It's worth playing with... I would be worried about loosing thrushes
and other spp in 3-5kHz range in the mix though, but if it worked well
would solve a huge problem.
Mike
Sent from my iPhone
On Aug 21, 2009, at 10:09 PM, Chris Tessaglia-Hymes <c...@cornell.edu>
wrote:
Aha! I did a little searching. It looks like it is possible to
cancel out the unwanted sounds, real-time. My first result was this,
which describes adaptive noise cancellation technology:
http://www.developer.com/java/other/article.php/3599661
Sincerely,
Chris T-H
Chris Tessaglia-Hymes wrote:
So, the question is: can the unwanted cricket and katydid sounds be
removed from the audio channel at the time of sound acquisition,
real-time, such that their acoustic signatures are minimized or
eliminated altogether from the collected sound data prior to an
automatic detector batch process?
Sincerely,
Chris T-H
Michael Lanzone wrote:
No software we have worked with gets near 100%. I have toyed
around with templates that got 95% of the calls, and detectors can
get ~90%, but more commonly get in the 60-80% range. In Louisiana
with the insects it would be on the low end of this. Katydids and
such are problematic for detectors...
Best,
Mike
Sent from my iPhone
On Aug 21, 2009, at 7:43 PM, Erik Johnson <ejoh...@tigers.lsu.edu>
wrote:
Hi All,
I've been recording from my home in south Louisiana with set-ups
like
Chris and David over the last few years. I've been using the
oldbird
software (tseep, etc), but only get about 20% of the flight calls
that
I would otherwise detect by ear (and visually on spectrographs).
Not
only is the detection software missing many calls, it's also
underestimating the richness that I could get. In one of my best
fall
nights I more than doubled the species richness by listening
through
the entire night compared to running it through the software.
What's
also frustrating is that I get a TON of trash clips - many more
than
birds clips. I've tried to filter out background noise (which is
mostly insects and air conditioning units) before running the file
through the auto-detect software, but it doesn't change the results
much. I haven't toyed with the other programs that have been
mentioned in this threat, but as I understand it, they also don't
get
near 100% - or am I wrong - it sounds like this technology
improving
quickly. This list serve is giving me new inspiration to hook up
the
mic this fall and to play around with more settings and
programs. I'm
eager to see the upcoming manuscript and to hear everyone's
thoughts
on this subject!
Happy listening,
Erik Johnson
Lafayette, LA
ejoh...@lsu.edu
--
NFC-L List Info:
http://www.NortheastBirding.com/NFC_WELCOME
http://www.NortheastBirding.com/NFC_RULES
http://www.mail-archive.com/nfc-l@cornell.edu/maillist.html
--
--
NFC-L List Info:
http://www.NortheastBirding.com/NFC_WELCOME
http://www.NortheastBirding.com/NFC_RULES
http://www.mail-archive.com/nfc-l@cornell.edu/maillist.html
--
--
=============================================
Christopher T. Tessaglia-Hymes
TARU Product Line Manager and Field Applications Engineer
Bioacoustics Research Program, Cornell Lab of Ornithology
159 Sapsucker Woods Road, Ithaca, New York 14850
Voice: 607-254-2418, FAX: 607-254-2460
http://www.birds.cornell.edu/brp mailto:c...@cornell.edu
=============================================
--
NFC-L List Info:
http://www.NortheastBirding.com/NFC_WELCOME
http://www.NortheastBirding.com/NFC_RULES
http://www.mail-archive.com/nfc-l@cornell.edu/maillist.html
--
--
NFC-L List Info:
http://www.NortheastBirding.com/NFC_WELCOME
http://www.NortheastBirding.com/NFC_RULES
http://www.mail-archive.com/nfc-l@cornell.edu/maillist.html
--