system.Speech Breaking changes

[Chris Smedley]

Perhaps I am the only person out here attemtping to work with the February
CTP of WinFx in the system.Speech area, but has anyone else found a number
of breaking changes in the grammarbuilder object.
Robert Brown's samples in the MDSN mag for example dont seem to be working
as we no longer have an AppendChoices or the AppendResultKeyValue methods as
described.

I am inferring that these are now withi the .Append method, but it would be
nice to get confirmation of this and perhaps some refactoring notes...

anyone have any insights ?

Thanks for your time

[Andrew Ahearne]

Hi Chris,

You're not on your own, I'm also using the Feb CTP version of SpeechFX. As
for any breaking changes that have occured, you're right they are all
covered under various overloaded versions of the Append method. Is there
anything these changes prevent you from doing?

Andrew

"Chris Smedley" <chris....@terracomputing.com> wrote in message
news:eD9ySIg...@TK2MSFTNGP12.phx.gbl...

[Chris Smedley]

Thanks for the reply Andrew.
At some point I would like to see more MSFT involvement here in this
newgroup, or perhaps vista speech needs its own forum as a lot of the issues
here are SAPI or SASDK related and I am not sure that Vista speech is
getting the audience ( not that SAPI and SASDK are not important stuff, use
that as well).

In general there is so little about the various functions that its a bit
like a black art at this point to intuit the developers intent.

The removal of AppendChoices seems to have been refactored into Append
(Choices), so that fairly straight forward.

But the AppendResultKeyValue seems to have been broekn into the need for two
separate calls one for the result key and one for the value. Perhaps is this
to pass multiple values for a specific key?

As I said earlier, notes in the Speech DLL other than generic call
informatioon and the stuff you get from the Object Browser would be a plus
IMHO

"Andrew Ahearne" <and...@news.microsoft.com> wrote in message
news:eRsJoOkU...@TK2MSFTNGP09.phx.gbl...

[Andrew Ahearne]

Yeah support is fairly thin on the ground but maybe that's because they're
busy writing documentation :-) If you have a question you can't find an
answer to then send a mail to lis...@microsoft.com and they may be able to
help you. Speaking of helping you, in your other post you asked about mixing
in dictation with your grammar. I have discovered that this is quite easy to
do. All you need to do is use GrammarBuilder.AppendDictation() where you
want to include it. If you need to specify a dictation in a W3C Grammar XML
file then it's simply a case of creating a ruleref with a
uri="grammar:dictation". Finally Chris I actually owe you thanks, in trying
to help you I actually found the solution to my own problem ;-)

Andrew

Oh by the way a trick I use frequently to discover what is going on is to
use the SrgsDocument.WriteSrgs(XmlWriter). This outputs the grammar into an
XML format you can use to figure out what's going on. In fact if you have
Visual Studio .NET 2003 and the Speech Application SDK installed somewhere
you can open a visual representation of your grammar and run test input
strings agains it, amongst other things.

"Chris Smedley" <chris....@terracomputing.com> wrote in message

news:Omm5CqmU...@TK2MSFTNGP12.phx.gbl...

[Chris Smedley]

Andrew, I sent you an email on this under separate cover. I found the same
thing this morning and was going to post the result only to find you had
already done so.. More information follows in the email on specifics how
this might be done.

Thanks for the collaboration

"Andrew Ahearne" <and...@news.microsoft.com> wrote in message

news:e6lqa$oUGHA...@TK2MSFTNGP14.phx.gbl...