Control

Reverb Control

Reverb control sets the wet mix of a convolution reverb from 0 to 100%. See how dry and wet mix, why mix level buries vocals, and values that keep lyrics clear.

Open the free editor: upload MP3, WAV, AAC, OGG or FLAC, adjust speed, reverb, bass, tone, width and space, then export MP3 or a ZIP of many tracks. Everything runs in your browser.

Reverb is the slider that most often ruins a slowed edit, and it ruins it in the same way every time: too much of it. The control runs from 0 to 100% and sets how much of the reverberated signal you hear against the untouched signal. It does not set how long the tail lasts; that is the separate Space slider. Keeping those two ideas apart is the single most useful thing to understand when you add reverb to a song with the reverb generator in this editor, and it is what this page is about.

What Reverb does in signal terms

After Speed, Bass and Tone, the signal is split into two paths. The dry path carries the sound unchanged. The wet path runs it through a convolution reverb, which multiplies the signal against an impulse response, a recording of how a space responds to a single click. The editor generates that impulse response in the browser from shaped noise, with a 10 ms pre-delay, a set of early reflections in the first 60 ms, and a tail whose brightness falls from a 9 kHz cutoff at the start to 1.2 kHz at the end. The two channels get slightly different early reflections so the tail is genuinely stereo.

The two paths are then mixed back together with a linear crossfade. Dry gain is 1 minus the mix; wet gain is the mix. So:

That last row matters. At 100% the direct signal is gone. You hear nothing but reflections, which is sometimes exactly right for an ambient intro and never right for a song with lyrics.

Why mix level, not decay, muddies vocals

Picture a singer in a large hall. If you stand close, you hear the voice clearly and the hall behind it. If you walk to the back, the voice is distant and the hall dominates. The hall did not change. Your distance from the source did. Mix level is distance.

Decay time is the size of the hall. A long decay in a quiet hall still lets you hear the singer, as long as you are standing close. In editor terms, a 5 s Space at 30% Reverb stays intelligible. A 1.2 s Space at 70% Reverb does not, even though the tail is short, because 70% of what reaches you is reflection and only 30% is the voice.

Consonants are the first casualty. The t, k, s and p sounds that let you parse words are short and quiet, and a reverb tail arriving from the previous syllable masks them. The vowels survive, so the song sounds like a song but the words become shapes. If you have ever wondered why an edit sounds fine for ten seconds and exhausting for three minutes, it is this. The guide on why slowed and reverb sounds muddy walks through the fix in order.

How Reverb interacts with the other controls

Space. The two are partners. A high mix needs a short decay to stay readable, which is why Haze Chamber pairs 54% with 1.2 s. A long decay needs a moderate mix, which is why Wave Drift pairs 2.8 s with 50% and why the cathedral-style variations use 40% or less.

Speed. Slowing a song opens up the gaps between notes. The same mix sounds less crowded at 0.80x than at 1.00x because the tail has more room before the next hit. Sped up edits have the opposite problem and rarely want more than 20%.

Bass and Tone. Both shelves sit before the reverb split, so they shape what goes into the reverb as well as the dry path. A +12 dB bass boost feeds a boomy low end into the convolver, which produces a low rumble in the tail. If a heavy preset sounds thick in the low mids, lowering Bass often cleans the reverb more than lowering the mix does. Likewise a dark Tone darkens the tail, which usually sounds good; a bright Tone makes the tail hissy.

Width. The wet signal is stereo with different left and right reflections, so increasing Width mostly widens the reverb. At high mix levels, Width has a bigger audible effect.

Recommended Reverb by content type

If a value from this table sounds wrong on your track, it is almost always because the source already has reverb baked in. Modern pop vocals often do. Subtract 10% and listen again.

A method for setting Reverb

1. Set Speed first and leave it.
2. Set Space to a sensible length for the material. 2.4 s is a fair middle.
3. Start Reverb at 0% and raise it in 10% steps while the chorus plays.
4. Stop at the first step where the vocal takes a step back. Lower the mix by 5%. That is usually your number.
5. Switch to the Dry preset for four bars, then back. If the words are harder to catch than you expected, drop another 5%.

Set it at the chorus, not the intro. Intros are sparse and will tolerate far more reverb than the busiest part of the song.

Common mistakes

• Reaching for Reverb to make it "bigger". Bigger is Space. Reverb makes it wetter. If you want a hall, lengthen the decay and keep the mix moderate.
• Setting it on headphones only. Phone speakers have no low end and compress the mid-range, so a tail that sounded balanced on headphones can swallow a vocal on a phone. Check both.
• Setting it on the intro. See above. The chorus is the test.
• Stacking reverb on a wet source. Many masters already carry a vocal reverb. Your mix adds to it, so start lower than the table suggests.
• Using 100% as a creative choice on a vocal track. It removes the direct sound entirely. Fine for a transition or an ambient bed, not for a song.
• Fixing mud with Tone. Darkening hides hiss, not masking. If words are lost, only lowering the mix brings them back.

Reverb and the export

The exported MP3 uses exactly the chain you previewed, at 96 kbps and 32 kHz, encoded in a web worker in your browser. Reverb tails are smooth, noise-like signals that MP3 encoders handle well, so heavy reverb does not cost much at this bitrate. The export is fine for YouTube, TikTok and Shorts because those platforms re-encode uploads. For a master intended for streaming distribution, process a WAV source and finish in a DAW, since the editor does not export WAV.