Dan Dugan wrote:
>>To do the conversion I simply feed the
>>sounds from the DAT to my iMac via analogue inputs, where the sound is
>>automatically taken up by the computer at 44.1 kHz.
>
>
> I wonder if you're clipping the signal; with something percussive
> like that frog it's hard to tell.
We found in a earlier discussion sometime back that the built in sound
in a iMac was pretty quirky. That was a discussion where we were working
to time stretch a recording and the problem was with the D/A. But it
makes me suspect this one may be some thing unique to the iMac as well.
If it's not a difference in the sound output equipment. The iMac version
is not playing through the DAT it's being compared to.
It's certainly easy to have some of these sharp frogcalls clip and not
see it. This is, however, only moderately sharp as frogs go and it's
characteristics are different from what I think of as percussive. It's
more complex in it's formation.
The call of this frog should be classified as a short trill rather than
a single pulse. It's median frequency is about 2.7khz. It is a series of
multifrequency pulses more or less 1/10 sec in duration each. The number
of pulses per call varies. Several frequencies are used ranging from 2
to 3.6 khz to make each pulse, minimum number of frequencies seems to be
6 and max is probably close to twice that. The energy in each pulse
varies as to it's time distribution from pulse to pulse, it is not all
identical pulses or calls. During the duration of the pulse the energy
rises and falls at about a .003 second interval, probably due to some
fundamental beat frequency. And you thought frog calls were simple...
In the case of this one, each sound pulse would have about 4400 samples
to characterize it. All of the info about the call I gave above is clear
in the CD recordings I have here. There's lots and lots of info present.
I have a number of frogs here that make trills that are similar in
general structure, it looks very familiar. Frogs often use frequency
composites to make sounds that sound more or less like one frequency.
Walt
________________________________________________________________________
________________________________________________________________________
|