An interesting method which I've looked into as a way to spread a single recorded surround channel accross two surround playback channels could also be used to spread mono material across the front stereo soundstage. I thought a lot about this a few years ago but haven't done anything with it yet. It is a way of using a constant phase shift (not a frequency variable phase shift like that introduced by a short delay) to decorelate two copies of the same signal. The method is attractive in that it introduces no delay, reverb or eq, sums back to mono perfecty with only reduced level depending on the amount of phase shift between channels, and is indetectable if one channel is listened to in isolation. This is the primary method all matrix surround decoders use (Dolby PLII, DTS neo6, Logic-7, Circle surround, etc) to synthesize multi-channel surround information from the mono difference signal of stereo source materials, spreading surround information to multiple surround speakers.
The technique is to mult the mono signal, then phase shift one copy forward by some amount and the other backwards the same amount (say +45º and -45º) for a total shift of 90º. Again, this is a constant phase shift across all frequencies. Note that the process as implemented by the Miles gear quoted below only shifts frequencies above 100Hz, probably as a way of minimizing potential bass cancellation. For a single channel mono recording you could simply shift one copy of the signal something like 90º, as there is no standard phase 'reference signal' like there is with multichannel applicaitons.
How to do it? The easiest way is probably a plugin. Note that this type of thing is
not the same at all as most Phase Shifter type guitar effcts.
This Voxengo plugin would do, there are probably others available form other sources:
PHA-979 is a professional audio AU and VST plugin which allows you to apply an arbitrary phase shift to sound material. What is meant by the phase shift here is simultaneous shifting of all frequencies across the active frequency range of the signal by the given value in degrees. This is achieved by linear-phase design.I asked the developer (Aleksey Vaneev) about such a use on the Voxengo forum back in '08.
Here is the thread and his suggestions. Here is
another related question on the same plug-in. I'm posting here, now, because I recenty came across a website of a now defunct manufacturer of matrix gear that has some information about using the technique:
http://www.milestech.com/Application Note #4
Studio and Production Applications of the MTI-3 TriSonic Imager
Synthesizing Stereo With SpreadSound
The SpreadSound function is a very effective way to synthesize a stereo signal from a monophonic program or signal source. Unlike other stereo synthesis techniques, it will not alter the frequency response of the signal nor create any audible delay or other type of distortion. It only adds a small amount of phase shift to the signal, which in itself is practically inaudible.
SpreadSound will create a big, spacious, yet totally clean version of a mono input signal. Either of the described connection approaches can provide excellent stereo synthesis. In addition to working perfectly with all stereo or TriSonic playback systems, the SpreadSound channels can be mixed back to mono with perfect results.Multisonic Info
The Multisonic® Imaging System:
Introduction and Technical Description
SpreadSound
..a high-fidelity method of intentionally and accurately localizing a sound source across a space, rather than at a point. SpreadSound is used at the surround outputs of the M86 to prevent the surround signal, which once separated out is a single monophonic signal, from localizing as a phantom center between adjacent surround loudspeakers. Phantom localization is desirable across the front soundstage with stereo sources but is usually considered a detriment at the sides and rear of a surround system.
SpreadSound works by shifting the phase of the audio signal a constant relative amount. A pair of SpreadSound outputs is designed to have a relative phase shift of 90 degrees at all frequencies above 100 Hz. By keeping the output signals in quadrature phase, a spreading effect is realized with no phantom localization. The effect is unnecessary below 100 Hz; at these lowest frequencies localization by humans does not occur. In this frequency range the signal is left in phase to maximize the low-frequency power output and the loudspeaker-array efficiency.
The SpreadSound circuit can be used with any monophonic program material to create a big, spread-out sound. With Multisonic Imaging, a mono source will focus tightly to the center. If a large soundstage is desired instead, such as with a mono music source, SpreadSound is a great way to achieve it without any of the detrimental side effects typical of stereo synthesis techniques.