[Athen] Captioning Web Videos - and some feedback from you all
jfoliot at stanford.edu
Wed Oct 8 13:36:15 PDT 2008
Apologies if you get this multiple times, as I am sending it out to a
number of web accessibility lists.
Creating captioned web media is something that is not easy to do right now
- this we already know. Getting captioned iTunes media is even harder,
although it is possible, as the following link(s) will illustrate:
(these are both also available via the iTunes store:
The exciting thing is however, that it can be done, although it remains a
bit of a manual challenge to produce these files - a simple "turn-key"
solution really doesn't exist right now (but there *is* work being done in
Along with the challenge of the technical part of this process however is
some other, less technical aspects, and the purpose of this note: In the
first referenced video above (atr13.mov) the video stream features some
rap lyrics in foreign languages. For the purpose of the
caption/transcription, my associate Kim simply noted "Japanese Rap",
"German Rap", etc. Given that she's already adding to her work flow by
ensuring her podcasts are now captioned, I congratulated her on a job well
done and started telling everyone about the files above.
But the question of multi-lingual content within a caption file remains
one that I am unsure of, exactly. Is this as much a "policy" kind of
question as it is one of technical discussion:
1) should we provide the foreign transcript in the caption file? (this
raises questions of i11n as well, especially with the Japanese rap and
foreign character sets)
2) should we provide, instead, a translation of the rap? (even if it is
somewhat out of context; plus then we are providing information that
non-caption choice users are not getting, so is this right?)
3) is Kim's choice of simply noting "Japanese Rap" or "German Rap"
4) What kind of support do the various media players have for i11n
switching on the fly? (If in fact this is what needs to happen, or will
all the foreign text/character sets simply follow through from the initial
transcript/caption file? - I am thinking of alternative output as well -
Braille for example). Can we (should we) be thinking about the lang
attribute here (DFXP can accept it) - and if yes, how, and will, Adaptive
Technology be able to take advantage of it?
5) are there any other aspects of multi-language multi-media that I've
overlooked that we should be thinking about?
Now I suppose that these questions are also ones posed within the
caption/transcription community already (TV and Movie for example), but I
have not found any definitive information about this so far (in my limited
Google quest), and the delivery via the web of multi-media has its own
unique abilities and constraints that other forms of media delivery may or
may not share.
So... thoughts? Comments? Ideas?
Realizing that you might not subscribe to all of the lists I'm sending
this out to, I will recap the discussions on Friday, Oct. 10th.
Thanks in advance for your participation
More information about the athen-list