Speed sets the playback rate between 50 and 120 percent. Pulling it down to 85 percent is the typical slowed setting; below about 70 percent most tracks start to smear. Because the rate change is applied to playback rather than resampled, slowing the track also lowers its pitch — that coupling is the sound this edit is built on, not a side effect to work around.
Slowed and Reverb Maker for Any Song
Turn any song into a slowed and reverb edit. Speed, reverb mix and tail length are adjustable, the render happens in your browser, and nothing is uploaded.
Pick a track, set how far to slow it and how much room to put it in, preview the result, then download. Everything is rendered on your own device, so there is no upload and no credit cost.
MP3, WAV, M4A or OGG · up to 10 minutes · stays on your device
01
What the Three Controls Do
Reverb mix is the balance between the dry signal and the reverberated one, from nothing to fully wet. Around 25 to 35 percent is where most slowed and reverb edits sit: enough to put the track in a room, not enough to lose the transients.
Tail length sets how long the reverb takes to decay, from a fifth of a second to six seconds. Short tails read as a small room, long ones as a hall or a cathedral. The render adds the tail length to the output duration so the final decay is not cut off mid-fade.
02
Why Slowing Also Lowers the Pitch
There are two ways to make audio play slower. Time-stretching keeps the pitch and rebuilds the timing with a phase vocoder, which is what you want when fitting a loop to a project tempo. Changing the playback rate slows the waveform itself, which stretches every wavelength and therefore drops the pitch — the tape-machine behaviour.
This tool deliberately uses the second method. The slowed and reverb sound came out of people playing records and files back slower than they were cut, and the lowered pitch is inseparable from the result. A pitch-preserving stretch at 85 percent sounds like a slow performance; a rate change at 85 percent sounds like the edit people mean.
The practical consequence is that a vocal moves down roughly three semitones at 85 percent, and a little over five at 70. If you were planning to sing or play along, work out the new key first — the key and tempo finder on this site reads the original from the same file.
03
How the Reverb Is Generated
The reverb is convolution-based, which means the dry signal is convolved with an impulse response: a recording of how a space answers a single sharp sound. Convolution is what makes a reverb sound like a real place rather than a chain of delays.
Rather than shipping impulse response files, the tool synthesises one on the fly from exponentially decaying noise, with the decay curve raised to the power of 2.5. That exponent matters more than it sounds: a linear decay reads as artificial, while a steeper curve runs out too quickly to sound like a room. It lands close to a medium hall without needing a megabyte of audio to be downloaded first.
The trade-off against a recorded impulse response is character. A real hall has early reflections at specific times and a frequency response that changes as the tail decays; synthesised noise is smooth and placeless. For this style that neutrality is usually an advantage, because the point is space rather than a particular room.
04
Previewing and Exporting
Preview renders the whole file and plays it back, so what you hear is exactly what the export will contain — there is no separate real-time path that might sound different. On a long track the render takes a few seconds before playback starts.
Each preview renders from scratch, so moving a control and previewing again always reflects the current settings. Playback stops when you change file or leave the page, and the audio context is closed rather than left running in the background.
The download is a 16-bit WAV at the source sample rate. As with the other tools here, exporting WAV avoids a second round of lossy encoding; convert once at the end if you need a smaller file for sharing.
05
What Works and What Does Not
Material with space already in it takes the edit best: ballads, ambient, anything with a sustained pad or a sung line. Dense, loud masters have less room for a reverb tail to sit in, and pushing the mix past about 40 percent on them mostly buries the rhythm.
Two settings tend to disappoint. Very low speeds below 60 percent expose the limits of a plain rate change — cymbals turn into washes and consonants smear. And very long tails on an up-tempo track overlap successive beats, which reads as mud rather than atmosphere.
Files are capped at ten minutes because the decoded audio and the render both sit in memory at once. A long file on a low-memory phone may fail where a laptop handles it without noticing.
06
Slowed and Reverb Questions
Is my file uploaded?
No. Decoding, rendering and preview all happen in your browser, and the file is read directly from your device. Nothing is sent to a server, which is why there is no upload bar and no credit cost.
Why does the song sound lower in pitch?
Because the playback rate is changed rather than time-stretched, which is what produces this particular sound. At 85 percent a vocal drops about three semitones.
Can I slow it down without changing the pitch?
Not here. That needs a phase vocoder and produces a different effect — a slow performance rather than a slowed and reverb edit. Use a DAW's time-stretch if that is what you want.
What settings should I start with?
Speed around 85 percent, reverb mix around 30 percent, tail about two seconds. That is close to the convention for the genre, and all three are easy to push from there.
Why is the exported file longer than the original?
Two reasons: slowing the playback stretches the duration, and the render appends the reverb tail so the final decay is not clipped.
Why is the download a WAV?
To avoid a second generation of lossy compression. If you need a small file, convert the WAV once after exporting rather than encoding twice.
Does it work offline?
Yes, once the page has loaded. Neither the render nor the preview needs a connection.