- From: Henri Sivonen <hsivonen@hsivonen.fi>
- Date: Fri, 14 Jun 2013 09:58:56 +0300
- To: Karl Dubost <karl@la-grange.net>
- Cc: John Foliot <john@foliot.ca>, Duncan Bayne <dhgbayne@fastmail.fm>, public-restrictedmedia@w3.org
On Fri, Jun 14, 2013 at 8:05 AM, Karl Dubost <karl@la-grange.net> wrote: > You can have accessibility with DRM, you can have accessibility without DRM, because it is unrelated. Not unrelated. Consider the case of book publishers choosing to turn off text-to-speech features on Amazon's Kindle. You might argue that it's not DRM that lets them turn off an accessibility feature but a contractual relationship with Amazon. However, DRM is what prevents a third party that doesn't have a contractual relationship with either Amazon or the publishers from creating a product that reads out loud DRMed Kindle books. One might argue that this is irrelevant to EME, since EME is not about text. However, if your computer is Google's data center, speaker-unspecific speech-to-text is already possible as seen in YouTube auto-captioning. It's conceivable that in the future deaf people could run speaker-unspecific speech-to-text on their own computer or, alternatively, pipe the audio stream into a data center for computing speech-to-text there. In that case, the compliance rules of the DRM applied to the soundtrack become accessibility-relevant. (I expect some readers of this list to take offense at the notion of computer-generated captions saying that users are entitled to human-generated captions anyway, but I think my point about DRM and accessibility not being unrelated stands despite people probably taking such offense.) -- Henri Sivonen hsivonen@hsivonen.fi http://hsivonen.iki.fi/
Received on Friday, 14 June 2013 06:59:24 UTC