Re: Caption synchronization tolerance

 >Meeting minutes say:
 >
 >>#67WC ask Geoff Freed or someone at WBGH.
 >>JW GV didn't want a number.
 >>Action WC: Ask Geoff.

GF:
Did I miss Wendy's question?  If so, I apologize.  Sometimes I'm a little 
too fast with the delete button.

 >>3. descriptions and captions are synchronized with the events they
 >>represent to within a tolerance of X. [Note: We need to research the
 >>tolerance. Any information on this is appreciated.]

GF:
In his reply, Joe hits most of the right points about captioning.  Here's 
another one:  in off-line captioning situations, professional captioners 
try to make captions change with shot changes.  This makes for smoother 
transitions.  However, with Web-based video technology, it's nearly 
impossible to *exactly* synchronize a caption with a shot change.  "Close" 
or "reasonable" synchronization, therefore, is, well, reasonable.  Someday 
this will improve.  We'd be making a mistake by setting a tolerance level 
of X frames or seconds, but a basic rule of captioning should be noted:  if 
someone is speaking, or if there's an important sound effect occuring, 
there should be text on the screen.  I recommend against anything more 
specific than that.

Descriptions are trickier, especially now that multimedia authors can pause 
video and program-audio tracks in order to insert extended 
descriptions.  In other words, we're no longer limited by the existing 
pauses in the soundtrack.  It is acceptable to describe events slightly 
before or after the fact.  But pausing the movie introduces a whole new set 
of advantages and problems which are just now being researched.  Thus, I 
recommend the guidelines say something simple like, "Use audio descriptions 
to describe important on-screen events," and let the author decide what is 
important, if the movie should be paused, etc.

Geoff Freed
WGBH/NCAM

Received on Friday, 24 August 2001 14:37:49 UTC