Live sound for both theatre and music events always seem to have an echoey aftertaste to them, and the aural image presented appears a bit out-of-focus. It is difficult to close your eyes and really know where the sound is coming from. Aside from a room having bad acoustics (that’s a whole other issue), one of the key aspects of sound is the arrival time. Sound does not travel instantaneously from one place to another, its speed varies by the transport medium. For sound travelling in air, the speed is about 1,126 feet per second (343 m/s), or about 1 foot per millisecond (mS). This varies slightly with temperature and humidity, but for the purposes of this discussion, we’ll assume it to be constant. Sound travels through other, denser materials like water and steel, much more quickly, and again, we are not going to concern ourselves with this today.

It’s about time

Once sound is converted into an electrical signal by a microphone its travel time can vary significantly. In the most pure situation, like a microphone connected to an amplifier that is connected to a loudspeaker, the delay is almost negligible – what comes out of the speaker happens almost instantaneously.

As the audio signal transmission path becomes more complicated, then the signal may be delayed by analog-to-digital (A/D) conversion, digital signal processing calculations, and again by the conversion from digital-to-analog (D/A). In a simple digital audio mixer this may be a few microseconds to as long as several milliseconds. If you think about time in terms of distance, then a couple of mS add-up quickly into a couple of feet. Us mere mortals don’t typically notice these things, unless you happen to be standing onstage in front of a wedge monitor, then it can become painfully obvious – the sound from the speaker arrives later than the sound from your instrument. As these arrival time differences coincide with the wavelength of a note / tone it can result in noticeable phase cancellations (dips in the perceived frequency response).

Dang those Vidiots

If you are sending the audio in synchronization with a video signal, then it may be delayed by 2-to-5 video frames (each video frame is about 1/30 of a second, or 33mS) as it passes through each piece of signal processing gear. This is why it is so important to keep the picture time-aligned to the sound, otherwise you get ‘Lip-Sync’ issues and people’s mouths don’t seem to be moving with the audio track.

Wait a sec . . .

On stage, this can affect the clarity and punch with which your sound is reproduced. If you are working with an analog audio mixing console, there is little you can do to affect this problem, unless you maybe have an outboard patchable signal delay. Most digital audio mixing consoles offer a signal delay adjustment for each individual input channel.

In the example shown above, we have a kick drum that is about 20 feet upstage of the sound reinforcement speakers. The sound picked-up by the microphone is (more-or-less) instantly reproduced by the loudspeaker, so the sound from the speaker(s) arrives at the listener’s ears BEFORE the sound directly from the kick drum. This subtle delay between the two signals is what gives a slightly ‘hollow’ sound to the mix and the subsequent signal cancellations due to the sound waves not being perfectly combined is what takes some of the ‘punch’ out of the sound.

In the second example, the signal from the drum is delayed so that the acoustic wave from the drum head arrives at the listener’s ears in unison with the reinforced sound from the loudspeaker system. KerWhump! In lieu of a dull ‘thud’.

Yes, but my actors aren’t kick drums

Agreed, but the basic premise still holds true. If you adjust the delay on the performers mics so that they are more closely aligned with the sound emanating from the loudspeakers, it will take some of the hollowness out of the system and as a result, the audience may not have to work quite as hard to have their brain correlate the disparate signal arrivals.

Actors move around, where do I set the delay? If you are working with a preset console that allows you to record cues, this can be done by setting the delay to approximately match the actor’s target blocking. If they miss their mark, it typically won’t be but by a few feet, and you are still ahead of the game. Most movement onstage traverses the acting area laterally, so they are working in a fairly narrow window up-stage/downstage relative to the speaker location(s). If the actors are ‘all over the place’, then you may not be able to keep up with their movement through manual adjustments. There are ways to actually track the performer’s movements in 3D real-time and cause their position to adjust the audio system delay. You can read about systems that do that at Cast Software’s BlackTrax web site and Outboard UK’s TiMax website.

Whoa there, Hoss.

The brain is a mysterious thing. The mind plays tricks on you. One of those tricks is that if you have two audio signals arrive at the ear within about 20 mS of each-other, the brain will lock onto the one that arrived first, not the one that was loudest. This is known as the Haas Effect and is named after the researcher that first documented it. For voices you can use this to your advantage. If you add about 10-20 mS more delay than you calculate that you need to align the performer to the speakers, then the actor’s voice is the one that the audience will focus upon in lieu of the sound arriving from the loudspeaker. This is why being exact with the signal delay setting and the actor’s blocking is not as critical as you might think.

Have fun in your adventures in time travel. Just don’t be late.