naturerecordists
[Top] [All Lists]

Re: [Nature Recordists] Re: High Sample Rates

To:
Subject: Re: [Nature Recordists] Re: High Sample Rates
From: Rob Danielson <>
At 3:20 PM -0700 8/17/09, Dan Dugan wrote:
>
>
>Steve Pelikan, you wrote,
>
>>=A0 the question of archiving recordings so that future researchers can
>>=A0 learn about and get access to them and get the associated data is a
>>=A0 significant one. The current best approach I know of is to donate or
>>=A0 make plans to donate to one of the acoustics archives associated
>>=A0 with a university or museum. Many have long-term plans to protect
>>=A0 their data. In this context of future researchers, detailed field
>>=A0 notes about the recording might be even more valuable than the 2x or
>>=A0 4x samples. I've often though of adopting a scheme to put data
>>=A0 =3Dpossibly large bits of text, photos, other documentation- into the
>>=A0 sound file itself. Something like RIFF would let us do that, I guess.
>
>I agree, archiving in an institution with useful metadata is the most
>valuable thing we can do. No one will listen to a recording unless
>they know what's on it.
>
>I'm trying to figure out how to create a workflow for efficiently
>annotating long unattended recordings. I'd like to be able to scan a
>spectrogram of the whole file, drop markers with text labels, and then
>automatically produce a table of contents from the markers that gives
>the position of each marker in absolute time from the beginning of the
>file. I think this text data could be saved back to the header of
>the .wav file so it would be self-documenting. I'd like it to be able
>to show all channels summed or multiple channels in parallel. And do
>it on a Mac.
>
>For years I've been using markers in Pro Tools, but they have no legs;
>they are only good in Pro Tools. Pro Tools has no spectrogram display,
>and the markers can't be printed out, or referenced to the start of
>the file.
>
>iZotope Rx gives me a nice spectrogram of the file, and has markers,
>but the markers have to be edited in a separate step and can't be
>exported or printed.
>
>Soundtrack Pro might be able to do something like this, but when I
>loaded an hour-long stereo file and switched to spectrogram display,
>it took a long time to calculate the spectrogram and then it took
>forever to do -anything-, even to pull down the help menu! This is on
>a Mac Pro with four processors. I had to give up.
>
>Suggestions?
>
>-Dan Dugan
>

Hi Dan--
Have you tried Amadeus Pro?=A0 It has robust meta tag capability, text
markers (exported as a .txt file with times from the start), handles
long recordings very well with a pretty quick sonogram option. Good
batch processing features. The .txt file can be converted to a
database-compatible format. The author has been very helpful with
some bugs I've encountered.

Would it be useful if the generated table of contents had playable
links? QuickTime Reference "movies"=A0 are the only solution I know of
for this feature. I bet someone could write an Applescript to take an
Amadeus Markers list and produce QT reference movie links.=A0 A script
to insert a sonogram as a still image to a Reference movie (sound)
should also be doable.

As for hybrid stereo/surround playback, iTunes will play QuickTime
Normal & Reference movies in stereo and surround.=A0 Once on a server,
stereo and surround files will also play in a browser on Mac (and
supposedly Windows with QT installed and configured). Here's a link
to 4 channel surround file I recorded last night. You can see if it
plays in your browser the way your audio card is currently configured:

http://tinyurl.com/npck3g (43 mb QuickTime)
(If not, here's the OS/card adjustment for mac OS X:=A0 Audio/Midi
set-up (an app in the Utilities folder) ->=A0 Default Output (Select
your 4+ channel card); Properties For (Select your 4+ channel card);
Configure Speakers -> Multi-channel -> Left Front =3D 1; Right Front =3D
2; Left Surround =3D3 and Right Surround =3D 4; Center 5 and LFE=3D6. Click=

"Apply."=A0 Sorry, I'm not sure how to do this configuration on PC with
QuickTime installed and a multi-channel card connected to four or
more speakers-- but it should be doable. Might try Firefox too?)

Here's a 2.6K "Reference" movie which should play only a part of the
above longer file (in 4 ch surround): http://tinyurl.com/mnbdwu

QT handles Amadeus' markers. AP also has good batch processing.

For me, the tallest unknown is the database side-- I want to be able
to "drop" a prepped sound file into the database (with tags or linked
to other data otherwise) and have the associated documentation
generated automatically in both the on-line and off-line formations.
There's still a lot of work in my daily "flow" at present.=A0 I'm just
trying to do it consistently so the conversion process will be fairly
uniform if not=A0 swift.

Anyone, maybe the subject-line should describe this new discussion if
you wish to continue?=A0 Rob D.

--



------------------------------------

"While a picture is worth a thousand words, a
sound is worth a thousand pictures." R. Murray Schafer via Bernie Krause
Yahoo! Groups Links













<Prev in Thread] Current Thread [Next in Thread>
  • Re: [Nature Recordists] Re: High Sample Rates, Rob Danielson <=
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