Erik Johnson wrote:  "What's also frustrating is that I get a TON of trash 
clips - many more than birds clips."

To be clear, I'm a hobbyist with limited time, so I use detectors *assuming* it 
will give acceptable results more quickly than viewing/listening to sound files 
directly.

Unfortunately, as Mike Lanzone points out, Trash-versus-Bird is one trade-off 
when using detectors.  However, this trade-off can be mitigated by an efficient 
tool to sift through the trash.  For the this discussion, I'll say the software 
detection process has two major phases: the software detection itself, and then 
the human classification phase (trash versus bird).

Not sure if others agree, but as others work to improve the detectors, I think 
a quick win is an improved tool for the 2nd phase, wheat-vs-chaff 
classification.  

For example, last night I ran a file through a Raven detector graciously 
forwarded by Mike Powers.  Examining the results with Glass-of-Fire, I labelled 
one sound out of 200+ detections as a bird (same as when I used Tseep/Thrush 
against the file).   This was quick and painless.

However, individual review of Raven detections revealed I *incorrectly* 
labelled 7 bird calls as Noise in Glass-of-Fire.  This is because Glass-of-Fire 
stretches spectrograms to a pre-defined size, rendering the bird calls visually 
unrecognizable.  So, the detector found birds, but the efficient classifier was 
inaccurate.

Manual review of each Raven detection was accurate, but highly inefficient:  
viewing hundreds of selections one-at-a-time is slow and tedious.  The bounding 
boxes effectively hide short sounds.  Keeping or deleting good/bad 
selections from the selection list is error prone.

Glass-of-Fire is a great format: view page-fulls of spectograms, and quickly 
classify them with key combos.  A great improvement would be to present 
spectrograms without stretching.  To use Raven detections with a Glass-of-Fire 
style viewer, it would be helpful to see more sound around the Raven 
detection.  For example, in the case of a longer bird call it 
successfully detected part of the call, without selecting the whole sound.  In 
the case of a short call, it's difficult to understand what you're looking at 
without seeing more context around the sound.

Regardless, I think increased efficiency during human classification should 
allow current detectors to flag even more sounds, catching more bird calls 
along with the trash.


Thanks,
Eric








________________________________
From: Chris Tessaglia-Hymes <[email protected]>
To: [email protected]
Sent: Friday, August 21, 2009 2:09:37 PM
Subject: [nfc-l] NFC Detectors and Equipment?

Hi everyone,

In the past, I have not used any detectors when going through my night 
recordings at home (Etna, NY). I have collected my sound data from the roof-top 
microphone (Evans-style, with a Knowles microphone element) piped into my home 
computer running Raven Pro, recording a continuous file sequence from start to 
finish with each file duration equal to 1 minute. The following day, I would 
browse through the sound file sequence by hand, again using Raven Pro, looking 
for sounds of interest. Once a sound of interest was worth saving, as an 
example of a good flight note for species x, or an interesting unidentified 
species flight call, I would cut-and-paste that sound file into a new window 
and save it with a time-stamped label, uniquely pairing it to the file/time it 
was copied from.

Now, this is all fine when you are a single person, operating your own home 
station, only recording on those nights which appear to have good night 
flights. But, when you begin operating to capture every night from multiple 
stations, or you want to really quantify most or all of the calls that night, 
the question of data storage and data processing becomes the big issues.

How do some of you out there collect your sound data?

What tools do you use for browsing sounds?

Do you only use detectors?

Here's a question for probably three people on this list:

What is the difference between the current Raven Pro detector that Mike Powers 
provided settings for and the old BirdCast transient detector? Is there a 
difference?

Getting back to an earlier posting from Tom Fowler (prior to the bloom in 
membership...140+ now!), what kind of equipment do you each use for recording 
or listening to your sounds?

I mentioned that I use a variation on the Bill Evans-style flowerpot 
microphone. I know that Andrew Farnsworth and Mike Powers use a microphone, 
pre-amp, and housing designed by engineers at Bioacoustics at the Cornell Lab 
of Ornithology, storing their night sounds on flash memory inside a SoundCache 
for analysis later, but what do others use?

What are your personal home recording setups like?

What obstacles or limitations have you encountered with your equipment setups 
or recordings?

I realize these are a lot of questions, but I wanted to pose these to the list 
in order to help initiate discussion along these lines.

Information about Bill Evans's flowerpot design can be found here: 
http://www.oldbird.org/ (click on Microphone Design in the left panel)

Information about the Raven software can be found here: 
http://www.birds.cornell.edu/brp/raven/RavenOverview.html

Another sound analysis software tool, Syrinx, can be found here: 
http://syrinxpc.com/

Thanks and good night listening!

Sincerely,
Chris T-H

-- Chris Tessaglia-Hymes
Listowner, NFC-L
Ithaca, New York
[email protected]
http://www.NortheastBirding.com/NFC_WELCOME
http://www.NortheastBirding.com/NFC_RULES


--
NFC-L List Info:
http://www.NortheastBirding.com/NFC_WELCOME
http://www.NortheastBirding.com/NFC_RULES

http://www.mail-archive.com/[email protected]/maillist.html
--



      
--
NFC-L List Info:
http://www.NortheastBirding.com/NFC_WELCOME
http://www.NortheastBirding.com/NFC_RULES

http://www.mail-archive.com/[email protected]/maillist.html
--

Reply via email to