Case Studies in Voice-Over Recording (Part 1)
April 15th, 2009
By Rudy Helm, Quality Assurance Tech, Visual Purple, LLC
I recall once upon a time a colleague described to me that on his PC, he was hearing within the recordings of a project’s Voice-Over talent (supporting an animated character) where she persistently pops her p’s – and my colleague began to wonder if he was crazy. He recognized that if he turned down his subwoofer, he didn’t hear it. So he asked me if that was something that could easily be fixed in their processing? He added that he didn’t hear that kind of problem in any other audio.
I answered to that, “EZ fix?”
Well… If she was the only popper, a couple things may have happened:
-She wasn’t recorded with the same circumstances as other VO talent; or
-She got too close to the microphone sometimes. A pop filter should have been used in all cases (this is a physical device which costs about 25 dollars….hangs on the microphone stand). But in software, Sonic Foundry (now Sony) has a nice preset which minimizes plosives (pops) or sibilance (“ess” sounds). Sometimes using both the physical device and the software filter in tandem can be helpful.
It is also a possibility that the software Equalization settings (EQ) in my colleague’s PC sound system were incorrectly configured for the situation. Many of today’s software drivers include a wide palette of listening-environment presets. Available parameters generally can be further tweaked by the end user. Consider that if a preset is configured to give an extra bass boost, one could notice artifacts in audio within the low-end frequency bands where plosives reside. But I didn’t believe this was the case at hand because my colleague didn’t notice artifacts in any but the recordings of this one specific VO talent.
Anyway, when you’re stuck in a situation where VO can’t be re-recorded, EQ’ing after the fact may be the last resort you have to take to arrive at a solution. In the case of the plosive p’s popping, running the digitized VO clips through a software editor with batch editing EQ capabilities might work out satisfactorily and may even save time.
When ‘fuzziness’ quality in VO can happen:
-when lips are too close to the microphone. This is when the proximity effect is most noticeable. (The proximity effect is that heavy, low frequency characteristic that you get when a microphone records a sound emitter very close to the microphones diaphragm. Record a voice too closely and you may have to use EQ later to thin out the heaviness). Or,
-when a windscreen was utilized, where a pop-filter should have been used instead.
Windscreens also tend to filter away some of the high end. And there is yet another factor that may need to be considered… The choice of microphone.
The proactive solution for any particular recording room is to rent several microphones of various brands and/or diaphragm sizes (condensers, dynamics, and lapels) and test-record VO talent, male and female (same test-script on each voice, each mic). Then choose to use (buy or rent) the microphone(s) that seems work best with available talent.
Even with cases like this, there are shades of gray. Nothing, it seems, is simply black or white. My colleague told me that his team’s recording department reported to him that the mic-to-talent distance was consistent. Well, the caveat about this might be, was the distance the same for all voice talent? If any one specific voice talent was involved, her consistency (regarding distance) would be the only aspect that matter…but managing a larger pool of voice talent is another thing. To be continued later in a Part 2 blog post….
Leave a Reply