Audio dilemma: no noise vs no quality

[Steve Siegel]

I have been told that my audio isn't "crisp" and I'm trying to figure out how to fix it. I shoot birds, outdoors, in all weather. There is always wind. I have a deadcat made expressly for the mic of the Canon XL series. Even with a fuzzy cat, it's impossible to eliminate all the ambient noise. I have a pretty good sound cleaner (Goldwave) that lets you isolate the noise in a part of the clip with no other sound, copy it, and remove it globally. But wind is not relegated to less than 150 hz, and wanted frequencies are affected too. Crispness goes out the window. Then there is that warble that's introduced.

So my question is this. Buy an expensive mic (like a Senheiser M66) and record such good sound that it can't be seriously degraded, or find a software that removes the wind but is gentle to everything else. Is there such a thing?

Any thoughts on this will be appreciated.

[Jon Fairhurst]

Get a better mic. The Goldwave noise reduction is pretty good. I also have Sound Forge, and it's NR isn't really any better.

May I recommend the ME-67 over the ME-66. It's more directional, and better for your application. I wouldn't go as far as a parabolic mic though. They're great for picking a voice out of a crowd when we want the information, but they don't sound very natural or pleasing.

You'll probably still want to roll off the low end, and the NR software will still help if there's a repetitive noise in the background, such as a motor.

There are other solutions as well. The Audio Technica AT815b is less expensive than the ME-67, in case budget is a concern. It's a bit "thin" for dialog, but it has a narrow pattern and is reasonably clean.

http://www.sennheiser.com/india/icm_eng.nsf/resources/ME_67_GB.pdf/$File/ME_67_GB.pdf

http://www.audio-technica.com/cms/re...5b_english.pdf

Looking at the patterns, the Sennheiser mic is really superior to the AT. The lobes are smoother, and at a much lower level.

[Colin McDonald]

Originally Posted by Steve Siegel

So my question is this. Buy an expensive mic (like a Senheiser M66) and record such good sound that it can't be seriously degraded, or find a software that removes the wind but is gentle to everything else. Is there such a thing?

Any thoughts on this will be appreciated.

My opinion:
Whatever mic you use outside it will need proper wind protection - Rycote full kit or whatever make, but not just a fluffy thing which won't help you enough in anything but the lightest breeze. Kill the wind noise as much as you can on the way in, tweaks in post are never as good as proper wind protection. Since this is what you do, go for it Rode NTG-2 or the new NTG-3 or Sennheiser M66 or whatever further upmarket - the actual mic used matters less than the wind protection in this scenario. You need a Zeppelin type kit in the work you do.

[Jon Fairhurst]

Colin is right. Removing the wind noise on the mic is critical.

You mentioned "ambient" noise. A longer mic won't remove the sound of wind buffeting the mic, but it will help attenuate the wind in the trees, traffic, streams and such. At least it will attenuate the high frequencies of the unwanted signals. I would think that with birdsong, you could roll off the LFs without too much degradation.

If you're doing documentary work, leaving in some of the wind noise is actually attractive to me - at least in somewhat hostile environments. It can give the feeling of "being there." I'd rather hear the occasional rumble than continuously warbly audio.

You might also do well with a compressor with a filtered side-chain. The idea is that you reduce the volume of the signal whenever the low frequencies are loud. You can assume that the low frequencies are due to a burst of wind, rather than the target sound. You'll still hear the buffeting, but you can play the birdsong louder between bursts. A multi-band compressor would let you reduce only the LFs when the LF signal is loud. FWIW, SoundForge 9 includes a four band multi-band compressor from Izotope.

Of course, you can also duck the levels manually during wind bursts, but that doesn't work well with tight deadlines - or a social life.

[Steve Siegel]

Jon and Colin,
Thanks for the input. I was hoping for something affordable (ie software), but it looks like the wind is going to go right through my pockets. What we do for love.

[Ty Ford]

Originally Posted by Jon Fairhurst

Get a better mic. The Goldwave noise reduction is pretty good. I also have Sound Forge, and it's NR isn't really any better.

May I recommend the ME-67 over the ME-66.

With all due respect. You want good sound, you get good mics, good preamps and a good recorder. The 66 and 67 mics are NOT great mics. They are what schools can afford to teach students. Raise your sights a bit; a lot really. MKH 60, Schoeps CMIT.

Steve,

What are you recording audio on now?

Regards,

Ty Ford

[Jon Fairhurst]

Ty,

Good to see that I was on the right track: the mic comes first. And, sure, the best mics will be best. Unfortunately, they come with a price tag, and it seems that Steve is on a budget.

Sure, the CMIT will sound sweeter than the ME-67 - especially on dialog, where our ears are very discriminating. But will it really be that much better for a birdsong application, where things are more notional?

It looks like the ME-67 has a tighter pattern. The CMIT has a more uniform fall off (which is, of course, is great for a natural sound when the source is a bit off axis.) Also, the ME-67 has slightly lower self-noise, which would be critical for nature shots. The ME-67 can't handle the SPL of the CMIT, but unless the wind is howling, that's not all that important in this application. The CMIT has lower THD, and that relates to the sweetness and accuracy.

I'm not arguing that the ME-67 is the better mic. (It isn't.) And I'm not arguing that the specs tell you how a mic sounds. (They don't, but they do illuminate areas of strength and weakness.) But for a nature application, would the CMIT really sound $1,000 (or whatever) better? Especially, if that blows the budget for a possible pre-amp/mixer and recorder?

For dialog, I'm with you. For the birdsong application, I'm not (yet) convinced.

That said, I spent some time at NAB at the Sennheiser booth testing various shotguns, and as I recall, the MKH-70 blew the ME-67 away when it came to picking up conversations across the hall. The gain was higher on the MKH-70, so that certainly colored my perception. (Humans naturally prefer the loudest sound and the brightest pictures.) Still the low noise, off axis rejection and clear sound were pretty impressive, even from 30 feet or more in a crowded, noisy hall.

[Ty Ford]

Jon,

um, yes, birdsongs or fly farts, it'll sound better. But please don't take my word for it. Inform yourself.

Do yourself a favor. Rent a CMIT and an MKH 70. Put them up against your 66 and 67 with a good mixer like the SD 302 or 442.

Regards,

Ty Ford

[Paul R Johnson]

I'm confused. We're talking about wind noise issues, then suddenly it's all the mic's fault?

I've used 416s and 816s, AT 815s and a cheap unbranded design obviously based on, but not sounding like, the Sennheisers.

The OP says his audio is rejected because it isn't crisp. Wind noise is at the bottom end, crispness is at the top? The hairy sausage, on a zeppelin style frame, in an isolated handle is pretty good at removing bottom end wind noise and rumble. Taking off the shielding inside and using the mic 'naked' does increase crispness, clarity, etc - because the HF suffers going through the windshield.

So if you want to get rid of wind, then you'll also lose a bit at the top end too. However, removing the bottom end in an effort to remove the rumble usually produces crisper audio anyway, so maybe we could have a sample of the problem to ponder over.

On the mic front, all these different microphones have a 'sound' of their own, but any kind of ultra directional mic is a compromise anyway. What I have noticed, from my own teaching experience, is that students easily produce dull sound whenever they go out with a mic and pole. The problem is simply that they don't aim them, just wave them generally in the right direction, and on these mics aim is critical - the off-axis is thin and weedy, but the audio gain ramps up to compensate.

[Seun Osewa]

You need a great microphone and proper handling for great sound. No way around that.

In this case, you need a Zepellin. If that doesn't work, then you'll just need to recreate the audio in the studio. Also, try using a high pass filter - it can make the sound crisper.

[David Beisner]

Originally Posted by Seun Osewa

you'll just need to recreate the audio in the studio.

What's he supposed to do, capture the birds and bring them into the studio to mic 'em? Haha!

No, but everyone's right so far... gotta go with a better mic. Software NR will never get you "all the way there" in terms of removing noise without affecting the whole track with that wobbly sound. Not even the expensive NR plugins for ProTools do it well (though they're a sight better than anything in Soundbooth!)

Also, if you're doing nature, you might try something even a little more outside-the-box in terms of your micing. If you have access to a really good small diaphragm studio condenser (something like the Neumann KM series), try that instead of a shotgun. Make sure you use a good windscreen, and only do it if you can get far away from roads or manmade noises. But you might find that you could get a good natural feel to your sound with something like that.

I know, it's way outside-the-box in terms of audio for video, but I'm one of those guys that likes to try crazy things and I'm rarely disappointed in my results.

[Jon Fairhurst]

Originally Posted by Ty Ford

Do yourself a favor. Rent a CMIT and an MKH 70. Put them up against your 66 and 67 with a good mixer like the SD 302 or 442.

I've got a project in August that I might just do that with.

Then again, we won't be doing nature recording. We might be shooting dialog indoors. If I rent, I'll probably get the Schoeps 641 paired with a 722 recorder. Unfortunately, it's budget first, equipment second, so we'll see...

[Ty Ford]

Jon,

That would be a good test for you.

Regards,

Ty Ford

[A. J. deLange]

Originally Posted by Jon Fairhurst

I wouldn't go as far as a parabolic mic though. They're great for picking a voice out of a crowd when we want the information, but they don't sound very natural or pleasing.

Given that their gain increases 6 dB per octave (10 dB at 500 Hz, 16 dB at 1 kHz, 22 dB at 2 kHz... for a 23" diameter) I would certainly expect them to sound funny but suppose you applied a -6dB/octave equalization curve? Some of the classic recordings of birds were made with the Dan Gibson parabolic mics.

[Jon Fairhurst]

Originally Posted by A. J. deLange

Given that their gain increases 6 dB per octave (10 dB at 500 Hz, 16 dB at 1 kHz, 22 dB at 2 kHz... for a 23" diameter) I would certainly expect them to sound funny but suppose you applied a -6dB/octave equalization curve? Some of the classic recordings of birds were made with the Dan Gibson parabolic mics.

I wonder if they apply the EQ correction on football games. Parabolics are often used for hits and grunts at the line of scrimmage, and sound really good when mixed in to the overall signal. However, when isolated, you can clearly hear the conch shell effect.

But it seems that they have been made to work for birdsong recordings.

They're probably not the thing to use in the wind though. Rather than a deadcat, you'd need a deadlion!