Deleting simple sound
traces.
Sample 1. Deleting
frequency modulated single harmonicum noise.
In computer
system “System Workshop” expert can select different time-segments with
different sonoobject traces. These segments may overlapped in time.
On the next
top picture you can see PASFilm of original sound with yellow dots (selected
sound traces) and red dots (unselected sound traces). On the bottom of the same
picture you can see PASFilm of the same original sound with deleted noise
traces.
To
listen original sound with frequency modulated single harmonicum noise (top
picture) click here.
Sample 2. Deleting
frequency modulated poliharmonicum noise.
This sample
is like the previous. But in this original signal (top PASFilm) you can not
even hear any traces of voice. They all masked by powerful frequency modulated
poliharmonicum noise. Please, be careful to listen this original signal. Save
your ears.
After
deleting by expert noise traces on PASFilm you can hear the voice and
understand speech.
To
listen original sound with frequency modulated poliharmonicum noise (top
picture) click here.
Compare power level
result.
To estimate
how deep speech signal was under the noise on original sound you can see the
dynamic of power on the top of the next pictures (1 grid on power axis = 10
dB). The most powerful segments of speech signal is weaker the noise on 20-25
dB.
