RE: Pronunciation Lexicon Markup Requirements

	I have a couple of comments on this draft:

	2.1 - some of the issues which this document treats as unresolved
(e.g. pronunciation alphabets and suprasegmental information) have already
been decided for the Speech Synthesis Mark-up spec: interoperability
presumably requires that these documents concur.

	4.3 - syntactic category is only a "should have", but if this is
absent i don't see how the treatment of homographs (4.6, a "must have") can
be achieved.

	5.10 - can we please distinguish between acronyms (those items, such
as DEC or NASA, which are made up of capital letters but are pronounced as a
word) and other sequences of capital letters such as all the examples given
in this draft?
	the provision of lexica for cases such as BT, MIT, ATT and NEC is
probably trivial - all that is required is a pronunciation for each letter
of the relevant alphabet (Roman, Greek, whatever).
	the handling of cases such as NASA should be a by-product of
allowing for variations in upper and lower case: i assume that in most Roman
alphabet languages the pronunciation of NASA, Nasa and nasa can be the same
(although there are problem cases such as US/us, MIT/mit, ...).
	syllable-based abbreviations (e.g. VoPo for Volkspolizei in German,
or the contraction of Japanese names) may be more problematic.

	6.7 - WHY?!
	other working groups have made exactly the opposite decision. if
vendor-specific alphabets are supported, how do you decide which ones?
(members of W3C? "the big guys"?) - you can't support them all! speaking as
a vendor, i see no problem in defining a vendor-independent standard and
then requiring vendors to provide a transform between their proprietary
alphabet and the standard. this could be part of the definition of
compliance.


	best,
		alex.

Received on Thursday, 22 March 2001 05:43:28 UTC