W3C home > Mailing lists > Public > wai-xtech@w3.org > August 2010

Re: TTS from web content?

From: Jason White <jason@jasonjgw.net>
Date: Mon, 23 Aug 2010 20:08:58 +1000
To: wai-xtech@w3.org
Message-ID: <20100823100858.GA13847@jdc.jasonjgw.net>
Cain, Sally <sally.cain@rnib.org.uk> wrote:
 
> The answer to all of my questions may be that it would be used loads and
> yes it is well worth it. However from the perspective of blind and
> partially sighted people if you are using a screen reader, you will
> probably continue to use your screen reader rather than change to using
> the speech within the page. 

In some scenarios there may not be anything resembling what is currently known
as a screen reader. I have in mind, in particular, the situation in which the
operating system exists only to run a Web user-agent, which then becomes the
complete desktop environment for the user. There have been proposals of late
involving minimal operating systems (for example, a kernel plus basic system
libraries plus an X11 environment and a minimal window manager) on top of
which the Web user-agent runs. All of the user's interactions are with the Web
interface, in this situation.

Now in this circumstance, the best solution might well be to implement braille
or speech interfaces entirely in Javascript, which could be downloaded from
the network as needed. (I am assuming that the text to speech functionality
would be available as part of the operating system, and likewise for the
interface to braille hardware, but there would be no screen reader - instead,
there would be Javascript interfaces to the relatively low-level braille and
speech functions, together with an entirely Javascript-based assistive
technology somewhat like Fire Vox, which could be downloaded in the same
manner as any other Web application, and cached locally to facilitate instant
access.

I don't know whether anyone is working on such a technology, but the point is
that it's entirely consistent with one of the directions in which the client
side of the Web is headed, and it wouldn't require any real technological
innovation to develop. Its main distinguishing feature is that it would
function entirely within the Web user-agent; it wouldn't interact with the
underlying user interface libraries of the operating system as screen readers
do; but since the Web would be the desktop environment, the fact that the
assistive technology would comprise a Web application need not be a
disadvantage. In fact, it would have better access to the internals of the Web
user agent than is provided today by accessibility APIs, so one can imagine
scope for innovation in entirely Javascript-based assistive technologies that
would be harder to achieve with screen readers in their present form.
Received on Monday, 23 August 2010 10:09:29 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 27 April 2012 13:16:09 GMT