Hey guys,

I recently shot an interview inside a building that had a pretty bad echo. My shotgun mic did a pretty good job, but there's still a bit of echo in the audio. Can anyone recommend a method to help reduce it?

Thanks in advance for any tips.

I don't know how to solve the problem you have at the moment... but in the future use a hypercardoid mic rather than a shotgun mic for indoors.... A lot of the echo is caused by the tube on the front of the mic.

Yep I'd usually do that, but unfortunately the shotgun was all I had with me. It was supposed to be an outdoor shoot, but the weather changed our plans as it often does.

Izotope RX3 Advanced. Expensive, but removes reverb just like you asked. It's a wonderful program to have if you can justify the cost.

the DYvision may be a lower priced alternative for reverb removal, and the trial version is well worth downloading. DyVision Reverb Remover ... VST audio plugin from DyVision Works (http://www.dyvision.co.uk/reverbremover.html). It worked well for me and might sort out your problem for you.

The tube of a shotgun mic shouldn't add echo to the sound, as it is designed to cut out side reflections. There should also be slots in the sides of the tube to stop internal tube echoes. The echo, or more likely reverb will be the residual room reverb which your ears automatically compensate for, but will be picked up by any mic more than a few inches from a normal talking voice source.

For best results with interviews, a lavalier mic gives the best pickup as it is close to the sound source although I sometimes use a PZM (boundary layer) mic which reduces reflective pickup if a lavalier would interfere with the shot. I always carry a wired lavalier in my camera bag as takes little space and is always there if needed.

Roger

If you are using Adobe CC, I recommend using the Center Channel Extractor in Audition. It might not get it perfect but it will be a huge improvement.

Bob

@ Roger: Interference tube microphones don't reduce sound reflections equally across the frequency spectrum and when used in bright rooms the resulting comb filtering (fancy term that I'll confess to not really understanding) is thought by many to introduce what sounds like an echo to your recording. Some shotguns are better than others for this, but hypercardiods are supposedly better if you can get them in close enough. The big question is which is better if you can't the mic in close enough. In my experience, it's usually the hypercardiod, but opinions differ.

I've never noticed any echo effect when using a shotgun mic, but to be fair I rarely use them. Given a choice between hypercardioid or a lavalier, I would use the hypercardiod in a dry room, but as a safe bet, the lavalier in just about every other interview circumstance.

It is all a matter of personal choice though :-)

Roger

Thanks for the tips all. I'll try the suggested software and report back.

And yes, a lav mic is next on my to-buy list :)

Comb filtering occurs when signals arrive at multiple locations displaced in time. So with two or more shotguns, the effect of combining them (a bodge often tried in theatre) introduces a kind of strange phasing sound. The uneven frequency response of shotguns from LF to HF making this worse. However, comb filtering happens when you mix two omni mics - often when you have two people close together miked up, and they go in close - for quiet dialogue maybe just before a fleeting kiss.

It's a pet hate of mine when the sound gets described as echo. It's reverberation until you can separate the sound into two distinct versions - as in a real echo. The 'blurring' of the sound is reflections from walls, and is always worse when the boom op is not on the ball, and is pointing a very directional microphone only vaguely at the sound source. The shotgun, at the HF end is very directional, so if it's pointing at the wall, rather than the mouth, it hears the reflection very strongly, and the off-axis real sound source less so - making a mess. The commonly cited rule about 'no shotguns indoors' is usually what happens when people used to outside booming get when they A. Are too far away, and B. Not accurate in their aiming. Outside, these things are not as critical because with no reflections you can up the gain and all is well. The bad technique being obscured. These people often seem to have problems with wind noise too. Funny that!

An awful lot of TV people get decent sound indoors with a 416. Other people don't - and blame the mic! Moving a shotgun closer in by even a small distance makes a huge difference.

You won't be able to completely remove the sound of the reflections from the dialogue, but a noise gate can remove the delayed "echo" and will clean things up a bit.

Yeah, I have used the CS3e indoors many times and it has worked well. People think the no shotgun indoors rule is a lot more concrete than it really is.

Yeah, I have used the CS3e indoors many times and it has worked well. People think the no shotgun indoors rule is a lot more concrete than it really is.

Once I thought the same way untill I was given a cheap hyper to try, I rubber banded my 416 to a Samson CO2 mic inside the blimp and recorded them on seperate tracks for every shoot I did for about 3 months and left it up to the editors / producers to choose between mic A or mic B and tell me which one they used (they didnt know what they were).
For ALL of the TVCommercials, news / current affairs interviews and Docos done with the dual mics the 416 was chosen for outdoor shots for all but one occasion, but indoors the Hyper was chosen EVERY time.
For that number of varing shoots to consistantly come up with the same results has TOTALLY convinced me to become an 'Indoor Hyper convert'

Yeah, I have used the CS3e indoors many times and it has worked well. People think the no shotgun indoors rule is a lot more concrete than it really is.

I've used my Rode NTG2 indoors a few times and the resulting audio has sounded fine to me. I'm no audiophile though..

There is a specific issue with using shotguns indoors. They are not unidirectional at low frequencies, but have a rather large lobe of sensitivity to the back.

If you'll look at the diagram of a polar plot below, first, realize that it's an oversimplification. The polar pattern of a shotgun varies with frequency. This plot is for low frequencies. The problem noted below reduces as frequency rises.

Now, imagine that sensitivity out the back, with the back of the mic pointed to a reflective ceiling or wall, and now we get a whole bunch of out of phase LF going into the recording. This is classic comb filtering in the audio spectrum, and it sounds BAD!

On the other hand, in a different room, it's going to be different. Specific reflections, overall reverberation, distance from the mic, loudness of the subject, all play a part. Someone pointed out that the closer the mic is to the subject, the more direct the sound is - that's true too, and if you follow the guidance of the inverse-square law, a small change in position can have a larger effect than one might expect.

(diagram attribution: wikimedia commons)

Seth, it's even worse at low frequencies than that plot suggests. This plot below is more representative (though not quite accurate across the whole range.)

At high frequencies, there is a tight lobe, and many smaller lobes off-axis that can cause comb filtering.

At mid frequencies, the pattern is wider and smoother. You can get a rear lobe too. (For whatever reason, this plot doesn't show the high and mid rear lobes to be as strong as they often are.)

At lower frequencies, the lobe gets wider. This is the real culprit. If you don't have bass traps in the room, the reverberation can pollute the sound with a heavy, underwater feel.

At even lower frequencies, the mic becomes omni-directional, which would be represented with a circle.

I've got a 36 ft x 36 ft barn that we shoot in. We added moving blankets, which remove the high frequency reflections, but not the lows. It sounds okay in there to the naked ear. By adding the blankets, hand claps even sound decent. But with a shotgun, the low frequency reverb is terrible. The poor-man's way of solving it is to EQ out the lows in post, but you still hear the "echo" and the voices become thin. The better solution, of course, is a hypercardioid mic, placed as close as possible.

This plot (follow the link), from the Sennheiser MKE400, shows the stronger rear lobes that can occur, even with very short shotguns...
http://d1c86o3nm55du5.cloudfront.net/prodimgdetail/235994_a.jpg
.
.

Alrighty. Well I tried the Izotope RX3 Advanced demo, and it got rid of the echo completely. Unfortunately when I go to render the effect in Avid MC, it spits out a whole bunch of errors and crashes.

I was going to try Dyvision Reverb Remover but I don't run Windows.

Does anyone else know of any other software that might help?

Thanks in advanmce.

I see I'm too late, but I was going to suggest the plugin Unveil by Zynaptiq. Works with many different programs, and they have a free demo. I think it runs 300 or so (so same as Izotope?). It works. It's amazing.

I see I'm too late, but I was going to suggest the plugin Unveil by Zynaptiq. Works with many different programs, and they have a free demo. I think it runs 300 or so (so same as Izotope?). It works. It's amazing.

Not too late! I'll give it a try. Thanks.