Common Audio Mistakes with Voice Tracks
by Dan Geocaris
In working with producers and editors on their video sweetening and mixing projects, I often get audio that I haven’t had the pleasure of recording. As a result, over the years I have seen... or heard it all.
I have had voice tracks come in so heavily compressed that they are on the verge of distortion and their waveforms looked like a Marine haircut. Or how about voice tracks from a video shoot layered with a nice grounding hum because the shooter never checked their audio before and during recording…”Hey, can you fix this?”
While these examples are gross mistakes, there are some simple things you can do as a producer to help insure that you have the cleanest audio tracks to work with when it comes time to mix.
Using the tips below will minimize some of the most common mistakes I see with voice tracks, and they are all easily preventable. The goal with these tips is to get you the best audio possible to work with.
1- Insist on Voice Track Files that are Unprocessed.
Unprocessed means no compression, no exciter, and no reverb. These are all things that can and should be added later in the mixing process if needed. And yes insist on unprocessed tracks before you begin your session, because a lot of times these are added automatically as a preset when they are setting up to record.
Everyone wants to sound good and most voiceover studios, and voice talent with home studios, will have favorite processor settings they run their tracks through to make them sound “pro”. The problem is that what they add is their opinion of what sounds good, and what they add, cannot be taken away. They are also making these carved in stone decisions without hearing what the voice tracks are going to be mixed with. What kind of music? What kind of nat sound?
It is best to let whoever is doing your final mix make those tweaks as they are doing the mix. That way your audio guru can take into account the tone and everything around the voice tracks when adding processing to give you a great sounding mix.
2- Don’t Mix Lav and Camera Mics Together.
If you work with video you know that when on-camera narration and interviews are shot they are usually shot using a lavalier and a camera or auxiliary microphone. When I get just a single track or 2 tracks that are exactly the same from an editor, I usually question if it is correct. Most of the time it is a mistake, they had both mic tracks centered before they exported them resulting in a mono sum of the two mics. Other times the editor will say, “Oh, I just mixed the two mics together for you.” In general practice you don’t want those two mics mixed together.
First of all if you zoom in on the waveforms you will see that they don’t line up. For every foot away from the talent the camera mic is, it adds just under 1 millisecond of delay. Though this delay only starts to become perceptible at about 30 milliseconds, mixing the mics together can create all sorts of uncontrolled phase colorations.
Secondly, the two mics will have totally different sounds. The lavalier will tend to sound darker with a more direct sound and less room influence. The camera or auxiliary mic will usually be brighter with a lot more of the room’s reflections and background noise.
In the mixing environment I pick just one of the tracks to work with depending on how they sound and which one I feel I can do the most with. Gut reaction is to take the brightest sounding track. But sometimes it is a lot easier to brighten and EQ the mud out of a lav, and add a little room convolution reverb, than it is to deal with everything the camera mic picks up. So give your audio person the choice.
3- Grab a Little Room Ambiance.
This one is fairly simple. When editing video a lot of times on-camera interviews and narrations are opened up to finish b-roll shots or for dramatic pauses. Unless your source material was recorded in a perfect studio, when it is opened up there are dropouts in the background ambiance. These dropouts are distracting, and you will notice background noise more if it comes and goes. Also, don’t think that just because there is going to be a music background that it will hide the problem.
The easy fix is to be sure to record a little room tone at the end of your interviews, or harvest a little from between takes, and give it to your audio person with the rest of your tracks. That way in the mixing process the dropouts can be filled in with the same room tone and background making it a lot less distracting to the viewer.
I have seen ordinary videos totally brought to life and saved by a good sound track. I have also seen really good videos ruined by a bad sound track. The voice track is such an important part of any video project that you don’t want to treat it as an afterthought. You want to keep control, and depending on the usage and feel of your program the audio may need to be treated differently, so leave your options open.
I hope these tips help you go into your mixing process with tracks that are easy to work with, saving you time and money, and you end up with a great mix.