DIY Active Noise Cancellation

[Al Skierkiewicz]

Originally Posted by DanArgent

On a side note, are you sure, that Condensers are $2000+ I work with a few sure condensers that I think were more in the range of $1000.

Dan,
The Shure's are electret, are you using the SM81? That is an electret. True, non electret condensors will be from manufacturers like AKG, Sennheiser, and of course the Neuman U67 tube mic. Many of them will be in large cases to hold the electronics needed to convert the condensor output to a useable signal. Many mics are coming out that look like classic mics with big cases and heads that are electrets made to look like a classic. Remember that each mic has it's own distinctive sound and good audio people can tell a mic (model) by listening to it.

[Pat McCarthy]

Does anybody know if there is an active noise cancellation program you can download that uses your computer's mic and speakers/headphones? If so, could you post a link to the site were it can be found?

-Pat

[Bharat Nain]

Originally Posted by PCPat

Does anybody know if there is an active noise cancellation program you can download that uses your computer's mic and speakers/headphones? If so, could you post a link to the site were it can be found?

-Pat

I don't know how possible it is for a program to cancel noise, just because it has no input for what noise is going into our ear. It doesnt know what to cancel. I'd really be amazed to find a program that actually cancels noise.
-Bharat

[Al Skierkiewicz]

Techno et al,
There are all sorts of ways to detect noise and cancel it out or pull data from it. Modern DSP (digital signal processing) can do marvelous things. For just about one hundred dollars there are DSP add ons for ham radios that reduce noise 20-30 dB. In these applications, the DSP is not noise canceling as much as signal enhancing. Noise cancelling for me means multiple loud sounds only one of which you want to hear. You filter out the noise, invert it and add it back in out of phase to "cancel" the unwanted sounds.
Now there are other methods for pulling signals out of a mixed environment. NASA uses these techniques in order to communicate with distant objects like the Mars explorers. It is easy when you have an unlimited power supply or huge antenna array, to crank out enough power to overcome the losses of distance in radio communications. But when you are limited by power and size, you must come up with other ways to insure "no data radio" will hamper you project. By using these techniques, NASA is able to demodulate signals that are below the noise floor. For standard voice communications, you need about 35-40 dB signal to noise for the parties to be understood. If both of you know that a limited vocabulary will be used or are willing to repeat messages until understood, then less than 30 dB S/N is tolerable. For data and morse code (with no DSP) a few dB S/N is OK. When you get to the point where the signal is 10-20 dB below noise, you better have a powerful DSP set and high gain antennas to get the data through.

[TBNAudioEnginer]

Hello All,

Well, this is interesting, Robot Comps? I think thats what I've stumbled upon...(scratches head) Anyway, My name is Jonathan, I'm the Audio Engineer for a TV Station in Canton, Ohio. I've been having a phase interference problem with trying to use 4 omni-directional microphones (on an interview set). Thus, I've been doing constant research about phasing concepts. Along the way, I found this forum. You might find a google search of "noisebuster" to be useful.

[Al Skierkiewicz]

Jonathon,
There is no real way to counteract the phasing you are experiencing other than riding the mics. The phasing is occurring simply by the length of time sounds takes to travel. So as one person speaks, his voice is picked up by the closest mic first and then the next closest, etc. Since all the mic signals are mixed together you are hearing four different signals slightly out of time alignment with each other. At audio frequencies, these slight time differences convert the mixed signal into a comb filter(s). As with any notch filter, phase changes rapidly going into and out of the notch frequency and that is what you hear. The depth of the notch(es) is determined by the level of each of the contributing sources and the frequency is determined by the acoustic distance between the mics. Two mics at equal level can add in phase to only +6dB but out of phase can add to an infinite depth notch. (i.e. >85dB) If you need any further info, please PM me. I am a broadcast maintenance engineer for WTTW/Chicago with specialty in audio.