At 08:32 AM 9/9/02 -0700, you wrote:
>I got word back in March that there is a team of two fellas at Oxford
>or Cambridge in the UK who have developed reasonable match-to-sample
>software but have been unable to track them down. Anyone know who
>that might be?
>
>Bernie
>
> > wrote:
> >
> >> This is an over simplified view of the process. The day when one can
> point a
> >> microphone at a vocalizing bird and get a digital readout
> >>identifying it will
> >> never be a practical reality in my opinion.
> >> John Arvin
> >
> >I doubt that will keep folks from trying. It's definitely a challenge.
> >
The trouble with any sample matching is pitch shifts and big-time timing
imprecision. The animal sound recognition seems to work more like a
loosely-specified phase-locked loop - it looks - correction, - it listens
for frequencies in certain ranges, locks to them, checks whether they go up
or down, start or stop, within broad guidelines, then decides. since 1967,
I have thought a hard-wired multiple PLL box for each species would work
much better than a digital approach, but who cares, I haven't built one yet.
But I agree with John Arvin. Back when the Pentium first hit the streets,
a fellow (?) from Hydro-Quebec claimed to me they and Cornell Bioacoustics
people had a Pentium box "and a lot of truck batteries" out in the woods,
censusing birds by sound as to species, location, date and time. I asked
for details and things got very vague from that point on.
I am not holding my breath.
In designing EnjoyBirds, I deliberately stayed away from "computer ID"
directions, and made it to be a teaching, planning, learning tool to help
PEOPLE become better birders. Who wants the computer to have all the fun,
anyway.
cheers,
Marty Michener
MIST Software Associates
PO Box 269, Hollis, NH 03049
coming soon : EnjoyBirds - software that migrates with you.
________________________________________________________________________
________________________________________________________________________
|