[music-dsp] Random Noise Reduction
arcane at arcanemethods.com
Thu Jan 15 09:05:01 EST 2004
Yaakov Stein wrote:
> > Looking for pointers on random noise reduction which, given
> > a good characterization of the spectrum,
> which spectrum - that of the noise or that of the signal ?
> > will minimally impact the quality of the sound. Looking for moderate
> > reduction within the quality constraint and also to minimize
> > the sound of noise pumping.
> > Trying to see what I can do to improve a noisy mic for low
> > SPL recording.
> If the noise is stationary and uncorrelated with the signal,
> and you know both spectra, then your best bet is a Wiener filter.
> This can be implemented in either time or frequency domains.
Yes, it is stationary and uncorrelated but I'm afraid little
can be assumed about the signal spectrum. I'll research
> If it is uncorrelated but nonstationary,
> and you believe that it develops in time according to some linear system,
> then Kalman filtering will help.
> If the noise is relatively low and your signal is well represented by an AR model
> (e.g. speech or harmonic music) then LPC prefiltering is the way to go.
Not sure about this. While the noise is realtively low,
about 35 dBA SPL equivalent, the signal can be any kind of
> If the noise is stationary and the signal is not always present,
> then spectral subtraction might help (but destroys phase).
Not sure about this either. The average signal over short
intervals can have wide dynamic range and, while I am
interested in a method whose effect is diminished when the
average signal gets larger and masks the noise, it should
not be at the expense of audible noise pumping.
Basically I am wondering if anything can be done practically
to compensate the relatively high self noise of very small
mic capsules. The acoustic self noise of a diaphragm goes
up as the diameter of the diaphragm goes down and that is
inescapable but there are some interesting designs for first
and second order directional mics that require tiny capsules
to work. I'm hoping to find a way to ameliorate this design
conundrum to some extent with DSP methods.
"Things should be described as simply as possible, but no
More information about the music-dsp