Re: [gentoo-user] [OT]: Denoising software ?

luis jure <ljc@××××××××××××.uy> [11-12-10 13:28]:
> on 2011-12-09 at 13:55 Michael Mol wrote:
>
> >I couldn't tell you if it's necessarily "good", but Audacity has a
> >noise filter.
>
> that's exactly what i was about to reply.
>
> and if you want to try a CLI tool, sox provides a similar utility.
>
> noiseprof [profile-file]
>         Calculate  a  profile  of the audio for use in noise reduction.
>         See the description of the noisered effect for details.
>
> noisered [profile-file [amount]]
>        Reduce noise in the audio signal by profiling and filtering.  This
>        effect is moderately effective at removing consistent background
>        noise such as hiss or hum.  To use it, first run SoX with the
>        noise‐ prof effect on a section of audio that ideally would contain
>        silence but in fact  contains  noise  - such sections are typically
>        found at the beginning or the end of a recording.  noiseprof will
>        write out a noise profile to profile-file, or to stdout if no
>        profile-file or if `-' is given. E.g. sox speech.wav -n trim 0 1.5
>        noiseprof speech.noise-profile To actually remove the noise, run
>        SoX again, this time  with  the noisered  effect;  noisered  will
>        reduce  noise according to a noise profile (which was generated by
>        noiseprof), from profile-file, or from stdin if no profile-file or
>        if `-' is given. E.g. sox speech.wav cleaned.wav noisered
>        speech.noise-profile 0.3 How much noise should be removed is
>        specified by amount-a number between 0 and 1 with a  default  of
>        0.5.  Higher numbers will remove more noise but present a greater
>        likelihood of removing wanted components of the audio signal.
>        Before replacing an original recording with a  noise-reduced
>        version, experiment with  different  amount values to find the
>        optimal one for your audio; use headphones to check that you are
>        happy with the results, paying particular attention to quieter
>        sections  of  the audio.
>
>        On most systems, the two stages - profiling and reduction
>        - can be combined using a pipe, e.g. sox noisy.wav -n
>        trim 0 1 noiseprof | play noisy.wav noisered
>
> never compared the results, if you do, i for one would be very interested
> in your experience.
>
> IMO, it's much better to remove noise by small amounts in successive
> passes (taking a new profile each time, of course), than trying to remove
> too much noise in one pass.
>
> let us know how it went!
>
> lj
>

I have some results now. "Measurement" was only done with my two ears
in a before/after-way of doing. No spectrum analysis or other high
advanced stuff. 

Device: Recording was done with a Tascam DR2D field recorder (SDcard
recordings), which has a nice sound and a fine stereo image ... and a
slightly high noise floor.

Test:
I set the internal mics of the  Tascam to mid
gain and pushed the level to max. I think (read: "I dont know for
sure") that the level is simply a variable resistor in the signal
path. The gain is the amplification level...so the noise comes from
this source beside others. No AGC was active.

I went into the kitchen, put the recorder on the desk and switch it
on. I let it record its own noise for a moment or two and then start 
to boil water (high freqs in the sound) and I produced other sounds
which contain high freqs.

After a while I stopped recording.

As exspected, the recording was filled with some noise, which was not
destructive but fairly audible.

I loaded the file into audacity, and used the denoising filter.
The result has noticeable less noise but it was still audible.

Then I used gwc in conjunction with pulseaudio. I leave all the setting
alone and only selected for the Window function "Hanning-overlap-add"
and for the Noise Suppression Method "Lober & Hoeldrich", both marked
with "Best" in the settings menu.

I denoised the same input as above.

The result? There was no noise at all anymore -- at least for my ears.
Both files were checked using earphones and played with the Tascam
instead being played through the PC loudspeakers.
In the beginning of the file there were some audible artifacts, which
may result from too less read data...but this is guessed.

Gwc is a little fragile due the longer period it was not maintained
anymore. Its alsa interface simply does not work. 

But it clearly wins when it comes to denoising. It also supports the
removal of other audio artifacts.

HTH!
Best regards,
mcc