Towards a Dubbing and Audio Description exchange format

W3C has begun work on an open standard exchange format for audio description and dubbing scripts and wants interested people to review the draft requirements first published on 2022-05-10.

This post is by one of the TTWG Chairs. It's about this work and why it is important, and why now is a good time to be doing it.

As a Chair not only of the W3C’s Audio Description Community Group (open to all) and the Timed Text Working Group, but also of the EBU’s Timed Text group, I’ve been privileged to see that there’s a growing interest in both audio description and dubbing. As well as the more established vendors, there is a growing cottage industry of small, you might almost say hand-made, web based authoring tools, that each seem to use its own bespoke proprietary format for saving and loading work.

From a client perspective, this means that these tools do not interoperate with each other, and it can be hard to move from one to another. This is a classic case where an open standard exchange format would solve real needs. From conversations I have had, I believe implementers would welcome an open standard format.

From a user perspective, anything that makes it more likely to get an accessible experience, especially for users who are watching videos without necessarily seeing the images, must be a good thing. Audio Description and Dubbing are both important in this area.

Audio Description helps explain what is happening in the video image directly, in case the video content does not describe it adequately in the audio.

Dubbing is an alternative to translation subtitles: traditionally it has seemed that some countries culturally prefer one or the other, but perhaps we can make it easier for content providers to offer both and allow the user to choose.

Finally, if we can provide the script data as text content to the player, this opens up alternative renderings that are neither visible nor audible, for example using Braille displays.

The W3C Timed Text Working Group has agreed to work on creating an open standard exchange format that supports both dubbing and audio description, and has just published a first public draft Note describing the requirements that such a format needs to support.

The DAPT Requirements Note first published earlier this week, on 2022-05-10, will be used to define the Recommendation track specification, which will be a profile of TTML2.

We have published this as a draft Note because getting the requirements right at the beginning is really important, and we want everyone who is interested to review it and tell us how they can be improved.

The way we derived the requirements was to consider firstly the production workflow, then the needs of each step in that workflow, and finally break that down into a granular set of requirements, against which we can check the resulting specification.

Please do review the requirements document and feed back - the header material at the beginning of the document says how to get in touch.