“Good prose is like a windowpane.”
To compress vocals, put a compressor on the vocal track and start at a 6:1 ratio. Lower the threshold until every word is clearly audible, set the attack to about 30 ms so each word keeps its punch, keep the release at 100 ms or less, then turn the gain up until the vocal is as loud as before.
Knowing how to compress vocals starts with a moment every producer knows. You record a great take, but in the mix the quiet words vanish under the beat and the loud ones jump out and sting. Turning the fader up only fixes half of it, because that uneven, now-you-hear-it, now-you-don’t sound is a job for a compressor.
These steps work in any DAW and with any compressor, because they all share the same main knobs. Below, each step is shown in FL Studio’s Fruity Limiter with screenshots from a video by Max, the founder of Mix Elite, along with the mistakes that make a compressed vocal sound flat, mushy or lifeless.
Table of Contents
What does compressing vocals do?
Vocal compression turns down the loudest parts of a vocal, so the loud and quiet words sit closer together and every word stays easy to hear over the beat. Two settings decide how much: the threshold, the volume where the compressor starts to work, and the ratio, how hard it pushes down anything louder than that.
Here’s the ratio in numbers: at 6:1, a word that goes 6 dB over the threshold comes out only 1 dB over it.
Vocals need this more than almost any other sound. A singer or rapper never stays at one volume, so some words jump out while others get lost under the beat. Compression is a big part of what makes a vocal sound great, and this method works for any genre, from pop and EDM to rap, drill, drum and bass and even voiceovers.
In Mix Elite’s free 4.5-hour FL Studio course, Max explains why it matters so much: “Our ears prefer consistency. Sudden changes in volume are distracting and mentally exhausting.” A well-compressed vocal lets the listener relax and follow every word.
How to compress vocals in FL Studio, step by step
The video uses Fruity Limiter, the limiter and compressor that comes with FL Studio. You can use any compressor you like, though, since they all have the same main controls. Here’s the order:
- Normalize the vocal so it reaches the compressor at a healthy level.
- Load Fruity Limiter on the vocal’s mixer track and switch it to COMP.
- Set the ratio to 6:1.
- Lower the threshold until every word is clear.
- Set the attack to about 30 ms.
- Bring the release down to 100 ms or less.
- Turn the gain up to match the old volume, then listen with the beat.
The video below shows every step on a fast drill vocal, with before-and-after comparisons you can hear.
Step 1: Normalize the vocal first
Before you touch a compressor, normalize the vocal. Normalizing turns the whole clip up as far as it can go without clipping, which makes the right threshold much easier to find. A quiet vocal makes you push every knob further than it should go. A healthy level gives you plenty of room.
Open the vocal clip’s channel settings and click Normalize (1) under Precomputed effects. If the volume jumps a lot from line to line, even it out with volume automation too. A compressor only works on what goes over the threshold, so a line that’s much quieter than the rest slips under it and never gets touched.
Step 2: Load Fruity Limiter and switch it to COMP
Press F9 to open the Mixer, select the vocal’s mixer track and load Fruity Limiter into an empty effect slot. It opens as a limiter, so click COMP (1) to turn it into a compressor. These are the knobs you’ll use:
- RATIO (2): how hard the compressor pushes down.
- THRES (3): the threshold, where compression starts.
- ATT (4): the attack, or how fast it grabs each word.
- REL (5): the release, or how fast it lets go again.
- GAIN (6): makeup gain, which turns the vocal back up at the end.
Leave SAT, KNEE and SUSTAIN where they are. This method doesn’t touch them. If you want to know what every knob does, our FL Studio compressor guide explains Fruity Limiter and Fruity Compressor one control at a time.
Step 3: Set the ratio to 6:1
Start with the ratio. For modern vocals like pop, EDM, rap or hip-hop, 2:1 or 3:1 is a little too gentle. Max says: “I usually just find myself doing 6 to 1. It’s a perfect starting point for all kinds of modern vocals.”
Don’t push it much further. Above 10:1, the compressor flattens the natural ups and downs of the voice until the waveform looks like a sausage. Those ups and downs carry the emotion, like a rapper getting louder on a line because they mean it. A very high ratio also drags up breaths and clutter from the quiet parts.
Treat 6:1 as a start, not a law. In Mix Elite’s videos, the free FL Studio course uses about 4:1, and the stock-plugin vocal mix goes anywhere from 5:1 to 20:1, with the high end saved for a rapper who hits a lot of loud accents.
Step 4: Lower the threshold until every word is clear
Play the vocal and pull THRES (1) down. Keep going until all the words are clearly audible. A handy rule from Mix Elite’s video on vocal mixing mistakes is to set the threshold as low as the quietest word in your lyrics, so the loud words get pulled down toward it.
Fruity Limiter’s display shows you what’s happening. The purple tips (2) are the part of each word the compressor is taking off.
So how far down should it go? There’s no single number, because every vocal is different. On the drill vocal in the video, the threshold ended up at about -20 dB once the attack was set. Trust your ears. Every word should be easy to hear, and no word should jump out.
Step 5: Set the attack to about 30 ms
Most people get this step wrong. With the ratio and threshold set, the vocal in the video still sounded mushy. The fix was the attack, which is how many milliseconds the compressor takes to grab a sound. The attack starts at 0, which means instant, so the compressor chops the front off every word and the words smear together.
Attack times sort like this: up to 8 ms is fast, 8 to about 25 ms is medium, and only past 25 ms is it slow. Some people call 1 ms a medium attack, but anything up to 8 ms is still a fast attack.
For a quick rap or drill vocal where every word should cut through, let roughly the first third of each word pass untouched. On the vocal in the video that’s about 27 to 30 ms, the red part (1) in the picture, and the compressor works on the rest of the word (2). So set ATT to 30 ms. Because those first peaks now get through, lower the threshold a little more, as the video does.
Want a softer vocal that sits back in the mix instead? Then a 0 ms attack is fine. Slower, sung vocals don’t always need this trick either.
Try this test: play one line with an instant attack, then again at 30 ms. At 0 ms the words blur into one long smear. At 30 ms the start of each word comes through, and the vocal cuts through the beat.
Step 6: Bring the release down to 100 ms or less
The release is how fast the compressor lets go after a word drops back under the threshold. In the video’s project it sat at about 240 ms, which is too slow for a fast vocal. Max’s rule: “I would usually go no longer than 100 milliseconds.” For a quick vocal like this one, you could easily go down to about 50 ms. Shorter means punchier.
Turn REL (1) down to 100 ms or less. ATT (2) stays at the 30 ms from the last step. Next to the default settings, the difference is night and day.
Step 7: Match the volume and listen with the beat
Compression makes the vocal quieter overall, so turn GAIN up until it’s about as loud as it was before. Then switch the compressor off and on while the beat plays. If both versions are equally loud, you hear what the compression changed instead of just a jump in volume.
Listen closely, because compression is harder to hear than EQ. If you boost 1 kHz with an EQ, you at least notice that something changed, while compression can slip by unless you compare it with the raw vocal. Judge it with the full beat playing, too. The video adds reverb after the compressor and checks the vocal in the mix, not on its own.
What are good starting settings for vocal compression?
Start at a 6:1 ratio, lower the threshold until every word is clear (about -20 dB on a normalized vocal), set the attack to about 30 ms and the release to 100 ms or less, then raise the gain until the vocal is as loud as before. That makes a lead vocal clear and punchy. Then adjust by ear.
Here are the same starting points in one place.
| Setting | Starting point | What to listen for |
|---|---|---|
| Ratio | 6:1 | 2:1 to 3:1 is too gentle for modern vocals, and past 10:1 the vocal goes flat |
| Threshold | Lower until every word is clear (about -20 dB on the video’s normalized vocal) | No word jumping out and no word lost |
| Attack | About 30 ms for fast rap, 0 ms for a softer vocal | Words that cut through instead of smearing |
| Release | 100 ms or less, about 50 ms for a quick vocal | Punch between the words |
| Gain | Up until it matches the volume with the compressor off | The same loudness on and off |
Different vocals need different settings. A fast rap or drill vocal gets the 30 ms attack and a short release. A slow, sung vocal may not need the slow attack at all. And a rapper with lots of loud accents can take a much higher ratio, which is why the stock-plugin vocal mix sometimes goes all the way up to 20:1.
How do you set attack and release without squashing the vocal?
Start the attack around 30 ms, so the first part of each word gets through before the compressor grabs it; at 0 ms the words smear together. Set the release to 100 ms or less, about 50 ms on a fast vocal, so the compressor lets go between words, and keep the ratio under 10:1 so the vocal doesn’t go flat.
If it still sounds squashed, raise the threshold until fewer words get pressed down. For a softer vocal that should sit back in the mix, a fast attack is fine. Your ears decide.
Why do my vocals get buried in the mix?
Usually because the vocal’s volume jumps around. The quiet words drop under the beat while the loud ones stick out, so no fader setting works. Compression evens the words out so you can turn the whole vocal up, and an attack around 25 to 40 ms lets the start of each word cut through.
Mix Elite’s free FL Studio course puts it simply: the vocal is usually the most important part of a song, so it belongs in front instead of fighting the beat for attention. Three more fixes help it get there:
- Duck the reverb and delay. Sidechain them to the dry vocal, so they drop while the singer sings and bloom in the gaps. Do it on every vocal; here’s the setup in FL Studio.
- Make room in the beat. Sidechain the instruments to the vocal, so they dip a little while it plays. A multiband sidechain dips only the frequencies the vocal sings in, so the beat keeps its low end. Our guide on how to sidechain in FL Studio shows the routing.
- Level the lines first. If one line is much quieter than the rest, lift it with volume automation before the compressor, so it doesn’t slip under the threshold.
Why not just use a limiter on vocals?
A limiter is a compressor with a very high ratio, so nothing gets past its ceiling. The video tries one on the same vocal, and the verdict is simple: the vocal becomes lifeless. The quiet and loud moments end up at the same level, and the feeling in the performance goes with them.
That’s also why the ratio matters so much. Anything past 10:1 starts to act like a limiter, and you get the sausage-shaped waveform from step 3. For the difference between all three tools, read our guide to compression vs clipping vs limiting.
How do you know if vocals are over-compressed?
Over-compression is easy to miss, because a compressed vocal sounds louder, and louder often sounds better at first. Watch for these signs:
- The vocal sounds flat and lifeless, with no loud or soft moments left.
- The waveform looks like a sausage, the same height from start to end.
- Breaths, mouth noise and room sound get louder between the words.
- The words smear together and sound mushy, which usually means the attack is too fast.
- Harsh S sounds stick out more than before.
Most fixes are simple. Raise the threshold or lower the ratio until the life comes back, and slow the attack down if the words smear. Harsh S sounds after compression are normal, and that’s why a de-esser comes next in the chain.
One warning: once compression clicks, you’ll want it on everything, because it makes sounds louder, glued and juicy. Louder isn’t always better. Compress the vocal until every word is clear, then stop.
Where does compression go in the vocal chain?
Compression comes after the clean-up work and before the effects. Mix Elite’s vocal formula, V.O.C.A.L., gives the order: volume automation, optimize with EQ, compression, attenuate harshness, and layer with effects. Attenuate just means turn down, which is the de-esser’s job, and the effects are things like reverb and delay.
Our guide on how to mix vocals walks through every step of that chain. And our stock-plugin vocal chain in FL Studio builds it with nothing but FL Studio’s own plugins.
Need a lot of compression? Many engineers split it across two compressors in a row, so each one only does part of the work. That tends to sound more natural than one compressor working very hard. The stock-plugin chain does something similar: after Fruity Limiter it adds OTT, a free multiband compressor, mixed in at about 20% for a more modern sound.
Another common trick is parallel compression. You send a copy of the vocal to a second mixer track, compress that copy hard and blend it in quietly under the original. The vocal gets thicker, while the original keeps its natural ups and downs.
How do you add vocals to an instrumental?
Put both in one project. Drag the instrumental into your DAW, then record the vocal on its own track or drag in one you already have, and line it up with the beat. Then mix the vocal on its own mixer track in this order: volume, EQ, compression, de-essing and effects.
That’s the homework in Mix Elite’s free 4.5-hour FL Studio course: drag a practice acapella and an instrumental into FL Studio and mix the vocal with the chain above. Recording your own voice? Our vocal recording tips help you get a clean take first.
DAY 5 OF THE 7-DAY MUSIC PRODUCER COURSE
Compression is one step. Day 5 covers the whole vocal.
Day 5 of the 7-Day course teaches recording and mixing vocals, from setting up your room and choosing recording equipment for any budget to the V.O.C.A.L. mixing formula. You also get 10 copy-and-paste vocal processing chains and practice acapellas, so you can try them even without a singer. It’s taught in FL Studio (the free trial works), about 45 minutes a day, and on day 7 a pro gives you feedback on your finished track. It’s $27, and you get 60 days to decide if it was worth it.
Frequently asked questions
What are good starting settings for vocal compression?
Start at a 6:1 ratio, lower the threshold until every word is clearly audible, set the attack to about 30 ms and the release to 100 ms or less, then raise the gain to match the old volume. Slower, sung vocals may not need the slow attack. Adjust everything by ear with the beat playing.
Why do my vocals get buried in the mix?
Usually the vocal’s level jumps around, so quiet words drop under the beat, or the reverb and the beat fill the space the voice needs. Compress the vocal so every word is steady, use an attack around 30 ms so words cut through, and sidechain the reverb, delay or beat so they dip while the vocal plays.
How do you set attack and release on a vocal compressor?
Start the attack around 30 ms so the first part of each word gets through; at 0 ms the words smear together. Set the release to 100 ms or less, about 50 ms for a fast vocal, so the compressor lets go between words. If the vocal still sounds squashed, raise the threshold or lower the ratio.
What does vocal compression do?
It turns down the loudest parts of a vocal so loud and quiet words sit closer together. Every word stays easy to hear over the beat, and the vocal is easier to listen to, because sudden jumps in volume are distracting and tiring for the ear.
How do you add vocals to an instrumental?
Put both in one project: drag the instrumental into your DAW, then record the vocal on its own track or drag in one you already have, and line it up with the beat. Then mix the vocal in this order: volume, EQ, compression, de-essing, and effects like reverb and delay.
