naturerecordists
[Top] [All Lists]

Re: Match-to-sample software?

Subject: Re: Match-to-sample software?
From: Walter Knapp <>
Date: Sun, 08 Sep 2002 16:30:43 -0400
Wild Sanctuary wrote:

> Does anyone know of a good match-to-sample software that is
> available? Looking for something that could be conformed to identify
> bird vox w/in the context of an entire biophony.

If you are trying for just any bird that might be out there, I somehow
doubt it. There are attempts underway. The frog folks are a bit ahead of
the birds, but I've seen nothing even for frogs that was a product
available publicly. It's all been custom software, generally aimed at
one or a few species.

The problem is approximately the same as computer voice recognition, and
the software much the same. Except you are dealing with the variations
of a lot more than just one species. Even characterizing the precise
necessary parts of calls to the degree the computer needs is probably
incomplete. So, for instance they may work on software for Ivorybills,
but we have so little info on their actual call that it could easily
miss them.

There is software that will react to a sound in a given frequency range
of a preset loudness, in fact there is a device that will do that
without a computer. It was originally designed as a froglogger, and the
frequency range is a bit low for all birds, though they will customize a
unit to your needs. They, in fact, did not even make the range as high
as I'd need if I wanted to to be settable for all species of frogs here.
The little grass frog goes 7.5-8kHz.
http://www.titley.com.au/ecopro.htm
I'd still love to have a bunch of these to set up.

This sort of logger, set to a frequency range and level could at least
minimize the length of recording you had to sift through.

Walt



________________________________________________________________________
________________________________________________________________________

<Prev in Thread] Current Thread [Next in Thread>
Admin

The University of NSW School of Computer and Engineering takes no responsibility for the contents of this archive. It is purely a compilation of material sent by many people to the naturerecordists mailing list. It has not been checked for accuracy nor its content verified in any way. If you wish to get material removed from the archive or have other queries about the archive e-mail Andrew Taylor at this address: andrewt@cse.unsw.EDU.AU