How to make a bass boosted version of a song

A bass-boosted song is just the original with a low shelf lifting everything under about 120 Hz by +6 to +12 dB, plus enough level pulled back that the extra energy doesn't run past full scale. In a browser it takes about two minutes: drop the volume to around 70 %, push the shelf up in the low-shelf bass booster, leave the soft limiter on, export a WAV. The order matters more than the numbers. Do those steps backwards and you get the sound people call "blown speaker", which is really just clipping smeared across the whole mix.

What "bass boosted" actually means

There are two different things people mean by the phrase, and they need different settings.

The first is the tasteful version: a track that sounds fuller in a car or on cheap earbuds, where small drivers give up below about 80 Hz. Maybe +4 to +6 dB of shelf and nothing else. Nobody should be able to tell you did anything, except that the kick has weight now.

The second is the deliberate internet version, where the 808 rattles and the whole mix ducks around it. That's +10 dB or more of low end, limiting on top, and a certain amount of audible saturation that is part of the sound. It isn't an accident. But there's a difference between saturation you chose and clipping you didn't notice, and the difference is entirely in the order of operations.

Either way the filter is a low shelf, not a knob that boosts one note. The shelf here has its corner at 120 Hz, so it lifts a broad region below that and tapers to nothing above it, leaving vocals and snare alone. The five presets are exact: Off is 0 dB, Light is +3, Medium is +6, High is +9, Max is +12.

The tasteful setting, for the car

1. In the volume booster, set the level to about 70 %. That's roughly −3 dB (20·log10(0.7) = −3.1), which buys back most of what the shelf is about to add.
2. Leave the soft limiter on. It's a tanh knee: transparent below 0.7 of full scale, bending asymptotically toward full scale above it.
3. In the bass booster, set the shelf to Medium, which is +6 dB with the corner at 120 Hz.
4. Hit the level-matched A/B button a few times. It swaps between processed and untouched, and it trims the processed side down to the original's peak first, so you're hearing the tone change instead of just "the new one is louder". Loudness fools everyone in an A/B. This one takes it off the table.
5. Export as 16-bit WAV.

The volume booster and the bass booster are separate pages, so the way to combine them is to export a WAV from one and drop that file into the other. WAV is lossless, so a hand-off costs you nothing; use 32-bit float for anything in the middle and 16-bit only for the file you actually keep.

If the result sounds boomy rather than solid, the problem is usually 150–250 Hz mud, not missing sub. Skip the shelf and use the five-band equalizer instead: +6 dB at 60 Hz and −2 dB at 230 Hz. Cutting mud reads as more bass than adding bass does, most of the time.

The extreme, over-the-top version

For the loud rattling version, people reach for +10 to +12 dB of low end. Here's a sequence that survives it.

The clip readout is live and it names the actual percentage of the file that would run past full scale at your current setting. That number is the one to watch. A handful of isolated samples you won't hear. Past a percent or two on a bass-heavy track it stops sounding like grit and starts sounding like a fault.

Why boosting first makes it crunch

Most bass-boost attempts fail for one reason. The song was already mastered close to full scale, with maybe 0.5 dB of headroom. You add +12 dB of shelf on top. Every low-frequency peak now wants to sit 12 dB past the ceiling and there's nowhere to go, so samples get flattened at ±1.0. Flat tops mean sharp corners, and sharp corners mean harmonics thrown all the way up the spectrum. That's the fizzy, harsh sound on the hi-hats and vocals in a bad boosted track. The damage isn't in the bass at all; it's smeared over everything above it.

Turning the level down first fixes it, because the shelf then has somewhere to push. The limiter catches whatever still pokes through, and because it bends rather than chops, it produces mostly low-order harmonics that read as warmth instead of breakage. You bring the level back up at the end, once the shape of the waveform has settled.

Make room, add bass, catch the peaks, then get loud. In that order.

Deliberate saturation vs accidental clipping

If you want the distorted aesthetic, get it on purpose. Leave the limiter on, push the volume past what Smart Boost suggested, and let the tanh knee round off the peaks. It'll sound thick and slightly squashed. That's a limiter working hard, which is a musical sound.

What you don't want is digital clipping, which is what you get with the limiter off and the volume higher than the file's headroom allows. Two tells: it gets brighter and nastier as it gets louder rather than fatter, and it gets worse on cymbals than on bass. If that's what you're hearing, back off 10 % at a time until the clip warning reads 0 %, then decide how far you want to push back into it.

Worth knowing: the limiter is a limiter, not a compressor. There's no attack or release to set, so it won't pump. It just refuses to let the waveform leave the box.

Exporting without wrecking it

Export is WAV only — 16-bit, 24-bit or 32-bit float, rendered offline faster than real time. No MP3 export, on purpose.

Here's why that matters for bass boosting specifically. If you start from a 128 or 192 kbps MP3 and encode the boosted result straight back to MP3, you've stacked a second lossy generation on the first. MP3 encoders throw away what's masked in the signal they're given. You just moved the balance by 12 dB, so the artifacts the first encoder hid under the old midrange are now sitting under a much quieter midrange where you can hear them. Heavy low end eats bitrate too, so the second pass at the same kbps has less left over for everything else.

WAV skips all of that. A 4-minute stereo 16-bit 44.1 kHz file is about 42 MB, which is fine for a USB stick in the car or an AirDrop. If you need an MP3 at the end, encode the WAV once, at 256 kbps or higher, in whatever converter you like. One generation, not two.

What happens when you upload it

If the boosted track is going to YouTube, Spotify or Apple Podcasts, know that those platforms normalise playback loudness — roughly −14 LUFS for Spotify and YouTube, −16 LUFS for Apple Podcasts. Crushing your bass-boosted file as loud as it will go doesn't make it play louder there. It gets turned back down, and all you're left with is the limiting.

Smart Boost measures integrated loudness with a real ITU-R BS.1770-4 implementation: K-weighting, 400 ms blocks at 75 % overlap, absolute gate at −70 LUFS and relative gate at −10 LU. It reports the number and tells you the gain that would land you on a target. It does not quietly re-master the file for you. You read the figure and move the slider, or click the suggestion. Adding 12 dB of low shelf usually raises integrated LUFS by a couple of dB less than you'd expect, because K-weighting deliberately discounts low frequencies the way ears do.

Doing it on your phone

The browser tools run in mobile Safari and Chrome, though a 10-minute track will make an older phone think for a few seconds while it decodes. Nothing is uploaded; the decode and the processing happen in the tab, and the page keeps working with Wi-Fi off. If you want your boosted versions saved with their settings so you can come back and adjust them, the Increase Volume app for iPhone and the Android version keep a library of your files with per-track EQ and bass settings, and work offline. Same five bands, same preset values. The app tops out at 200 % volume and has no limiter and no level-matched A/B, so for extreme boosts the website is the better tool right now.

One thing neither can do: raise your phone's actual output past its maximum. No app can, on any platform. What you're changing is the file, which is why a bass-boosted WAV played at the same system volume sounds bigger than the original did.

If the track you're boosting is the audio in a video, the video volume tool will play an MP4, MOV or WebM back with the boost applied and export the boosted audio as a WAV. It does not re-encode the video file. You'd need to mux the new audio track back in with something else.