AUD · Audio tools

Mid/Side Encoder and Decoder

Continue editing Open in workflow
Often followed by + Stereo width + Equalizer + Limiter

Shareable link

Settings are written to the URL as you change them. Nothing differs from the defaults yet.

Sum and difference

Left and right is one way to hold a stereo signal. Mid and side is another, and it is the same information: mid is what both channels share, side is what only one of them has. The conversion is two additions, and it reverses exactly.

Mid carries the centred material, which in most music is the lead vocal, the kick, the snare, and the bass. Side carries the width: room, reverb, wide guitars, and anything panned hard.

Three modes

Process converts to M/S, applies the two gains, and converts back. The file goes in as stereo and comes out as stereo. This is the everyday mode.

Encode stops halfway and writes M and S as the two channels of the output. Opened in any player, the left channel plays the centre of the mix and the right channel plays the difference, which sounds strange and is meant to. It is a handoff format for another editor.

Decode does the second half on a file that already holds M and S, turning it back into normal stereo.

What separate gains buy you

Dropping the side narrows the image; raising it widens it. Dropping the mid pushes centred material back without touching the room around it, which is closer to a rough vocal-down than an EQ move can get.

Small amounts do a lot. Beyond about 6 dB of side boost, the centre starts to sound hollow and the mix loses its mono fold.

Check the mono fold

A wide file summed to mono loses the side signal entirely, because +S and -S cancel. Anything you push into the side is therefore invisible to a phone speaker, a smart speaker, or a club PA in mono.

Measure it rather than guess: the stereo field analyzer reports mid and side levels and phase correlation for a file, which tells you how much survives the fold before you commit.

Frequently Asked Questions

It rewrites a stereo pair as a sum and a difference. Mid is (L+R)/2, the part both channels share. Side is (L-R)/2, the part they disagree on. Decoding is L = M+S and R = M-S, which returns the original.

It writes M and S as the two channels of the output file, so another editor can treat them as separate tracks. Bring the result back here in decode mode to turn it into normal stereo again.

Stereo width gives one percentage that scales the side signal. This gives separate gains for mid and side, plus the matrix conversion itself, which is what you need to hand M and S to something else.

The maths is exact in theory and rounds at the last bit in floating point. Encoding and decoding without touching the gains returns the same audio to well below anything audible. Re-encoding to a lossy format in between is where real loss would come from.

A mono file is all mid and has no side, so only the mid gain applies and no matrix runs.

It does not, but it raises everything around it. A centred vocal lives entirely in the mid, so pushing the side up buries it in relative terms. Drop the mid instead if you want the opposite.

It can show you one. If the side signal is nearly as loud as the mid, the two channels disagree badly, which is what a poor mono fold sounds like. The stereo field analyzer measures that directly.

Explore Our Tools

Browse all tools