ELAN language archiving technology

[ROOD DAVID S]

Mark, I have been using ELAN since it was first introduced in a very 
experimental form in the DOBES program in 2000.  It is being steadily 
improved.  I have had no complaints since they fixed some of the really 
horrid bugs from the early versions.  I am using it only for annotating 
Wichita videos, though I have also put one Lakota audio text into it in a 
preliminary way.  We're using it for the Lakota videos in my current 
project, too.

You can set up an unlimited number of tiers for different kinds of 
information about the data.  In the Wichita, I use tiers for individual 
speakers, Wichita transcription, morphological analysis, glosses, grammar 
coding, and comments (some tapes require different things, depending on 
what's on them).  In the Lakota we're using one tier for an "underlying" 
transcripton and another for phonetic (fast speech) deviations.

I don't know what other similar programs are out there, but for aligning 
video, audio, and transcription, I think ELAN is great.  For precise 
phonetic analysis, however, I go elsewhere (Praat, Sound Forge).  I 
haven't used this as a database, i.e. I don't know much about the search 
functions, but I did get a very nice printout of a Lakota text formatted 
exactly the way I wanted to display it for classroom use.  Keep in mind 
that I am probably the most computer illiterate Siouanist on the list -- 
if I can figure it out, I think anyone else can, too.

I haven't seen any more "labor intensive" phenomenon than would be needed 
for any other program.  If you don't have anything in writing, you have to 
write it somewhere at least once, and putting it into ELAN is no worse 
than anything else.

There is one counter-intuitive feature that bothers me.  To save whatever 
you have just typed, you need to hit "control enter" -- otherwise your 
typing disappears.  A few experiences with that problem are enough to 
inspire learning to do it right, however.

Best,
 	David


David S. Rood
Dept. of Linguistics
Univ. of Colorado
295 UCB
Boulder, CO 80309-0295
USA
rood at colorado.edu

On Wed, 3 Mar 2010, Mark J Awakuni-Swetland wrote:

> http://www.lat-mpi.eu/tools/elan/
>
> Aloha all,
>
> Are any of y'all familiar with ELAN, a language archiving program?
>
> The UNL Center for Digital Research in the Humanities (CDRH) folks are
> suggesting I look at it for possible applications to the Omaha database
> work in progress.
>
> My CDRH guy give me an initial demo today..
>
> The program requires an operator to insert transcription directly linked
> to audio video files.
>
> It is XML based, so supposedly is UNICODE compliant.
>
> It looks like the straight text transcription can have a second line with
> parts of speech, and a third line that renders everything in IPA.
>
> I just sent over a short wav file with Omaha elicitation to see how we can
> deal with an Omaha orthography.
>
> It is possible to search on text words (presumably Omaha, too, if that
> works).
>
> The visual ability with the wav files does make me think that prosody
> studies could be done.
>
> Dumping the wav files into Audacity or other software could give you pitch
> contours and such, like what Rory Larson has been playing with recently.
>
> Bottom line, It looks really labor intensive up front... but might have
> some uses down the road
>
> Ideas?
>
> Thanks
> Mark
>
> Mark Awakuni-Swetland, Ph.D.
> Assistant Professor of Anthropology
> and Ethnic Studies (Native American Studies)
> University of Nebraska
> Lincoln, NE 68588-0368
>
> http://omahalanguage.unl.edu
> http://omahaponca.unl.edu
> Phone 402-472-3455
> FAX: 402-472-9642