Re: mathml to plain text | audio?

Hi Naylor,

>
> On Tue, 1 Jun 2004, Helder Ferreira wrote:
>
> >
> > Hi Andreas,
> >
> > True. MathML Content Markup seems at first better. However there are 2
open
> > issues:
> >
> > (1) MathML Content Markup only covers basic math. The operator's
dictionary
> > and support is not very big. OpenMath is doing a better job on that.
>
> However MathML Content Markup may reference OpenMath content dictionary
> and is therefore as 'big' as OpenMath using the semantics or csymbol
> elements.
>

True. But instead of supporting only simple MathML Presentation or Content,
you'll have to support OpenMath notation, or even other types of notations
(for instance: Latex, Mathematica, Maple, ...) as well.
Example using OpenMath:

<csymbol encoding="OpenMath"
definitionURL="http://www.openmath.org/cd/setname1#N">N</csymbol>

Now you need to go to the definitionURL and understand what N means in
OpenMath. This escape sequence from MathML was a nice way of include the
world around it, but openned new issues. Such as the interpretation of
whatever is on the other side of the definitionURL.

The other example using OpenMath is using xml-annotation:

<annotation-xml encoding="OpenMath">
  <OMA xmlns="http://www.openMath.org/OpenMath">
  <OMI>N</OMI>
  </OMA>
</annotation-xml>

But this also implies that you know OpenMath and convert it.

Note: I'm not saying that we shouldn´t use Content Markup. I'm only saying
that its use has a few issues .

> > (2) Who uses MathML Content Markup in WebPages? Noone. The major
> > contribution on the web is based on Presentation-Markup. So seems to me
that
> > a program that speaks MathML needs to understand Presentation Markup
too.
>
> so including content MathML in web pages is simply a matter of including
> the required XSLT stylesheet in an appropriate place and include a
> processing instruction pointing to that stylesheet, more information may
> be found about this (including stylesheets:David Carlisles) at:
> http://www.w3.org/Math/XSL/Overview-tech.html
>
> hope this helps,
>
> Bill
>
> >
> > Concluding the following:
> >
> > 1. A program that does MathML Audio Rendering needs to support both
Content
> > and Presentation Markup.
> > 2. By parsing Presentation and interpreting the markup we are in deed
doing
> > a kind of Presentation to Content Markup (which is not an easy task, and
not
> > being implemented yet), and then converting the meaning of the
mathematical
> > expression into plain text and therefore in audio using a TTS.
> >
> > Regards,
> >
> > Helder
> >
> >
> > ----- Original Message ----- 
> > From: "Andreas Strotmann" <Strotmann@rrz.uni-koeln.de>
> > To: "Helder Ferreira" <hfilipe@fe.up.pt>
> > Cc: <David.Pawson@rnib.org.uk>; <www-math@w3.org>; "Diamantino Freitas"
> > <dfreitas@fe.up.pt>
> > Sent: Tuesday, June 01, 2004 2:24 PM
> > Subject: Re: mathml to plain text | audio?
> >
> >
> > > Note that even though Aster is used to render LaTeX, T.V. Raman's
> > > dissertation makes it clear that his system does an enormous amount of
> > > work to recognize the logical structure of the formulas it reads.
> > >
> > > If I understand T.V.Raman correctly, for text / audio rendering,
> > > MathML-Content should therefore be a much better-quality input than
> > > MathML-Presentation, and I suspect that MathML-Presentation text/audio
> > > rendering should really be done as a two-phase process that first does
> > > MathML-Presentation to MathML-Content recognition.
> > >
> > >   -- Andreas
> > >
> > > Helder Ferreira wrote:
> > >
> > > > You can also look into the T.V. Raman's work about ASTER.-
> > > > http://www.cs.cornell.edu/Info/People/raman/aster/demo.html .
> > > > However ASTER works for Latex documents. Nevertheless, you can
transform
> > > > Latex to MathML (see
> > http://www.orcca.on.ca/MathML/texmml/textomml.html ).
> > > >
> > > >
> > > >
> > > >
> > > > ----- Original Message ----- 
> > > > From: <David.Pawson@rnib.org.uk>
> > > > To: <www-math@w3.org>
> > > > Sent: Tuesday, June 01, 2004 12:52 PM
> > > > Subject: mathml to plain text | audio?
> > > >
> > > >
> > > >
> > > >>
> > > >>www.daisy.org is a group of organisations who produce smil based
> > > >>'talking books'. Synchronised text and audio for blind and partially
> > > >>sighted people.
> > > >>
> > > >>We are starting to look at incorporating mathml into our schema,
> > > >>and that's fine for producing braille, just hard work.
> > > >>
> > > >>However, for 'spoken' math(s), it presents a problem  of automation.
> > > >>For more complex stuff, human readers tend to make mistakes, hence
> > > >>synthetic speech is often used.
> > > >>
> > > >>Put simply, mathml isn't 'readable' in its XML format (my view :-).
> > > >>Certainly if serialised it wouldn't sound very good via a TTS
engine.
> > > >>
> > > >>I'm informed others have looked at mathml to 'plain text'
transforms,
> > > >> perhaps with paragraph level formatting.
> > > >>I'm wondering if any such transforms are available open source, or
could
> > > >>be licenced to the daisy group for our readers?
> > > >>
> > > >>Any pointers appreciated.
> > > >>
> > > >>Regards DaveP.
> > > >>www.rnib.org.uk
> > > >>
> > > >>Not subscribed to this list, so please copy me in.
> > > >>
> > > >>
> > > >>**** snip here *****
> > > >>
> > > >>-- 
> > > >>DISCLAIMER:
> > > >>
> > > >>NOTICE: The information contained in this email and any attachments
is
> > > >>confidential and may be privileged. If you are not the intended
> > > >>recipient you should not use, disclose, distribute or copy any of
the
> > > >>content of it or of any attachment; you are requested to notify the
> > > >>sender immediately of your receipt of the email and then to delete
it
> > > >>and any attachments from your system.
> > > >>
> > > >>RNIB endeavours to ensure that emails and any attachments generated
by
> > > >>its staff are free from viruses or other contaminants. However, it
> > > >>cannot accept any responsibility for any  such which are
transmitted.
> > > >>We therefore recommend you scan all attachments.
> > > >>
> > > >>Please note that the statements and views expressed in this email
and
> > > >>any attachments are those of the author and do not necessarily
represent
> > > >>those of RNIB.
> > > >>
> > > >>RNIB Registered Charity Number: 226227
> > > >>
> > > >>Website: http://www.rnib.org.uk
> > > >>
> > > >>
> > > >>
> > > >
> > > >
> > > >
> > > >
> > >
> > >
> >
> >
> >
> >
> >
>
>
>

Received on Tuesday, 1 June 2004 10:39:58 UTC