Reducing Sound Leakage
December 10, 2009
I was asked this on Twitter yesterday:
Rob, I’m trying to figure out how to eliminate mic phasing for podcasting. Using 2 AT2020 USB mics in a “normal” room. Any tips?
For the purposes of this conversation, I’m going to widen the scope of the question and include sound leakage between the mics. When sound intended for one microphone is able to be heard in a second microphone, you’re getting sound leakage. When the delay is short enough that it doesn’t sound like a delay, but instead alters how the original source sounds, you’re hearing phasing. When you reduce the amount of leakage, you’re also reducing the amount of phasing. These problems can be solved in a number of non-technical ways. Because I don’t have photos of Alex’s setup, I’ll address as many as I can think of. I’m going to address the issue without advising you to separate the microphones, as that can ruin the feel of a recording very quickly.
You can use the pickup pattern of the microphones you are using to help cancel out sound. The AT2020 has a cardioid pattern, which picks up sound in the direction the mic is aimed, as well as sound from the right and the left. The pattern naturally ignores sound from 180 degrees where the microphone is aimed. If the microphones are placed facing each other, a good portion of direct sound from the other person will be cancelled out. The image to the right is of a cardioid pickup pattern. The thick black line represents how sensitive the microphone is in 360 degrees, with a dearth of sensitivity facing in the opposite direction of the microphone.
Facing the microphones away from each other solves the problem of direct sound, but Alex mentions that he is recording in a “normal” room, which probably means a room that has some natural reverb. A second method for eliminating leaking and phasing in audio recordings is to speak closer into the microphone. Speaking closer will not only give you the sexy-sounding proximity effect, but it will allow you to lower the gain on your microphone pre-amp. When there is more direct sound going towards the microphone, it will pick up more of your voice. With the gain on the pre-amp lowered, you’re lowering the amount of noise from the room.
A real world example of a closer source is the sound of a phone handset vs the sound of a speakerphone. When someone is speaking directly into the microphone of their handset, the ratio of the person talking to the background noise and room reverberation is high. When you’re on a speakerphone, the person speaking is further away and the phone has to compensate by being more sensitive to sounds in the room, which is why you’ll hear lots of reverb, toe tapping and typing on the keyboard from that guy who isn’t paying attention.
Using these two techniques, you can dramatically improve the audio quality of your show without spending a dime on sound dampeners, isolators or new microphones. Nor will you have to dramatically change how your show is recorded.
If you have a question for me, you can fill out the contact form on the site or send me an @robblatt on Twitter.

