Audio tips when interviewing in a noisy space? (conventions etc)

A

analog_addict

Well-known member
I have quite a big job next week shooting talking heads at a big conference location. Ideally we would conduct the interviews in a separate room but I don't think that's an option. They had issues with the previous talking heads (not shot by me) due to background chatter of people milling around.

Does anyone have any tips on how to minimise background chatter? I will be using a Rodelink wireless lav. Is it a case of reducing the sensitivity on the transmitter?

The only other mic I have is a NTG2 but I presume the lav will be better? I do not have a boom arm to hang the shotgun overhead.
 
The ratio of the SIGNAL (the subjects speech) vs. the NOISE (the ambient sounds) is called SNR (Signal to Noise Ratio).
Adjusting the gain of a wireless transmitter has NOTHING to do with SNR.
Using wired vs. wireless microphone has NOTHING to do with SNR.

Maximizing the speech and minimizing the ambient noise is the very definiton of SNR.

There are two ways of doing that:
1) Get the microphone as close as practictical to the subject's mouth.
2) Use a very directional microphone to maximize rejection from off-axis noise.

Note that (1) is FAR more effective than (2).

You did not reveal HOW you are conducting these interviews?
Is there an on-camera interviewer? You will very often see on-camera interviewers hand-holding a microphone right to the mouth of the subject. There is an extremely good reason for that. It is the very best way of getting decent audio in bad conditions.

How is your NTG-2 microphone for handline noise?
If all I had was a clip-on wireless lav and an NTG-2, I would be tempted to hand-hold the NTG-2.

Is there an off-camera interviewer? (Or producer or director, etc. asking questions)?
They can hold the microphone up to the subject's mouth just as if they were on-camera.

Do you have enough time to clip a mic in the right place on the subject? (And then remove it when you are done) If so, that is the second-best way of getting a microphone reasonably close to the subject's mouth. But it will seldom equal the performance of holding an NTG-2 up to their mouth.
 
Richard Crowley said:
The ratio of the SIGNAL (the subjects speech) vs. the NOISE (the ambient sounds) is called SNR (Signal to Noise Ratio).
Adjusting the gain of a wireless transmitter has NOTHING to do with SNR.
Using wired vs. wireless microphone has NOTHING to do with SNR.

Maximizing the speech and minimizing the ambient noise is the very definiton of SNR.

There are two ways of doing that:
1) Get the microphone as close as practictical to the subject's mouth.
2) Use a very directional microphone to maximize rejection from off-axis noise.

Note that (1) is FAR more effective than (2).

You did not reveal HOW you are conducting these interviews?
Is there an on-camera interviewer? You will very often see on-camera interviewers hand-holding a microphone right to the mouth of the subject. There is an extremely good reason for that. It is the very best way of getting decent audio in bad conditions.

How is your NTG-2 microphone for handline noise?
If all I had was a clip-on wireless lav and an NTG-2, I would be tempted to hand-hold the NTG-2.

Is there an off-camera interviewer? (Or producer or director, etc. asking questions)?
They can hold the microphone up to the subject's mouth just as if they were on-camera.

Do you have enough time to clip a mic in the right place on the subject? (And then remove it when you are done) If so, that is the second-best way of getting a microphone reasonably close to the subject's mouth. But it will seldom equal the performance of holding an NTG-2 up to their mouth.
Click to expand...

Very useful information, thanks.

The interviews are a 2-camera set-up with a producer off camera asking questions. Hand holding the microphone isn't an option unfortunately.

They are pre-arranged interviews so there will be time to light the subject and clip the mic correctly. All their videos must follow the same format so it's all fairly rigid.

I think they like doing the interviews so you can see the hustle and bustle of the event, rather than a dead room but it does give some audio issues. I will suggest shooting in a separate room but it might fall on deaf ears.
 
There's nothing wrong with background hustle and bustle on the audio either. If that is the atmosphere that the client wants, its a good thing. While it would be great to be able to add the background in a controlled manner (i.e. a good clean interview track with separate track picking up room tone), you can do a decent job by controlling the SNR as Richard described.
 
Good advice from Richard as usual! Most Lav mics are omni directional, I carry a uni directional Lav for just this situation. If you have the subject's back to the crowd that will help block some of the ambient sound. Do a test and listen to it in a quite space, it is hard to get a good sense of the ambient while live monitoring. Look for any existing sound dampening... curtains, office dividers, carpet on floor or walls.

Good luck
 
Richard Crowley said:
The ratio of the SIGNAL (the subjects speech) vs. the NOISE (the ambient sounds) is called SNR (Signal to Noise Ratio).
Adjusting the gain of a wireless transmitter has NOTHING to do with SNR.
Using wired vs. wireless microphone has NOTHING to do with SNR.

Maximizing the speech and minimizing the ambient noise is the very definiton of SNR.
Click to expand...

????
SNR in the audio world has nothing to do with background sounds except in a very abstract way and I think using it in this context could be very confusing to inexperienced readers.

SNR refers to actual noise, background sounds are not noise. They may well as in this case be unwanted sounds but that is NOT noise.

I know what you mean and if we were chatting I wouldn't object since I know that you know what actual noise is and I can roll with the abstraction. But if someone takes your statement literally then getting a mic with a good SNR would be expected to select out what I want to record and ignore the rest and we both know that is not the way it works.

New folks are constantly confusing "noise" and background sound so maybe I'm being over sensitive. But in a category where most equipment has a published SNR, using the term in a completely different context can really thow new users for a loop. Especially using it in a completely non standard (for the audio industry) way.

In a more general usage yes SNR is used as separating what you want from what you don't but in 35 years I have never heard it used in that context in sound except by newcomers who were confused on the topic and were trying to use noise reduction software to get rid of background sounds.
 
Richard Crowley said:
The ratio of the SIGNAL (the subjects speech) vs. the NOISE (the ambient sounds) is called SNR (Signal to Noise Ratio).
Adjusting the gain of a wireless transmitter has NOTHING to do with SNR.
Using wired vs. wireless microphone has NOTHING to do with SNR.

Maximizing the speech and minimizing the ambient noise is the very definition of SNR.

There are two ways of doing that:
1) Get the microphone as close as practical to the subject's mouth.
2) Use a very directional microphone to maximize rejection from off-axis noise.

Note that (1) is FAR more effective than (2).
Click to expand...

Yep. This.

I've done quite a few interviews on noisy convention floors using a simple Electro-Voice RE50N/D. An omni. 10 cm from the mouth of the speaker. The speaker's voice at that range is a huge signal and completely overwhelms the background noise. And this makes it imminently intelligible. Never had a complaint.

EDIT: If you gotta use a lav, get it close to the mouth (one button down from the collar instead of the normal two), and turn the guy to put his back to the biggest noise sources (use his back as a gobo to block the noise getting to the mic on his front). Don't pull him into a corner (reflections are likely worse), but even a step or two away from the worst of the noise will help some. Think, and listen. Trust your ears. Use those headphones. Focus on what you're doing.
 
Last edited:
Noiz2 said:
????
SNR in the audio world has nothing to do with background sounds except in a very abstract way and I think using it in this context could be very confusing to inexperienced readers.

SNR refers to actual noise, background sounds are not noise. They may well as in this case be unwanted sounds but that is NOT noise.
Click to expand...

Scott is an entirely practical guy with huge experience. And a gift for explaining complicated concepts in simple and redily understandable ways. And a desire to help and the patience to actually do it.

I rarely even consider disagreeing with him. But this is one of those times. Noise is, buy my definition, any sound I don't want. That can be babble in a restaurant, the squealing brakes of a metro train, or the gaussian hiss from microphone preamp working too hard. It can also be the voice of my wife interrupting me while I'm trying to listen to a Bach violin sonata at the end of the day. :violin:

This is similar to the definition of weeds. My wife once planted this interesting and quite beautiful passion fruit vine in the back yard. For a while it was indeed a beautiful plant. Then it started to spread. And rapidly. It came to the point where she couldn't control it. It became a weed, even to she who loved it and nurtured it. She gave up on it; it took me seven years (!) to eradicate it from the yard. Was it a "background sound" (lovely flower) or "noise" (weed)? Depends on who you ask, and when you ask.
 
I concur with my knowledgeable friends' statements.
My first choice would be an RE50 or other quality H/H omni. In lieu of that, a lavaliere would likely be best. However the common mistake of placing it up closer to the mouth in the neck line area should be avoided, unless the wardrobe dictates otherwise. A cardioid lav may help some, but is risky with it's other inherent characteristics.
If your expecting a silent BG in a noisy convention center scenario, it ain't gonna happen. So as long as you can 'see' the environment, I usually not too concerned about BG noise level.. as long as the subject is intelligible.
 
I was using SNR in the informal sense as a convenient measurement of desired vs. undesired sound.

​Signal-to-noise ratio is sometimes used informally to refer to the ratio of useful information to false or irrelevant data in a conversation or exchange. For example, in online discussion forums and other online communities, off-topic posts and spam are regarded as "noise" that interferes with the "signal" of appropriate discussion.
https://en.wikipedia.org/wiki/Signal-to-noise_ratio
Click to expand...

I can't say that I have ever seen a microphone with a published "SNR" specification. To be sure some microphones have published "self-noise" specs, but even there the ultimate "SNR" can only be determined if one also knows the SIGNAL level. And THAT is determined by how you USE it.
 
Last edited:
Bruce Watson said:
It can also be the voice of my wife interrupting me while I'm trying to listen to a Bach violin sonata at the end of the day. :violin:

This is due to you not having developed your Selective Hearing; practise makes perfect.

And thanks to everyone for their replies; I too have a job this week were this information will be very very useful.
Click to expand...
 
Richard Crowley said:
I was using SNR in the informal sense as a convenient measurement of desired vs. undesired sound.



I can't say that I have ever seen a microphone with a published "SNR" specification. To be sure some microphones have published "self-noise" specs, but even there the ultimate "SNR" can only be determined if one also knows the SIGNAL level. And THAT is determined by how you USE it.
Click to expand...

You are completely correct. It would have to be the recorder that filtered out what I didn't want ;~)
 
Bruce Watson said:
Scott is an entirely practical guy with huge experience. And a gift for explaining complicated concepts in simple and redily understandable ways. And a desire to help and the patience to actually do it.

I rarely even consider disagreeing with him. ...
Depends on who you ask, and when you ask.
Click to expand...
After blushing... Practical? I work in post and moved to Detroit! Insane maybe, not sure about practical ;~)

Anyway,
It is fine for you to think of it that way. My worry on this usage was demonstrated only a few days/ weeks ago on this list where someone was asking how to remove the noise on a recording and after six or seven responses about noise reduction someone asked about the "noise" and it was background sound not noise.

The reason I would disagree with your definition is not that it's intrinsically wrong but that it is too lose a definition to be really useful for communicating with other people. Allowing for a bit of overlap the tools one might try to remove actual noise are quite different from the ones you might try to minimize background sounds. Actual noise is fairly removable, background sounds (depending on the nature) almost impossible.

What I'm kind of trying to avoid is repeating this discussion every time a question about noise comes up to find out what the poster is actually talking about. I can call a recorder a storage device, but it make for really confusing discussions when I as which storage device works best in a bag.

We had pink Jasmine in SF that tried to take over the yard...
 
Back to the issue at hand...

If you are doing a two-camera setup with lights and with an interviewer, why is it impossible to add one more stand and get a hyper just out of frame? If the producer/director wants well-lit interviews, how is he/she unconcerned with audio?

If you want the best shot at this, use a hyper or short shotgun, as close as possible - I often use a blimp with the front cell screwed off which seems to block more side and back noise but gives me a really crisp voice (and with hypers, I use one of those XLR on-off switch barrels to get the mic right up to the front of the blimp). Without a blimp, you can sometimes even get the tip of the mic in frame (it's usually black) and it doesn't really read visually, or if it's a static shot, key in some piece of BG to hide it (sounds crazy but man, it can really work, especially if the BG is soft focus).

Then on a 2nd channel add your lav. Omni lavs will pickup a lot of noise; Cardioid lavs - you need to be aware of what the input direction is - can often be run in a shirt collar. Taping them to the inside of the "outside" fold of the collar will often point them right at the mouth. This will give you 2 feeds to choose from; you may find using the low end of the lav and the high end of the mic (or some sort of split-EQ strategy) will give you the best sound (watch for phase issues). A cardioid lav that close to the face may be pretty bassy anyway.

Wondering though - has anyone tried, in a high-ambient noise situation, to use two identical lavs - place one normally for the best voice pickup, and a 2nd one far enough away to have less voice, but aimed the same direction… then flip the phase of one mic in post and bring up the more distant mic and see if phase cancellation brings the ambient noise down? I know, it might sound like a 1960's flanger running, but in theory...
 
If the producer/director wants well-lit interviews, how is he/she unconcerned with audio?
Click to expand...

If I could answer that I wouldn't need to work. The reality is that most directors (and people in general) are much more visually oriented and just take sound for granted. To be fair not a lot of people understand how actively your brain filters and enhances what you "hear" based on context and emotional state. Many folks fall into the trap of "I can understand it so the audience wont have a problem" which always comes back to bite you in your hind quarters.

THe phase flip gets very problematic because of the distance between the capsules. To actually work the stuff you don't want needs to hit both capsules at the same time and in the same phase so you can flip one to cancel. The wave length of a 4K wave is just over 3 inches so if your second capsule is 3.4 inches from the first it would do a pretty good job of canceling out stuff around 4K. But 2K is twice as long so you would be close to 180 out of phase and you would be boosting 2K signals by quite a bit. This would also be the case for sounds around 8K.

It's the distance between capsules that people forget to take into account.
 
Another vote for the RE50/ND held close. This works as well as anything I know of for the OP purpose.

For a low budget, decent sounding and durable lav, you might also consider the Countryman EMW.

Grant
 
You must log in or register to reply here.
Back
Top