First "noisy mic" doesn't mean anything, you need to listen and describe the noise to understand it. broadband or narrow band? periodic or intermittent? hum? crackles? 
And I'm not trying to be funny but I would really eliminate the noise before the signal hits Pd. If it is for live input you can't really use an fft based noise reducing process and filtering out parts of your signal only makes the sound even worse. There are no noisy mics, perhaps yours is broken and you need to replace it. A reasonable dynamic mic doesn't cost a lot, so do quality soldered cables.

Jurgen

2010/5/28 András Murányi <muranyia@gmail.com>
On Thu, May 27, 2010 at 12:15 PM, Andrew Faraday <jbturgid@hotmail.com> wrote:
Your hip~ at the end isn't needed if you've already got one at 50hz,
I'd be very cautious of trying to band pass just the fundamental tones of voice too, there's a lot of harmonics in the human voice, not to mention bits of 'noise' signal (like the 's' and 't' sounds) which need treble to be heard as language.

...given that, a band pass can still be useful, but the upper limit shall be set far above 255hz.
Actually, i think even the lower limit will be higher than 255hz, or around that.
My recommendation is to try a 300-3000hz band, and of course to listen and adjust accordingly.

Andras

_______________________________________________
Pd-list@iem.at mailing list
UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list