around 40 to 80 Hz, a phone has nothing to reproduce. On full-range monitors or in a room with a bass buildup, you never noticed, because the energy was there.
A second cause is aggressive low-pass filtering or heavy saturation-free processing on the bass that strips out the upper harmonics. The ear can infer a low note from its harmonics even when the fundamental is missing, an effect often called the missing fundamental. Remove the harmonics and there is nothing for the ear to work with.
The fix
- Listen on the small single speaker while adjusting. Do not judge this fix on the monitors.
- Add harmonic content to the bass with gentle saturation, a parallel distorted copy blended low, or an upper-harmonic enhancer. Aim for audibility on the small speaker without audible grit on the monitors.
- Give the kick a defined click or beater attack in the upper mids so it registers on small speakers, and let its low thump do the work on larger systems.
- Check the arrangement. If the only element carrying the chord root is a sub bass, consider doubling the part an octave up quietly with another instrument.
- Recheck on monitors and headphones to make sure the added harmonics have not made the low mids congested.
Quick fix: Duplicate the bass track, run the copy through saturation, high-pass it well above the fundamental, and blend it in underneath the original until the bass line becomes audible on a phone. The original keeps the weight; the copy carries the pitch to small speakers.
How to prevent it
Choose bass sounds with some harmonic character from the start and check on a small speaker early in the mix, not only at the end. When licensing music for video, audition candidate tracks on a phone as well as on monitors; our guide to choosing music for video covers what else to listen for.
Symptom: Boomy, Muddy Low End in Cars and on Earbuds
How to recognize it
This is the opposite complaint. In the car, the bass overwhelms everything, the voice sounds as if it is behind a blanket, and the listener turns the bass control down. On consumer earbuds, which are often tuned with a bass lift, the low end feels bloated and the kick and bass blur together.
Likely causes
- A room null at the mix position. Untreated rooms create peaks and dips in the low end depending on where you sit. If you sit in a dip at, say, 100 Hz, you keep adding energy there to compensate, and every other system plays it back as boom.
- Monitors that cannot reproduce the lowest octave. If your speakers roll off early, you cannot hear sub-bass accumulating from kick, bass, pads and effects returns, and it piles up unseen.
- Low-mid buildup. Many sources contribute energy in the 150 to 400 Hz region. Individually they sound fine; summed, they turn the mix muddy.
- Reverbs and effects returns left full range. Low-frequency reverb adds wash without adding anything useful.
The fix
First confirm the problem on a second trusted system, such as good headphones, because the headphone presentation removes room effects. Then look at a spectrum analyzer on the mix bus and compare the low-end contour with a loudness-matched reference in the same genre. You are not trying to copy the curve, only to spot a region that is clearly out of family. Cut before you boost: high-pass sources that have no useful low end, such as vocals, guitars, overheads and effects returns, being careful not to thin them. Use narrow cuts on resonant buildups and a gentle broad cut in the low mids if the whole mix is thick. Any processing you apply to the whole mix should be conservative; our guide to mix bus processing explains why broad moves on the bus are safer than aggressive ones.
How to prevent it
Measure your room, even roughly, with a measurement microphone and free analysis software, so you know where the peaks and nulls are. Improve placement and treatment where you can. Until the room is trustworthy, lean more on headphones and a car check for low-end decisions, and less on the monitors.
Symptom: Vocals, Instruments or Effects Drop Out in Mono
How to recognize it
Many playback systems sum to mono: phones held in portrait, smart speakers, many laptop and television speakers in practice, public address systems, and some broadcast and streaming chains. When the mix is folded to mono, something gets quieter, thinner or hollow and phasey. A double-tracked guitar loses its body, a stereo piano sounds like it is in a tube, or, worst case, a lead element partly cancels.
Likely causes
Phase problems are often inaudible in stereo and obvious in mono. In stereo, the ear interprets differences between the channels as width or space. In mono, those same differences add and subtract as waveforms, and components that are out of phase cancel. Common sources include:
- Two microphones on one source at different distances, such as a guitar amp with a close and a room mic, or a lavalier and a boom on the same speaker in a video shoot.
- A polarity-inverted cable or microphone in a multi-mic setup.
- Stereo samples or synth patches built with out-of-phase layers or detuned voices.
- Very short delays, under roughly 30 milliseconds, used to widen a mono source, which produce comb filtering when summed.
The fix
Put a mono switch on your monitor path or a utility plugin on the master that folds to mono, and use it deliberately. Solo pairs of related tracks and fold them to mono to find the offender. For multi-mic sources, zoom in on a transient and align the tracks by nudging one against the other, or use an alignment plugin, then try flipping polarity and keep whichever setting gives the fuller sound. For dialogue recorded on two microphones, choosing one mic per line is often cleaner than blending both. A correlation meter helps: sustained readings toward negative indicate content that will cancel in mono.
If an element thins out in mono and you cannot find the cause quickly, solo it and toggle its polarity while listening in mono. If that does not help, reduce its width, or replace the stereo version with one side of it panned where you need it.
How to prevent it
Check the mix in mono as a matter of routine, not as a final ritual. A mono check after each significant change takes seconds. The choices that decide how a mix behaves in mono start at recording and arrangement; our guide to stereo imaging and mono compatibility covers those decisions in depth.
Related symptom: wide stereo effects vanish or cancel
Recognizing vanishing width
On headphones the mix is spacious and immersive. On a single speaker, the reverb disappears and the mix turns dry and small, a backing vocal pad vanishes, or a synth that sounded enormous becomes a thin line. Sometimes the result is worse than simple loss: the element becomes phasey and unpleasant.
Why wideners fail in mono
Stereo wideners typically work by boosting the difference between left and right, the side signal. Side information is exactly what disappears when the channels are summed. Aggressive widening, mid/side processing that pushes the sides far above the center, and some stereo reverbs with strongly decorrelated outputs all make effects that exist only in the difference between channels. Headphones exaggerate this width further, which is one reason mixing exclusively on headphones is a common mistake: the widths that sound tasteful on headphones are often excessive everywhere else.
Fixing width without losing mono
- Pull back wideners until the element still reads in mono, then decide whether the stereo version still sounds wide enough. It usually does.
- Create width with arrangement instead: two genuinely different performances panned left and right survive mono far better than one performance artificially widened.
- Use reverbs and delays whose stereo spread comes from different early reflections or delay times long enough to avoid comb filtering, rather than from phase tricks.
- In mid/side processing, keep the low end in the mid channel only; wide bass is rarely useful and often causes trouble on vinyl cutting, club systems and mono devices.
Preventing width problems
Set a personal rule: no widening plugin goes on without a mono check immediately after. For video work, remember that stereo decisions interact with dialogue clarity, localization and downmixing; a clean, centered dialogue stem also makes alternate language versions and downmixes far easier to produce.
Myth: Almost everyone listens on headphones now, so mono compatibility no longer matters and a headphone mix is good enough.
Reality: Headphone listening is common, but so is playback on phone speakers, smart speakers, laptops, televisions, cars and public spaces, many of which effectively sum to mono or narrow the image heavily. A headphone-only mix is tuned for the one system that exaggerates width the most, which is precisely why it fails elsewhere.
Symptom: The Mix Is Harsh and Tiring on Earbuds and Small Speakers
How to recognize it
Listeners describe the audio as shrill, piercing or tiring, and they turn it down within a minute. Sibilance stabs on "s" and "t" sounds, cymbals splash, and a bright synth or an edgy electric guitar bites. On monitors at your usual level, it may sound merely present and exciting.
Likely causes
- Too much energy in the upper mids, roughly 2 to 5 kHz, where hearing is most sensitive. Small speakers often have their strongest output in this region, so an excess that is tolerable on monitors becomes unbearable on a phone.
- Mixing too loud. Ironically, very loud listening can dull your perception of harshness through fatigue, so you add more brightness to compensate.
- Overuse of saturation and clipping added for loudness, which generates upper harmonics.
- Stacked brightening EQ on many individual tracks, each reasonable in isolation.
The fix
Drop to a moderate or low listening level and listen on the small speaker and on earbuds. Identify which element is harsh by muting candidates, rather than cutting the whole mix. Use a de-esser on vocals and dialogue, dynamic EQ on elements that are only harsh when they get loud, and narrow static cuts on resonances. If the whole mix is bright, a small broad reduction on the bus is fine, but first check whether one or two tracks are responsible. Resources from iZotope on mixing and reference practice describe how to use spectrum and masking displays to locate these conflicts, but always confirm with your ears on the system that exposed the problem.
Quick fix: Mute elements one at a time while listening on earbuds at a low level. The harshness usually belongs to one or two tracks; a de-esser or a dynamic EQ band on those tracks fixes it without dulling the rest of the mix.
How to prevent it
Take regular breaks, keep a moderate level, and avoid solving every perceived dullness with a high-shelf boost. Brightness is relative: a mix often sounds dull only because you have just been listening to something brighter or louder.
Symptom: Dialogue Gets Buried on TVs, Laptops and in Noisy Places
How to recognize it
For video, podcasts, e-learning and ads, the most damaging translation failure is not tonal but intelligibility. On full-range monitors in a quiet room, the voice and music sit nicely together. On a laptop in a café, or a TV across a living room, the listener catches every third word.
Likely causes
Music beds often share the voice's critical range, roughly the upper mids where consonants live. On monitors, the stereo image and full bandwidth help the ear separate them. On a small, mono, band-limited speaker in a noisy environment, those cues are gone and the music masks the voice. Wide dynamic range makes it worse: quiet lines fall under the background noise of the listening environment.
The fix
- Ride the music bed under the voice with automation, rather than setting a single static level. Our guide to volume automation and clip gain covers how to shape dialogue and music levels precisely.
- Carve a modest dip in the music in the upper mids while the voice is present, using dynamic EQ or sidechain-triggered EQ.
- Control the dynamic range of the dialogue so quiet words stay above typical background noise, without squashing it.
- Keep the dialogue centered and consistent; do not widen it.
Content for mobile and social feeds is especially exposed, because much of it is heard on phone speakers at low volume; audio for short-form video covers the specific choices for that format.
How to prevent it
Do an intelligibility pass on the small speaker at a low level with some background noise in the room, such as a fan or an open window. If you can follow every word, most listeners will.
Symptom: Your Mix Sounds Small and Dull Next to Commercial References
How to recognize it
You switch from your mix to a commercial release and yours instantly seems quieter, narrower, duller and weaker in the low end. The instinct is to reach for EQ, compression and a limiter.
Likely causes
Very often, at least part of this impression is a level difference. Commercial masters are often delivered loud, and the louder source always sounds bigger and brighter. Comparing against a reference that is louder is one of the most common mistakes in mixing, and it leads directly to over-compression and harshness. The remaining difference, once levels match, is real and useful information: genuine tonal balance, density and dynamics.
The fix
Measure integrated loudness on both and turn the reference down to match your mix, not your mix up to match the reference. Then compare section by section: the densest chorus of the reference against your densest chorus, the intro against your intro. Only after matching should you decide whether the low end, the vocal level or the top end genuinely differs. Our guide to using reference tracks explains how to choose references that fit the job and how to compare them without being misled.
Also check the reference on each of your playback systems. It will have its own weaknesses on the phone and in the car. Its value is showing you how a finished product behaves on each system, so you can judge whether your mix is out of family or simply behaving like everything else on that device.
How to prevent it
Build loudness matching into your session template so every comparison is valid by default, and choose your final delivery loudness based on the platform or specification, not on the reference.
Watch out: Many problems that seem to call for more processing are really listening problems. Before you add a compressor, limiter or EQ boost in response to a comparison, confirm three things:
- The reference and the mix are matched in integrated loudness.
- You are listening at your moderate reference level, not louder.
- You have not been listening continuously for hours.
Symptom: It Sounded Great Last Night and Wrong This Morning
How to recognize it
You finished late, felt confident, and sent it for approval. The next morning, on the same monitors, the vocal is too loud, the top end is fizzy and the reverb is excessive. Or the client plays it back on their system and asks why it sounds so different from the version they heard in the session.
Likely causes
Hearing adapts. After hours of listening, especially at high level, sensitivity to high frequencies drops and the ear grows accustomed to whatever balance it has been hearing. Familiarity also dulls judgment: after dozens of passes, you stop hearing the mix as a listener would. Approving a mix at the end of a long session is a reliable way to ship these errors.
The fix
Leave time between finishing and approving a mix, ideally overnight, and at minimum long enough to rest your ears and do something else. Return and listen once through from the start without stopping, on one system, taking notes but not touching anything. Then make targeted changes. Guidance on listening conditions from bodies such as the Audio Engineering Society reinforces the same principle: consistent, controlled conditions and rested ears produce more reliable judgments than long, loud sessions.
How to prevent it
Schedule the approval listen into the project timeline, rather than squeezing it into the end of the mixing day. Take short breaks every hour or so during mixing. When clients attend sessions, set expectations that a final version will follow after a fresh-ear pass.
Worked Example: Tracing a Translation Problem in a Product Video Mix
The following is an illustrative example, not a real client project. It shows how the symptom-first method works in practice on a typical brief.
A team is finishing a 90-second product video for social feeds and a website hero section. The mix contains a voiceover, a licensed electronic music bed with a sub-heavy bass, a few sound effects and a stereo-widened synth pad. The editor mixed on headphones at home and on a pair of small monitors in an untreated office. Delivery is a stereo file measured to the loudness the team uses for web video, around -16 LUFS integrated, with true peak below -1 dBTP; the exact target depends on the platform and the team's own specification.
The marketing lead reports three complaints after watching on a phone: "the music has no energy," "the voice sounds a bit hollow in places," and "it's quieter than our competitor's video." The editor runs the check routine and records findings on each system.
| System | What the editor heard | Symptom category | Action taken |
|---|---|---|---|
| Headphones | Balanced, wide, strong bass | None apparent | No change; headphones were the only system used previously |
| Office monitors | Bass slightly light, pad very wide | Possible room null in the low end | Low-end decisions deferred to headphones and a car check |
| Mono fold on monitors | Voice hollow when pad is playing; pad nearly disappears | Phase and width | Widener on pad reduced; pad's low mids cut; voice now solid in mono |
| Phone speaker | Bass line inaudible; kick only a faint tick | Missing harmonics | Saturated high-passed copy of the bass blended in; kick attack lifted around the upper mids |
| Car | Low end full but acceptable after bass change | None after fix | Confirmed bass copy did not cause boom |
| Competitor comparison | Competitor about 3 LU louder before matching; similar tone after matching | Loudness mismatch | No tonal change; delivery loudness checked against the team's specification |
Three conclusions came out of the check. The "hollow voice" was not a voice problem at all: the widened pad contained out-of-phase content in the same range as the voice, and in mono it partly cancelled the voice's body. The "no energy" complaint was the missing-fundamental problem: the bass lived around 45 Hz with almost no harmonics, so the phone reproduced nothing. And "quieter than the competitor" was a straight loudness difference, which disappeared when the two were matched, so the editor made no tonal change for it and simply confirmed the delivery level met the agreed specification.
Total time for the investigation and fixes was a little over an hour. Had the editor responded to the original complaints directly, the likely moves would have been a broad bass boost, a vocal EQ boost and more limiting, which would have made the mix boomy in the car, harsher on earbuds and more fatiguing everywhere, while leaving the phase problem untouched.
Preventing Problems When Mixing for Multiple Playback Systems
Every symptom above is cheaper to prevent than to fix. Prevention is mostly a matter of routine: building the checks into your session template and schedule so they happen automatically instead of when someone complains.
- Set and mark a moderate reference listening level, and return to it after every louder or quieter check.
- Keep a mono fold switch or plugin in every session and use it after each significant change.
- Load one or two loudness-matched references into the session template before mixing starts.
- Check on monitors, headphones and a small single speaker at least once per mixing session, not only at the end.
- Add delivery-specific systems: a car for radio, a handset for phone audio, a phone at low volume for social video.
- Run a mono check immediately after adding any stereo widener, wide reverb or mid/side processing.
- Check multi-mic sources for alignment and polarity before mixing.
- Take a break every hour or so, and never approve a mix at the end of a long session.
- Measure integrated loudness and true peak against the delivery specification before export.
- Write notes per system during checks, then change one thing at a time.
If your work regularly ends up in specialized environments, such as meeting rooms, assistive listening loops or clinical phone lines, add those to your checks as well. Each has constraints of its own, and a mix that is fine on a phone speaker can still fail on a hearing loop or a conference system.
When to Bring In a Mixing Engineer
Most translation problems are solvable in-house with the method above. There are three situations where outside help earns its cost quickly.
- The mix translates badly and the cause is unclear. If you have run the checks, matched levels, looked at mono and still cannot explain why the mix falls apart on a given system, a fresh set of experienced ears on a trusted system will usually find it faster than more trial and error.
- There is no treated room available. Without an accurate monitoring environment, low-end and balance decisions are partly guesswork. Headphones and car checks help, but they do not replace a room you can trust for final decisions.
- A delivery must meet a specification you cannot verify. Broadcast loudness, platform delivery requirements, stem formats and downmix behavior need proper metering and experience with how each specification is checked. Failing a delivery check costs time and sometimes a slot.
When you brief an engineer, send the current mix, the stems if you have them, the reference tracks you are aiming for, the delivery specification and, most useful of all, notes on which systems the problem appears on. That context turns an open-ended request into a targeted job. If you would rather hand the whole finish over, our audio editing and production service covers mixing, translation checks and delivery to specification.
Verdict Mixing for multiple playback systems is less about owning many speakers than about disciplined listening: match loudness, keep a moderate level, fold to mono routinely, check on a small single speaker as well as monitors and headphones, and approve with rested ears. Diagnose each complaint by the system it appears on before changing the mix, and most translation problems become quick, targeted fixes instead of a spiral of processing.
Where this comes from
- Audio Engineering Society — Monitoring practice
- European Broadcasting Union — Listening conditions
- iZotope — Mixing and reference guides
The figures and practices above come from the sources listed.
Working on something like this?
We take on Audio Editing & Production work for teams who want it done once, properly. Tell us what you are building and we will tell you honestly whether we are the right studio for it. Start a project.
Where to go next
Spotted something wrong? Report an error on this page. We correct on the page and say what changed.
