And Down the Stretch They Come! | A Sound Education from Dynamix

And Down the Stretch They Come!

Pasted Graphic 6

And Down the Stretch They Come!

A jam-packed field. Each running neck-and-neck trying to lead the pack. The whole mass moving toward the finish line with breathtaking speed. Finally it ends, but it's too close to call!!!

No, it's not the Kentucky Derby. It's an overcrowded commercial with too much information that's been crammed into 30 seconds. Sound familiar? We all want to get our message across, but we often say too much. I'm guilty of it (as many of my friends would attest).

What's my point? Your audio engineer at Dynamix Productions is not one-dimensional. We can also be a part of your creative team; your copy writing staff; your proof reader; your client liaison. In other words, we're not just button pushers. During your recording session, we can offer scripting advice so that your message is not cluttered. If one sentence sounds better rewritten as two sentences, we can suggest how. If your brand is not standing out enough, we can point out ways for more emphasis. We are experts at understanding that scripting for narration is much different than for visual mediums like print and web. We want your message to be clear and concise, without getting bumped, ran into the rail, blocked out, or flat out beaten. Oh, and horse racing lingo? We do a little bit of that, too.

Dynamix Tech Notes

So what do you do when you do need to cram too much information in a short amount of time? "Time Squeeze" to the rescue! In the old analog days, engineers used to speed up a tape or record to make it run faster. However, the audio was pitched up and started sounding like Mickey Mouse. Digital time-squeezing solves the pitch problem by mathematically throwing out data. Smoother sounds can be attained by matching the right algorithm with the content. Although voice-over is less complex than time compressing music, it can perhaps be the trickiest. Our ears (and brains) are designed to be experts at listening to the human voice. Any deviation from "normal" can be detected almost instantly. Therefore, we like to quote the "ten-percent" rule when digitally reducing time. Once a voice is time-squeezed more than around ten-percent, it starts to sound fake. Therefore, a recording that is 33 seconds long can be time-squeezed by 3.3 seconds to about 29.7 seconds without sounding too processed.

However, when you time squeeze this much, you will lose the spacing and leave sentences butted-up against each other. We like to build in spacing with breaths deleted and tightened before the final "squeeze." This at least retains a little "air" and personality.

Just like all rules, it doesn't always hold true. Hard and short sounds like "k," "t, and "p" are shortened even more and sometimes disappear all together. If we are planning a dramatic time-squeeze, as in a legal disclaimer for instance, the narrator will read slowly with exaggerated emphasis on consonants so they won't get too truncated.

What about adding time? "Time-stretching" does the reverse and adds in information. Here we use the "six--percent" rule. Anything longer makes the narrator sound drunk and makes for some good laughs. But it can be a little eerie when you push it too far and it doesn't make much of a difference. That guy must have drank his lunch!

Neil Kesterson
UA-25904086-1 GSN-354889-D