Re: [ARSCLIST] voice recorders with speech-to-text software

[andy kolovos <akolovos@xxxxxxxxxxxxxxxxxxxxxxxxx>]

Could you elaborate more?  Are these digital palm recorders (the digital
equivalent of the old micro-cassette recorders) that have the Dragon
software function built in?  Could you supply some model numbers so I can
check them out?

If they are indeed digital palmcorders, most people I know who are involved
in producing and archiving oral history/ethnographic interviews (me
included) really discourage the use of these units for field recording for
a variety of reasons.  The big three are as follows:

1. Their record quality generally isn't very good, and recording quality is
extremely important to our (Vermont Folklife Center) mission.  I argue
everyone interested in doing this work and saving recordings should be
concerned about this as well.  The better the recording, the greater its
value to future researchers.

2. Most seem to lack a mic-level input to attach an external mic

3. They generally create audio files in proprietary formats and/or
compressed formats, which flies in the face of standards-based digital
preservation techniques

As far as the Dragon (and other voice recognition) software goes, and this
is all based on hearsay as we are dedicated to keeping a human being
employed as a transcriptionist, my impression is that at this point in its
development voice recognition software still has to take a bit of time to
"learn" a voice before it can translate it into text.  If you are
interviewing slews of different people, the way we do, it would have to
"learn" each new voice--and when you get into dialect and such, I imagine
you'd have a hell of a time cleaning up the mess.  All this strikes me as
time better spent trusting our transcriptionist's ear and typing skills.

Down the road, who knows?  I imagine voice recognition software will just
get better over time, but for now I don't think it is well suited to oral
history/ethnogrpahic interview transcription.

If you're using it just for your own dictation, however, that's a totally
different matter.

andy

p.s. you might also want to check the archives of the HNET Oral History
list on the topic of voice recognition software and the Dragon product in
particular, I think you can access the list archives via this page:

http://www.h-net.msu.edu/~oralhist/

There has also been a lot of discussion on digitil voice recorders as well.



At 08:04 AM 07/03/2004 +0200, you wrote:
>Did any of the archives dealing with oral history experiment with these
>devices? The latest Sony models, equipped with the Dragon software, are
>being offered at very attractive prices and having gone through twice
>recently to the long and painful transcription process I was thinking of
>trying one.
>Anybody did already?
>Francesco Martinelli
>(Siena Jazz Archive)
>
>
>
>
>---
>Outgoing mail is certified Virus Free.
>Checked by AVG anti-virus system (http://www.grisoft.com).
>Version: 6.0.712 / Virus Database: 468 - Release Date: 27.06.2004
>
>
*********************************
Andy Kolovos
Archivist/Folklorist
Vermont Folklife Center
P.O. Box 442
Middlebury, VT 05753
(802) 388-4964
akolovos@xxxxxxxxxxxxxxxxxxxxxxxxx
http://www.vermontfolklifecenter.org