In search of closed captioning on web video

In the web environment I develop for – a busy hospital – having some kind of captioning on presentations that include an audio track is not optional. Videos and narrated presentations are watched in common areas, and it’s not always possible to locate a pair of headphones or even convenient to use them if there are other things going on in the room that need to be heard.

There are also a few people with hearing disabilities that need the captions. So I’ve been searching for a captioning method for all our streaming video that is efficient, and creates web-based video that is compatible with Windows and Macintosh computers.

Although the Virage system will be ideal for this (voice recognition and screen-scraping create searchable, linked chunks of text associated with the exact locations in the video where they occurred.), it will not be in place soon enough.

Creating captions on VHS/DVD formats would be a nice capability, as well.

Continue reading

Subcategories in MoveableType

David Raynes just came out with a great new plugin for MoveableType: “SubCategories” Get itHere

Code for the subcategories list on my main index page


<MTSubCategories show_empty="0">
<MTSubCatIsFirst><ul></MTSubCatIsFirst>
<li>
<a href="<$MTCategoryArchiveLink$>"><$MTCategoryLabel$></a><br>
<MTSubCatsRecurse>
</li>
<MTSubCatIsLast></ul></MTSubCatIsLast>
</MTSubCategories>

Continue reading

Virtual Reality Field trip

mu3d_tour_02.jpg

The VR gurus at the University of Michigan took us on a short trip into their realm a couple of days ago (just before the power failure!). They demonstrated three viewing options for their VRML files: The Cave, a rear projection screen, and standard web VRML plugins like the Cosmo player.

The Cave is a huge cubical room, about 10 feet on a side, with one wall removed. Each wall is a rear projection screen, except the floor, which gets its image from a mirrored projection unit placed in the ceiling (at least the mirror is – I am not sure where the projection came from.) The offset images are polarized, apparently they are alternated about 100 times a second, so if you view them without glasses, you will see a slightly doubled image. The viewpoint is determined by the “leader” who wears a set of polarized glasses (kind of like Oakleys on steroids) wired into the system. When the leader turns their head, the perspective – and separation of the images – changes to suit them. The glasses have shutters on each eye -they alternate open and shut, in sync with the image. The 3D effect is quite good. Everyone else in the room wears shuttered lenses also, but not wired in, so the viewpoint may be a little off for some. They demonstrated an architectural model of “the Barcelona House”, and a human heart.

Continue reading