Creating a powerful sense of space and presence in your recorded music is a primary function of the mixing process. The spatial balance in the mix is typically discussed in terms of “width” of the stereo image and “depth” of the acoustic listening environment. The former is primarily set by the Left (L) – Center (C) – Right (R) panning of the individual tracks to the Main Output stereo bus. The latter is primarily set by time-based effects, such as Reverb and Delay, that are added to the mix.
I’ve discussed previously many topics associated with creating and manipulating the stereo field image in posts on sound source localization, stereo image in mastering, mid/side processing, and stereo microphone techniques. For this post and the next, I want to touch on the subject of the stereo imaging of individual audio tracks, versus the stereo imaging of the Left and Right channels of the main mix output bus. In this post, I’ll look at “stereoizing” a mono audio track -- in other words, creating a wide, thick stereo sound from a thin, lifeless mono sound.
A commonly used approach for stereoizing a mono track involves utilizing the psychoacoustic phenomenon known as the “precedence effect”, discovered by Dr. Helmut Haas in 1949. Usually called the Haas Effect, it states that when one sound is followed by another with a delay time of approximately 40 milliseconds (ms) or less (below humans’ echo threshold), the two are perceived as a single sound. Neuroscientists believe that sound localization in the horizontal plane relies on two “cues”: the sound amplitude (loudness) difference between the two ears (inter-aural level difference ILD), and the time difference (delay) of sound reaching each ear (inter-aural time difference ITD). Two sounds with a short delay between them are perceived as one sound. The first arriving sound will localize the sound source, whereas the slightly delayed sound gives the human brain a sense of spaciousness, regardless of where this second sound comes from.
This Haas Effect can be used to create the perception of dimension to an otherwise “flat” mono audio track. This process of stereoizing a mono track goes as follows:
1. Duplicate a mono audio track,
2. Pan one track left and the other track right,
3. Add a short delay to one of the tracks.
Delays of 10 ms or less actually tend to enhance directionality, and yield “out-of-phase” type sounds that are hollow, thin, and sometimes dropping out completely. I’ve mentioned these phasing problems before when talking about extending the stereo field beyond the speaker locations. More examples of deleterious phasing effects are provided in the videos below. Delays between 10 ms and 40 ms, on the other hand, can produce the desired spaciousness. Choosing the right amount of delay is the trick. Ultimately, a stereoized track that sounds good in mono playback will ensure a solid translation regardless of the playback device and listening environment. A critical listening to the mix in mono serves as a quality assurance check for the sound. If phasing problems are present, this becomes clear when you hit the mono switch – the stereoized track shrinks and loses presence.
Joe Gilder is a songwriter, musician and sound engineer based in Nashville, TN. On YouTube, he offers a whole series of expert tutorials on music production using the PreSonus Studio One digital audio workstation. In the video below on stereoizing a mono track, he demonstrates what happens when you record a single input source (a guitar, for example) into a stereo track with the phase of one input inverted. In stereo playback of this track, there is a “push-pull” effect of the sound waves coming from the left and right speakers since one of the waveforms is an inverted version of the other. This is uncomfortable for the listener. And when the track is played back in mono, the sound disappears completely ! This happens because there is total cancellation of the two waveforms when summed. Now, mono playback is not as uncommon as you might think -- it effectively happens when listening to your phone speaker or in a large venue when you are at great distance from the left and right speakers. It can happen in your car. So, having this guitar part drop out completely is a big mistake.
OK, so what does Joe Gilder think about the stereoization technique of duplicating tracks with one channel slightly delayed from the other ? Take a look at this next video.
Well, the stereoized track sounds pretty good. But Joe is NOT happy with the mono playback -- there are phasing problems that lead to his hearing muddy bass tones and high-frequency ‘metallic’ sounds. Perhaps employing a greater delay in the 10 – 40 ms range might alleviate these problems ?
The goal of doubling a single source instrument part should be to obtain two separate, distinct waveforms, i.e., decorrelated waveforms. This should minimize the unwanted waveform phasing effects and cancellations. In essence, doubling an instrument part should sound like making two different instrument parts. This can be done by inserting a splitter plug-in in the stereo track, separating the duplicated parts to left and right channels with different signal processing chains. Different waveforms can be achieved in each channel by inserting any or all of these signal processing plug-ins in each channel and using different parameter settings in the plug-ins:
1. Spectral processing – Equalization (EQ)
2. Dynamics processing – Compression
4. Time-based Effects (FX) – Delay
5. Modeling FX – amplifier
An example of this stereoization approach is shown in the following video. Joe Gilder records a single source guitar part to a stereo track, and inserts a channel splitter with two different amplifier plug-in sounds on the separated channels.
OK, Joe is now pleased with both the stereo and mono playbacks !
A Final Word
Since the goal of stereoization is to obtain two different waveforms of the same instrument part to form a stereo image, why not just do the following up front:
1. Use stereo microphone techniques to record to a stereo track in the first place, as shown in the lead photo at the top of this post,
2. Actually record a second performance of that instrument part to a separate track -- this is truly doubling the part with two separate, distinct waveforms ! This approach is demonstrated very nicely in Joe’s second video above (beginning at the 4:15 time stamp), and, in his expert opinion, is absolutely the best way to get a rich, spacious stereo sound in an audio track.
In the next post, I’ll discuss several ways to manipulate the stereo image in an existing stereo audio track.