[Kde-accessibility] XML... I think

[Bill Haneman]

On Fri, 2003-01-10 at 21:38, Pupeno wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> On Friday 10 January 2003 08:27, Bill Haneman wrote:
> > regarding JSML and W3C, to follow up, I should have included some links
> > for you:
> >
> > w3c "Voice Browser" homepage:
> > http://www.w3.org/Voice/
> >
> > w3c comments on the original submission of JSML to w3c:
> > http://www.w3.org/Submission/2000/06/Comment
> >
> > Note that the 'speech synthesis' section of w3c's "Speech Synthesis
> > Markup Language" (aka 'JSML-2', aka 'VoiceXML') is now in "Last Call"
> > phase, so it should be a w3c Candidate Recommendation as a standard very
> > soon.
> Let me see as there are LOTs of diferent mark ups for doing similar things 
> with diferent overlaping functionality.
> SSML (http://www.w3.org/TR/speech-synthesis/) which is 'speaking' part of 
> VoiceXML (http://www.w3.org/TR/voicexml20/) which is a dialog specification 
> that uses SSML, SRGS, CCXML, etc is the same as JSML-2 ? (question too long, 
> in short, SSML is JSML-2 ?)

SSML looks to be the synthesis parts of JSML.

I don't really think an XLST adapter approach makes sense, since latency
is a big concern in text-to-speech for screenreaders.  One can't afford
the processing time.

I think we should both (gnome-speech and kde tts) try to use the w3c
speech-synthesis standard, for reasons that I think are fairly simple. 
As it stands, SABLE just isn't going to be that standard.  At the time
that the festival support for SABLE was designed, the w3c's activities
hadn't focussed on an XML language yet.

I notice that the festvox.org pages dealing with Sable are not working
anymore, and the festival sable docs make it clear that at the time the
sable plugin was designed, sable was not a stable API.  It appears that
the sable plugin was designed as a starting point for compliance with
some "future" XML standard; and I think the future is now, and the XML
standard will be w3c's SSML.  So I don't think that the sable spec is a
viable 'standard' to use.  I would not be surprised if an upcoming
festival release included SSML support, I certainly think from reading
the previous festival/XML docs that the intention is to support the
dominant, standardized XML DTD for speech synthesis.

Note also that the official SABLE mailing lists are/were hosted at Sun,
and Sun has moved its efforts to the w3c-sponsored SSML activity.  It
may be that SABLE's overall effort and momentum has been transferred to
the SSML w3c spec anyhow.  I can make some enquiries among my friends in
the speech synthesis section of SunLabs, but that's impression at the
moment.

regards,

Bill

> - -- 
> Pupeno: pupeno@kde.org
> KDE Accessibility co-maintainer
> http://accessibility.kde.org
> - ---
> Help the hungry children of Argentina,
> please go to (and make it your homepage):
> http://www.porloschicos.com/servlet/PorLosChicos?comando=donar
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.0.7 (GNU/Linux)
> 
> iD8DBQE+Hz1ELr8z5XzmSDQRAqIHAJ0aeYch2jTy5Fj483D5XQ5snyIuvwCdHc7K
> Kj9f9RK+Xie4YI1UVHSRJNw=
> =o46y
> -----END PGP SIGNATURE-----
> 
> _______________________________________________
> kde-accessibility mailing list
> kde-accessibility@mail.kde.org
> http://mail.kde.org/mailman/listinfo/kde-accessibility
-- 
Bill Haneman <bill.haneman@sun.com>