Compensating for lousy miking

[Adam Gold]

I know this can't really be done, but I'd like to do what I can to improve our latest audio debacle for the shows we do (pro-bono) for our kids' High School. Those of you who may have read some of my posts over the years know our theatre people just won't let us mic the shows properly so we have to make do with what we have. I'd use shotguns and/or wireless lavs if I could but they won't let us. In this latest case we had four boundary mics on the front lip of the stage, with orchestra behind and performers in front, fed through the theatre's mixer into our Zoom H4n. We set the levels during warm-ups but as we used the loudest levels (full chorus and orchestra really belting it out), the majority of the show, like 95% of it, comes in at about -40dB. The loudest parts are at -6 dB (set by the Auto Levels function on the Zoom during warm-ups but then locked into manual for the actual recording).

It sounds just as awful as you would suspect. The orchestra sounds pretty good but the vocals are thin and boomy at the same time, lots of nasty reverb/echo and only marginally better than the on-cam mics (used really only for sync purposes, but we mix in a little from the back of the house for presence and applause).

I can attach a file but just imagine your worst nightmare and that's pretty much how it sounds.

For the past two weeks, I've been slicing the audio track from the Zoom to bring up the levels during the soft parts, and also playing with the following in Premiere:

--EQ to boost the low-mids (for warmth) and the high-mids (for sparkle), and also using the low-cut (to help minimize rumble) and high-cut (to help kill the hiss a little)

--Compression and limiting to help reduce the dynamic range, bringing the softer sounds up and trying to avoid overloading during the loud parts

--Using a gate to help maybe reduce a little of the reverb.

But the more I do the worse it sounds. Everyone either begins to sound like they are underwater or Auto-tuned robots. I'm probably doing everything backwards or am on the right track but am doing too much.

I'm trying to do all this using the tools in Premiere, because at a certain point it just can't be worth all this effort for so little results, and considering the powers that be don't care nearly as much about this as I do, I'd love to get the sound just this much better.

Any creative thoughts out there? Specifically, what tools and techniques do you typically use to clean up audio like this? Are EQ and Compression the right tools or are there better ones?

[Vincent Oliver]

For plays I tend to have one mike up close to the stage, this picks up the voices and blows out the chorus. Another mike halfway up the hall, this records the chorus, but loses individual voices. My camera mike picks up all the ambient noise, claps and general hall sounds.

I bring in the video together with audio recorded by the camera and use this as my guide track for the entire production. Next I take the best vocals from mike 1 and the chorus from mike two and place each of them on a new audio track. Now I cut between the three (or four) sound tracks to pick out the best. I keep the camera audio track running but at a lower volume, it produces depth to the sound.

I don't mess about too much with auto dynamics, instead I use gain to bring up or down sections so I get a good sounding track.

It works :-)

[Adam Gold]

I guess I should have mentioned that this is exactly what I'm already doing, mixing the primary audio track from the stage mics (4 mics mixed down while recording to two channels on 1 track) with the on-cam tracks (six more, blended in as needed, always only using the best sound). For example, we'd blend in the cam mics when the performers move into the audience and the stage mics pick up, literally, nothing. I'm just trying to get that primary track to sound better.

Thanks for the tips; it's good to know I'm not the only one doing the mix this way.

[Warren Kawamoto]

Have you considered using a shotgun for the front instead of boundry mics? We used a shotgun with good results. The best results, however, has been when each performer wears their own wireless mic. These are the kids I shot. The first choir was 1 shotgun only, the rest of the shows were 16 wireless mics on performers.
YouTube - HEARTS..Every Child Deserves an ARTS Education!

Here is one scene with all wireless.
http://www.youtube.com/watch?v=AbH0SMMGjRE

[Vincent Oliver]

A nice sound track, and well recorded. I suspect that 16 radio mikes may be beyond most peoples budgets, especially for school productions. Professional theatre companies may not always allow you to attach mikes to performers, especially for a period production.

Nevertheless, it does work for you and you have produced a great sound track. (Nice footage too, what was it shot on and what settings do you use?)

[Warren Kawamoto]

Those 16 Sennheizer wireless mics belong to the theater, they are available for use by the cast of every production. All cast and orchestra mics were fed to a programmable mixer, and I took a feed from it. The shotgun for the choir is a Rode NTG3. On camera mic NTG3 picked up the audience. Close up camera was Sony EX-1, wide shots was Sony HXR-MC1. White balance was 3200K, preset. I occupied 3 seats in the center of the theater for my two cameras, and shot everything solo, while seated in the audience.

[Vincent Oliver]

Your expertise has certainly paid off, excellent theatre shots. I presume you used a IR filter on your EX

[Adam Gold]

Really nice videos and they sound great. It'd be great if we could do any of that, but we can't. We were stuck with the boundary mics for this production. I'd never use them if I had a choice. The alternative was no mics at all.

My task now is to make a silk purse out of this sow's ear.

This is my fault because I wasn't specific enough about my questions in my first post. These are great suggestions on miking tools and techniques but they are of more use for future shows. Right now I'd love some advice about what software tools, methods and techniques, if any, can help bring some life back to our crappy sound, if anyone's had some experience doing this because they were forced to record under less than optimal circumstances. I know there are a lot of post-production audio mixing experts out there and I'd really appreciate some advice.

[Steve House]

I have to ask, why do they want you to shoot the show yet are so adamant against you doing the sound properly? Don't they underdstand that a decent video that people will want to watch requires both good pictures and good sound? "Theatre people" doing professional stage productions go to extraordinary lengths to make sure the sound is up to par. It's particularly puzzling why they wouldn't let you use unobtrusive shotguns where appropriate.

You can't make a silk purse out of a sow's ear and the notion one can make a bad recording sound good by manipulating it in post is one of the great fallacies of the industry.

[Stan Harkleroad]

Why boundary mics also? Why not a couple sets of decent omnidirectional mics? Split them on the sides of the stage with one pointing in and one pointing out on each side. That would work better than boundaries I'm certain. Still not perfect but I'd guarantee it would be better.

[Jay Massengill]

I believe a portion of your trouble may be due to mixing 4 mics down to 2 channels, especially if in reality it's 4 mics getting mixed down to one signal that's simply recorded on both tracks of the Zoom.
In either case, unless a skilled operator is manually mixing the 4 mics live, it's going to be muddled sound if all 4 mics are up full all the time.

I'd suggest using one low-noise small diaphragm cardioid mic for the stage (powered by a high-quality preamp or mixer), recorded on one track of the Zoom. And one low-noise cardioid for the chorus, recorded on the second track of the Zoom.

If you use multiple mics to cover the width of the stage, you have to mix them live in order to get the benefit of multiple mics without introducing terrible phase problems from the mics that are further away from the action at that moment in time. Or if you can't mix them live you have to record each mic to a separate track and mix them properly in post.

[John Saunders]

Originally Posted by Stan Harkleroad

Why boundary mics also? Why not a couple sets of decent omnidirectional mics? Split them on the sides of the stage with one pointing in and one pointing out on each side. That would work better than boundaries I'm certain. Still not perfect but I'd guarantee it would be better.

Don't quite get how "pointing" a omni would work....

[Colin McDonald]

Sorry I have nothing really to suggest here Adam but I do wish someone would actually try to answer your question (which I thought was perfectly clear BTW) instead of telling you how you should have done it.

I have similar situations where I was not able to record audio properly. In one case I ended up with an on-screen apology and subtitles which is what the BBC do when the audio is crap for whatever reason.

[Adam Gold]

Originally Posted by Steve House

I have to ask, why do they want you to shoot the show yet are so adamant against you doing the sound properly? Don't they understand that a decent video that people will want to watch requires both good pictures and good sound? "Theatre people" doing professional stage productions go to extraordinary lengths to make sure the sound is up to par. It's particularly puzzling why they wouldn't let you use unobtrusive shotguns where appropriate.

You can't make a silk purse out of a sow's ear and the notion one can make a bad recording sound good by manipulating it in post is one of the great fallacies of the industry.

Steve, that's a great and valid question and I ask myself the same thing with every show. The shows themselves are extraordinary and they used to just hire a guy with a VX2000 to sit in the back, push the button and take a nap. This was completely unacceptable to me so I came in with my gear and have really been able to produce stuff that looks pretty great. But they are adamant that "we are not doing a TV show," only an "archival record" of the show and our mandate is to be invisible. My weakness here is I care about this more than they do. They're certainly happy with our DVDs and are verbally appreciative, but if I died tomorrow they'd likely go back to the old way. In any negotiation, he who cares least, wins.

And after reading your posts and others from wise sound experts over the years, I do know that you can't make bad sound good in post. My goal was just to make it a bit better so it wouldn't suck quite as much.

Stan and Jay, we normally use a mixture of mics on stands and the Zoom at the front of the stage but for this show, a period piece, they wanted nothing visible that would make the theatre look modern. We actually had to move two of our cameras that normally would have been up front to the back and sides, so no audience member during the show could see any evidence we were there. We couldn't even have the Zoom in the pit pointing towards the stage because of the little red LED that's on while it is recording. Apparently the lights on the music stands were not sufficiently contemporary looking to destroy the illusion, but mics on stands were.

Believe me, from the beginning, four years ago, I've been begging for them to let me use wireless lavs on the kids, at least for the Upper School (rejected), then for more and better mics and a proper mixer (not the house mix, which is unreliable at best and is never the same as a recording mix) (rejected) (And did I mention it's kids running everything? They don't have the skills, and they don't care). Hanging mics low the way you might record a symphony orchestra? (rejected.) Our normal system of mics on stands plus the Zoom works pretty well for Choir, Orchestra and Band concerts, but not so much for Plays and Musicals. It's actually quite a beautiful fairly new theatre designed by some big shot sound professional and sounds quite good live, but they don't understand that recording audio isn't the same as listening to it live. Or maybe they do but they just don't care. Leaving all four mics open all the time was bad, but the alternative was worse, because the kid running the board couldn't get it straight which mic was on which bus and EVERY LINE had the first four seconds cut off when he was riding levels. So I begged him just to leave all the mics open all the time for our final taping.

So I still do it because I can't stand the thought of someone else doing an even worse job than I do.

Overnight I did some playing around with Soundbooth with the audio tracks. Sounds experts will howl at this, but there's a setting that basically says "make everything the same volume" and despite my misgivings I applied it and on balance, it helped. Sure, because of those awful boundary mics, every time someone moved a set piece it sounded like a freight train, and every dance number is dominated by the thundering herd of footsteps, but it took care of at least making the softer vocals more audible during otherwise quiet times. Then I applied a preset called Vocal: Increase Clarity, which applied both compression and EQ. It sounded pretty decent, so I rendered those out and added them to the mix.

The problem was it killed the peaks, as you might expect, leading to a bizarre warbly cutting out of the sound if, say, a single piano note was louder than the vocal. So I recalled something that was recommended for a different purpose, which was to lay this heavily compressed track next to the original, which still had the peaks. So I did this, applied a little EQ to the original and boosted the gain on the original by 9 db (it was recorded with peaks at -6 dB), deciding to live with the occasional overload.

Does it sound great, or even good? No. Would anyone who knows anything about sound cringe? Sure. But at least now you can hear everyone, even if it sounds more like FM radio than a live performance.

I can live with that.

Thanks for everyone's thoughts so far, and if anyone has additional advice before I burn this sucker this weekend I'd be immensely grateful.

[Steve House]

Originally Posted by Colin McDonald

Sorry I have nothing really to suggest here Adam but I do wish someone would actually try to answer your question (which I thought was perfectly clear BTW) instead of telling you how you should have done it.

I have similar situations where I was not able to record audio properly. In one case I ended up with an on-screen apology and subtitles which is what the BBC do when the audio is crap for whatever reason.

His question was clear but the answer is not because of the simple fact that there might not BE an answer. Given that he has 4 boundary mics at a distance from the sound sources, there's going to be a lot of room "air" in the recorded audio and there are no filters that can remove it. Given that he is mixing 4 mics down to two channels, there are going to be arrival time difference at each mic that, when mixed, will produce phase interactions leading to comb filtering etc which cannot be fixed in post. It may very well be that the raw recording is just about as good as he's going to be able to get.