LCR vs Mid-Side

In audio terminology LCR means Left-Center-Right and can refer to a mixing style where all mono tracks are panned hard left, hard right, or equally to the left and right bus for a center signal. This was actually borne of necessity decades ago when mix consoles had only the ability to send a track to either the left or right output buses or to both. It is also used to denote channels in a sound reproduction system such as 3.1 which has left, center and right channels (3) and a subwoofer (.1).

Mid-Side can refer to a very useful miking technique (we won’t discuss here) and to a signal processing method often used in audio processing during mixing or mastering to help focus a process such as compression or EQ toward the center or out to the left and right channels. It is also sometimes used for analyzing the side content of commercial mixes. While Mid-Side conversion can be useful, it does not produce a clean left, center and right audio stream – in fact it yields only two audio streams, a mono Mid signal and a mono Side signal. Mid-Side conversion should really be termed “Sum Difference” conversion since it is simply a matter of addition and subtraction: M = L+ R and S = L- R where L is the Left channel, R the Right, M the Mid and S the Side channel.

Actually Mid-Side processing doesn’t provide very good separation and can actually create phase issues with strong signals on the sides – for example, if a signal is only on the right channel, then the left channel L = 0, and calculating the Mid = L+R = 0+R = R. Calculating the Side = L-R = 0-R = -R. So the right channel signal ends up in both the Mid signal and Side signal, but with the signal inverted in the Side signal! With a signal only on the left channel you again end up with it in both the Mid signal and Side signal, but not inverted. Is this a problem? Certainly! It can cause anything from mild phase issues to extreme (and almost painful) audio effects.

A stereo track of a Jazz trio processed with Mid-Side channel extraction.

Note in the waveform plot above that you can see both the left and right channel signals showing up in the extracted Mid channel at reduced, but still very audible levels. The Side channel combines both the original left and right channel signals with the right channel inverted (although you can’t see that detail here).

You can experience this inverted effect directly using Mid-Side conversion on a stereo mix by listening to the side signal only. If you use a Mid-Side plug-in that provides inline conversion from L/R stereo to Mid-Side, and then mute the Mid signal, you will experience an annoying sensation. Due to the flipped polarity elements there is still a “phasey” central sound and very odd left and right elements that make it feel like it’s sucking the sound out of your head! There is no way a normal Mid-Side process can yield clean left, right and center signals. Mid-Side techniques used too heavily on stereo buses can destroy phase relationships between instruments, upset frequency balances, impair transient relationships, and degrade the tonality of an instrument itself.

A New Way to LCR

CenterOne provides a new approach that can actually split a stereo signal into three audio channels, left, center, and right. Using it to mute the center channel presents a stereo image of only the left and right sides that sounds quite natural, with few low level artifacts, and with the center sounds missing! Sounds on the right are still on the right and sounds on the left are on the left. Muting the left and right channels with the center on yields a very clean center channel with left and right sounds almost entirely removed. Compared to Mid-Side processing, CenterOne yields much cleaner results as well as separate left, right and center signals.

Jazz trio stereo track processed with CenterOne – left guitar, center bass and right drums are clearly separated in this example.

Below is a closer look at the Mid-Side mid signal and the CenterOne center signal compared to the original single bass track from the mix. You can clearly see that the Mid-Side process includes significant left and right channel sounds.

Comparison of original centered bass guitar waveform with CenterOne and a Mid-Side process.

For analyzing commercial mixes CenterOne is far superior to Mid-Side processing and enables you to hear nuances in the left and right channels that are totally lost using Mid-Side techniques. For example, some mixes place the reverb for a centered vocal mostly on one side, and may place the reverb for bass on the other side! You can’t hear such details using Mid-Side processing.

Using Mid-Side processing to “expand” a stereo image, to make it wider, can quickly lead to unnatural phasing issues as mentioned earlier – CenterOne yields more natural sound widening of a stereo image as well as being able to provide many other sound stage adjustments.

There are several controls in CenterOne to help “shape” the separation. One fader adjusts the overall prominence of the center signal relative to the stereo image which can be used, for instance, to bring a vocal forward in a mix or to push it back. There are separate L(eft), C(enter) and R(ight) faders to adjust these three elements independently which enables modifying the sound stage in many ways. There is a center bandpass EQ control that permits tuning the frequency used for the center channel extraction algorithm – frequencies in the selected range are sent to the center channel while frequencies outside this range stay in the left and right channels. This enables effectively reducing, even removing, a vocal from the center while still playing centered kick drum and bass guitar.

Furthermore, CenterOne has a Center Channel Width control that adds further adjustment capability. This slider controls the width of the extracted center channel, from extracting “phantom center” sounds located only exactly in the middle of a stereo stream to encompassing audio that is slightly left and right of the center stereo image.

CenterOne also has five output channels: the processed stereo signal, a left output, a center output and a right output. The processed stereo outputs are useful for quick enhancement of the stereo image where more focus is required on the center sounds or where more space is desired by enhancing the left and right material. The separate left, center and right outputs enable unique processing capabilities in a DAW where you can route the separate outputs to separate buses for processing, and then bring the processed signals back into the final mix. This enables applying different compression levels or different EQ to the center and sides without the possibility of phase cancellation which may occur with Mid-Side techniques.

If you have used Mid-Side processing in the past you will find CenterOne provides new and extremely valuable capabilities for audio manipulation.

Written by Dennis J Wilkins.