W3C home > Mailing lists > Public > www-voice@w3.org > January to March 2002

Comments on W3C Working Draft, 23 October 2001

From: randy <rsolton@zygot.com>
Date: Tue, 19 Feb 2002 16:41:07 -0500 (EST)
Message-ID: <003a01c1b97d$d6009ee0$6401a8c0@pacbell.net>
To: <www-voice@w3.org>
There are three related issues that I would like clarification on:

(1) In section 2.3.1.1 Fields Using built-in Grammars,  in the table where the various built-in types are listed, there appears to be incomplete information regarding how field values are spoken:

For types boolean, digits, currency and time, there is a mention of how the values are spoken, If the field value is subsequently used in a prompt. But for types date, number, phone, no mention of how field values are spoken.

(2) In section 4.1.4, <value> Element, it states "The manner in which the value attribute is played is controlled by the surrounding speech synthesis markup.".  Is this always the case?  What if the <value> expr is a field name?  Wouldn't it then follow the rules stated in section 2.3.1.1?

(3) In 4.1.2.6 <say-as>, there are several types listed.  There seems to be some overlap with the built-in grammar types listed in section 2.3.1.1.  Some types listed in 2.3.1.1 (e.g. boolean), seem appropriate for <say-as> type but are not included.  Also, some <say-as> types (e.g. measure) seem appropriate as built-in field grammars.  Finally why is there is a <say-as> type called telephone, while the equivalent built-in field grammar is called "phone"?

Is there any thought of aligning the built-in field grammar types with the <say-as> types?

Thanks,
-- Randy
Received on Wednesday, 20 February 2002 05:38:43 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 30 October 2006 12:48:54 GMT