> and the demo file with a single bird vocalisation, starts
with the insect din in place, they are removed during the
bird call and are present at the end minus the background /
just the insect cacophony.
http://urlme.net/audio/bugswat.wav
Mike,
It's an excellent noise reduction and I had to have a play
with this one. :-)
I amplified it and slowed it down 10 times and did a
sonogram at 4098 samples:
www.stowford.org/sounds/bugswat10pc.mp3
www.stowford.org/sounds/sonogram10pc4096.gif
As before the main harmonics are approx 3300 to 7700 which
is typical of a croak call with low first and second
harmonics. Not suggesting what made the croak. :-)
Then I listened to the main part of the call on a loop with
cans. Ears are still the best tool even though they are a
bit slow for most wildlife. As with the original, there are
a lot of echoes along with multiple pathways "smudging" the
call. You can also see these on the sonogram along with
possible periodicities at about 30ms or 10 metres path
length differences but these may be speed change artifacts.
There are also some inevitable artifacts like the 15KHz
sharp cutoff ringing, but the insects are remarkably low
with a constant frequency of around 3650 (original speed)
just visible. How much was the NR software?
David
David Brinicombe
North Devon, UK
Cogito cogito ergo cogito sum - Ambrose Bierce
|