SOUND CRAFT BLOG / 9 MIN READ

How do I make my audio sound professional?

Professional audio is not a plugin you buy, it is a short chain you run in order: a clean source, a tamed room, the voice on top, then loudness on target. Here is the chain, link by link.

5links in the chain
−14 LUFSloudness target
−1 dBTPtrue peak ceiling
0–100craft score

By Thomas, founder of CutScore · Updated June 2026

AUDIO CHECK · voiceover_v3.mp4
Two people leaning over a lit audio mixing board, dialling in levels, the kind of careful sound pass that separates professional audio from a clip recorded and forgotten.
CRAFT SCORE
FIXES ADVISED
listening on the bad speakers
Music over voice · pull music −5 dB00:12
True peak in spec · −1.4 dBTP
Mix a touch quiet · −17 LUFS, lift +3whole file
The 30-second answer To make your audio sound professional, run it as a chain in order, not a single magic plugin. One, record a close clean source: mic about a foot from your mouth, slightly off to the side. Two, tame the room so it is not bare and echoey. Three, in the edit, clean the noise floor and high-pass the rumble. Four, balance the mix so the voice clearly sits on top of any music. Five, normalise the whole thing to around −14 LUFS with a true peak at or below −1 dBTP. Get those five right and cheap gear sounds expensive. If checking them by ear sounds unreliable, that is exactly what CutScore measures in one pass.
WHY YOUR AUDIO BETRAYS YOU

Here is the part nobody wants to hear. People will forgive a soft shot, a slightly green skin tone, even a wobbly handheld pan. They will not forgive bad audio. Sound is the sense your viewers cannot consciously fault but instantly distrust. When the voice is thin, distant, or buried under a stock-music bed, the brain quietly decides "amateur" before a single word has fully landed. I have shipped videos exactly like that and watched the retention graph fall off a cliff in the first ten seconds.

The trap is your monitoring. You edit on nice headphones in a quiet room, late at night, leaning in. Of course it sounds fine. Then someone watches on a phone speaker on a bus, and your carefully placed music is now a wall and your voice is a rumour behind it. The room you recorded in adds its own tax: bare walls bounce sound back into the mic, and that reflected energy is what we hear as "hollow" or "cheap."

None of this requires a studio to fix. Professional-sounding audio is mostly decisions, not money. The decisions just have to happen in a specific order, because each one builds on the last. Get the source clean and the rest is easy. Skip the source and no plugin will save you. So let us go link by link.

THE AUDIO CHAIN

The five links that make audio sound professional.

Run them top to bottom. Each link has a target you can actually hit, and skipping any one of them is something a listener will notice, even if they cannot name it.

Link in the chainTarget to hitWhat it costs you if you skip it
Source distancemic ≈ 1 ft, off-axisA distant mic records more room than voice, and that reads as hollow and cheap.
Room reflectionssoft, not bareBare walls bounce sound back into the mic and add an echoey "stairwell" tone.
Noise floorlow, steadyHiss, hum and fan noise read as amateur before you have said a word.
High-pass filtercut below ~80 HzLow rumble eats headroom and muddies the voice without you noticing.
Voice vs musicvoice clearly on topMusic burying speech is the single most common amateur tell there is.
Loudness≈ −14 LUFSToo quiet and your video feels timid next to everything else in the feed.
True peak≤ −1 dBTPHot peaks crackle and distort once the platform re-encodes your file.
The honest shortcutIf you only have time for one move, fix the voice-versus-music balance. Pull the music down four or five decibels until you hear every word on bad speakers. It does more for "professional" than any plugin you could buy today.
SKIP THE GUESSWORK

Checking loudness, peaks and the voice-music balance on every video by ear is slow and unreliable. CutScore measures all of it in one pass and hands back the exact gain changes.

Join the waitlist
HOW TO HIT EACH LINK

Capture clean, then fix in the edit.

1. Get close, and tame the room first

Distance is the loudest enemy of professional audio, and it costs nothing to fix. Roughly speaking, doubling the gap between you and the mic halves the direct sound and lets the room take over, which is exactly what "hollow" means. Get the capsule about a foot from your mouth, slightly off to the side so plosives do not thump it. Then soften the space: a rug, a sofa, a wardrobe of clothes, anything absorbent kills the reflections that make a bare room sound like a bathroom. A budget mic up close in a soft room beats an expensive one across a tiled kitchen, every single time. This is the cheapest professional upgrade you will ever make.

2. Clean the floor and cut the rumble

Once the take is in, the first edit move is subtraction, not addition. Apply a light noise-reduction pass to knock down steady hiss, hum and fan noise, but go gently: push it too hard and the voice starts to sound underwater and robotic. Next, set a high-pass filter to roll off everything below about 80 Hz, where the rumble of traffic, air conditioning and desk thumps lives. You will not miss those frequencies in a voice, and removing them frees up headroom so the words can sit louder and clearer. If your audio still sounds muffled rather than noisy, the cause is usually the room and the mic distance, and there is a whole separate fix for muffled audio.

A close shot of hands on a mixing console, fingers resting on the channel faders, the moment where the voice gets balanced against the music before anything is exported.
The mix is where amateur audio is won or lost: voice on top, music underneath. Photo: Tima Miroshnichenko / Pexels.

3. Put the voice on top of the music

This is the one that fools people into thinking they need better gear. They do not, they need to pull the music down. The most common amateur tell in all of online video is a stock-music bed sitting at the same level as the voice, or louder. In my own early videos I set the music by how it felt while editing on headphones, and it was always five or six decibels too loud on a phone. The rule that works: lower the music until you can hear every word clearly on the worst speakers you own, then drop it one more decibel for safety. If the music wants to swell, duck it under speech with a simple sidechain or a few manual keyframes. Here is the full breakdown on why music ends up louder than your voice.

4. Set the loudness, then guard the peaks

Now make the whole mix the right size. Normalise the full export toward −14 LUFS, which is where YouTube normalises loudness, so your video does not feel timid next to the next one in the feed. Short-form feeds tend to run a touch hotter in practice, but −14 LUFS is a safe professional anchor across platforms. Then guard the ceiling: keep your true peak at or below −1 dBTP. That headroom matters because platforms re-encode your file, and a peak that touched zero on your timeline can crackle and distort after compression. Loud and clean is the goal, not loud and crunchy. If you are unsure where your file sits, check your audio levels before uploading.

5. The contrarian last step: listen on the worst device

Most "professional audio" advice ends at the mix. The professionals I trust do the opposite of obsessing over their nice monitors: they finish by checking on the worst speaker they can find. A phone held at arm's length, a single laptop speaker, cheap earbuds in a noisy room. That is where most of your audience actually is, and it is the only honest test of whether the voice still cuts through. If your words survive the bad speaker, they will sound great everywhere. If they vanish, your mix is too quiet, your music is too loud, or your voice never had enough top end to begin with. Fix it for the worst case and the best case takes care of itself.

RATHER SEE IT THAN READ IT?

Here is a real CutScore report for an everyday video: loudness, true peak and the voice-music balance, all measured, with timestamps and the exact gain changes to make.

See a sample report
SHORT ON TIME

If you only fix three things.

Most of the jump from "homemade" to "this person knows what they are doing" comes from these three audio moves. Fix them first.

1
2-MIN FIXAUDIO
Pull the music below the voice
Music sitting on top of speech is the number-one amateur tell, and it has nothing to do with your microphone. Lower the music until every word is clear on bad speakers, then drop it one more decibel. This single move sounds more "professional" than any plugin.
How Listen on a phone speaker, not headphones. If the music ever wins, it is too loud. See music vs voice.
2
QUICKCAPTURE
Get the mic about a foot from your mouth
Distance, not price, is what makes audio sound cheap. A close mic in a softened room records voice instead of room reflection, which is the whole difference between "hollow" and "present." Move in close, point it slightly off-axis, and soften the space with anything absorbent.
How Hand-span distance, off to one side. Add a rug or a sofa to kill the echo.
3
EDITLOUDNESS
Normalise to −14 LUFS, peak under −1 dBTP
Quiet audio feels weak next to everything else in the feed, and hot audio crackles after the platform re-encodes it. Set the whole mix to around −14 LUFS with a true peak at or below −1 dBTP and your video sits at a confident, professional level.
How Run a loudness meter over your export, or let CutScore measure it and give you the exact gain change.
THREE WAYS TO GET THERE

By ear, by meter, or in one pass.

OPTION 01

By ear, on the worst speaker

Free, and better than nothing. The catch is the one we opened with: your monitoring flatters you. It works best when you finish on a phone speaker or cheap earbuds, not your nice headphones. Use the chain above so you are testing against targets, not vibes.

OPTION 02

With a loudness meter

Accurate and honest. An EBU R128 loudness meter and a true-peak readout tell you exactly where your mix sits. The cost is time and knowledge: you have to know the targets, open the meter, and read it correctly for every single export. Great if you enjoy this. Most people do not.

OPTION 03

With a coach in one pass

Hand the file (or a link) to CutScore. It measures loudness, true peak, noise and the voice-music balance against the right standard for your genre, then gives you a 0 to 100 score with timestamped evidence and the fixes. No meters to read. See a sample report.

How CutScore checks your audio for you CutScore is an AI video quality coach, and sound is one of the 13 families it measures. It computes loudness with an EBU R128 meter, reads true peak, estimates the noise floor, and flags when music is sitting over the voice, all deterministically rather than by guesswork. You get one score, the evidence behind each number, and a prioritised list of fixes like "lift the mix +3 dB" or "pull the music −5 dB at 00:12," before anyone else hears the video. It judges the craft of the file itself, so it sits next to a growth tool rather than competing with one. More on the method and the standards.
QUESTIONS

Frequently asked.

Treat it as a chain in order. Record a close, clean source with the mic a foot from your mouth in a room that is not bare and echoey. In the edit, clean the noise floor, set a gentle high-pass to cut rumble, balance the voice so it sits clearly above any music, then normalise the whole mix to around −14 LUFS with a true peak at or below −1 dBTP. Each link matters more than any single plugin.
Music sitting on top of the voice. It is the fastest tell there is, and it has nothing to do with your gear. Pull the music down four or five decibels until you can hear every word clearly on bad speakers, and the same recording instantly sounds more professional. Distance from the mic is the close second.
No. A cheap mic six inches from your mouth in a softened room beats an expensive mic across a tiled bathroom every time. Distance and room reflections do more damage than the price of the capsule. Get close, kill the echo, and a budget mic will sound far more professional than people expect.
For most online video, aim the full mix at around −14 LUFS integrated, which is where YouTube normalises, and keep the true peak at or below −1 dBTP so nothing crackles after the platform re-encodes. Short-form feeds run a little hotter in practice, but −14 LUFS with peaks under −1 dBTP is a safe, professional starting point.
EARLY ACCESS

Make every video sound like you meant it.

CutScore measures loudness, true peak, noise and the voice-music balance, then tells you exactly what to fix, with the evidence to back it up. Join the waitlist for early access.

Join the waitlist