- From: Silvia Pfeiffer <silviapfeiffer1@gmail.com>
- Date: Mon, 8 Oct 2007 10:14:05 +1000
Hi Chris, this is a very good discussion to have and I would be curious about the opinions of people. CMML has been developed with an aim to provide "html"-type timed text annotations for audio/video - in particular hyperlinks and annotations to temporal sections of videos. This is both, more generic than captions, and less generic in that captions have formatting and are displayed in a particular way. One option is to extend CMML to provide the caption functionality inside CMML. This would not be difficult and in fact, the current "desc" tag is already being used for such functionality in xine. It is however suboptimal since it mixes aims. A better way would be to invent a "caption" tag for CMML which would have some formatting functionality (colours, alignment etc. - the things that the EBU subtitling standard http://www.limeboy.com/support.php?kbID=12 is providing). Another option would be to disregard CMML completely and invent a new timed text logical bitstream for Ogg which would just have the subtitles. This could use any existing time text format and would just require a bitstream mapping for Ogg, which should not be hard to do at all. Now for Ogg Skeleton: Ogg Skeleton will indeed have a part to play in this, however not directly for specification of the timed text annotations. Ogg Skeleton is a track that describes what is inside the Ogg file. So, assuming we would have a multitrack video file with a video track, an audio track, an alternate audio track (e.g. speex as suggested by you for accessibility to blind people), a CMML track (for hyperlinking into and out of the video), and several caption tracks (for different languages), then Ogg Skeleton would explain exactly that these exist without the need for a program to decode the Ogg file fully. I think we need to understand exactly what we expect from the caption tracks before being able to suggest an optimal solution. If e.g. we want caption tracks with hyperlinks on a temporal basis and some more metadata around that which is machine readable, then an extension of CMML would make the most sense. Regards, Silvia. On 10/8/07, Chris Double <chris.double at double.co.nz> wrote: > The video element description states that Theora, Voribis and Ogg > container should be supported. How should closed captions and audio > description tracks for accessibility be supported using video and > these formats? > > I was pointed to a page outlining some previous discussion on the issue: > > http://wiki.whatwg.org/wiki/Video_accessibility > > Is there a way of identifying which track is the closed caption track, > which is the alternate audio track, etc? How are other implementors of > the video element handling this issue? > > Is CMML for the closed captions viable? Or a speex track for the > alternate audio? Or using Ogg Skeleton in some way to get information > about the other tracks? > > Chris > -- > http://www.bluishcoder.co.nz >
Received on Sunday, 7 October 2007 17:14:05 UTC