We've been looking at something along these lines but as we understand it, voice recognition typically get's you only about 80% of the way there. We never thought that was acceptable. We just outsource our captioning.


Christina Salazar
Systems Librarian
John Spoor Broome Library
California State University, Channel Islands
805/437-3198
________________________________________
From: Code for Libraries [[log in to unmask]] on behalf of John Wynstra [[log in to unmask]]
Sent: Wednesday, February 06, 2013 1:16 PM
To: [log in to unmask]
Subject: [CODE4LIB] On-the-fly Closed Captioning

I have been asked to find out whether there are software or hardware
solutions for on-the-fly closed captioning.  We currently work with
University IT production house on campus to perform this task.  I'm not
involved in any aspect of this at this time, but have been asked to
investigate.

Workflow is like this:
1) purchase a separate VHS copy of movie for captioning purpose (license
issues I believe)
2) view show and write a transcript (probably time consuming)
3) Campus IT production creates a closed captioned digital copy using
transcript and movie.

This is costly and time consuming for what often amounts to a single
viewing of an education resource that is not closed captioned out of the
box.

So basically, I'm asking if there is a "magic black box" that will allow us
to  bypass steps 1,2,and 3.  Just play the VHS and caption it on the fly
using voice recognition software and maybe a cray supercomputer on the back
end or even IBM's Watson if it is not playing jeopardy or going to school.

Thoughts?

--
<><><><><><><><><><><><><><><><><><><>
John Wynstra
Library Information Systems Specialist
Rod Library
University of Northern Iowa
Cedar Falls, IA  50613
[log in to unmask]
(319)273-6399
<><><><><><><><><><><><><><><><><><><>