3

I've hit a stumbling block on a relatively simple project. I have a helmet, a Contour camera on which I have piggybacked a microphone interface to put a mic inside my helmet, to be able to speak and reduce wind noise a bit. However, it is still proving too noisy for my liking.

I have three options, and I was wondering if a fourth existed. They are as follows:

  • Post-processing. Intensive, difficult and annoying. Would work as a last resort or to fine-tune.
  • Dual-microphone set-up, one within the helmet, the other one outside it. This requires an active IC, which I'd rather avoid (no easy access to the Contour battery :-( )
  • High-pass filter. Gets tricky at low speeds, works fine at high speeds.

My question is simple: is there a way to use microphone correlation (or lack of it) using just a passive circuit, effectively substracting one feed from another, without requiring an IC?

Sébastien Renauld
  • 173
  • 1
  • 4
  • 2
    Have you tried using wind screens or wind muffs, http://www.bhphotovideo.com/find/newsLetter/pro_audio_minimize_windnoise_July07.jsp – Garrett Fogerlie Dec 28 '12 at 13:52
  • 2
    How about multiple microphones spread out a bit but still near the mouth. Each would get somewhat different wind noise, so the overall wind noise will be attenuated when all the signals are averaged, but the voice component will presumably be the same for all microphones, so it won't attenuate after averaging. – Olin Lathrop Dec 28 '12 at 14:13
  • You can add microphones passively if you must. – Russell McMahon Dec 28 '12 at 15:02
  • @OlinLathrop: The wind noise will be decorrelated at the various microphones only at frequencies for which the spacing among the microphones is a significant fraction of the wavelength. In other words, this is just a complex way of creating a low-pass filter (which is what I assume the OP really meant with option #3). Also, with this method, the SNR only improves with the square root of the number of microphones used. – Dave Tweed Dec 28 '12 at 15:51
  • @Dave: Yes, multiple mics only works down to some frequency related to the spacing, but it can still help. Let's say you can get 5 inch separation. That's about the wavelength of 1.3 kHz, which means it won't help below half that or about 650 Hz. That doesn't solve the whole problem, but is still a useful result. As you say, you get diminishing returns with more microphones, but one on either side of the mouth with good separation will help and is cheap and easy and can be used together with other techniques. – Olin Lathrop Dec 28 '12 at 16:45

1 Answers1

2

For this type of application, you need a different kind of microphone known as a throat microphone, which doesn't depend on picking up sound from the air in the first place.

Dave Tweed
  • 172,781
  • 17
  • 234
  • 402