<speech> tag
Printed From: MSSpeech-Forum
Category: Windows™ Speech Recognition Forums
Forum Name: Macros and Macro Development
Forum Description: WSR Macros, Tools, Files, and new Macro requests
URL: https://www.msspeech-forum.com/forum_posts.asp?TID=189
Printed Date: 23/Nov/2024 at 4:01pm
Software Version: Web Wiz Forums 12.02 - http://www.webwizforums.com
Topic: <speech> tag
Posted By: edo
Subject: <speech> tag
Date Posted: 07/Oct/2010 at 12:09am
|
OK, trying again. I'm having trouble using the <speech> tag in WSR Macros; I can't seem to properly run the examples with that tag in the User's Guide that I just purchased. Here's the problem. I try to run a simple macro such as this: <?xml version="1.0" encoding="UTF-16"?> <speechMacros> <command> <listenFor>good day</listenFor> <waitFor seconds=".25"/> <setTextFeedback>good day to you too</setTextFeedback> </command> [signature] </speechMacros> Instead of the expected audio output "good day to you too", the macro returns that phrase in the WSR window. That is, somehow <speech> is being interpreted as <setTextFeedback> (Macros with <setTextFeedback> run as expected.) I've tried different command and response phrases, the response still appears in the WRS window instead of driving my speakers. I've reinstalled WSR Macros, to no effect. I'm running Vista x64 SP2. Input is via Andrea Superbeam mic. and USB-SA. Output is via RealTek mainboard sound and an external audio deck. Speech recognition is quite good. I'm doing these tests in notepad, wordpad, or MS WORD. I've shut down WSRToolkit and Macro Express. I've disabled spellcheckers and other processes that might interfere, and I don't see anything in Task Manager or Services that would interfere with WSR or WSRMacros. (Of course I may be missing something.) In trying to solve this problem I found something else weird about WSR, but I'll save that for another post. Can anyone point help me solve this problem?? - edo |
Replies:
Posted By: Gherkin
Date Posted: 07/Oct/2010 at 12:45am
|
Try this: <?xml version="1.0" encoding="UTF-16"?>
<speechMacros> <command> <listenFor>Hello</listenFor> <speak>Hello Master!</speak> </command> </Signature> </speechMacros> I believe you may have misread the manual, take another look at page 25. The speak command is to tell WSR to "say" something The setText is for displaying text.
Good Luck
|
Posted By: edo
Date Posted: 07/Oct/2010 at 2:09am
|
Phooey - I see I copied in the problem macro incorrectly in my query. The macro in question does have the correct tags: <?xml version="1.0" encoding="UTF-16"?> <speechMacros> <command> <listenFor>good day</listenFor> <waitFor seconds=".25"/> <speak>good day to you too</speak> </command> [signature] </speechMacros> This is the one that looks like it's running <setTalkFeedback>. So, as in my original question, why is <speak> running incorrectly? Sorry for adding to the confusion. - edo |
Posted By: edo
Date Posted: 07/Oct/2010 at 2:20am
| And I just rechecked with the <speak> macro - it doesn't return the response <speak>...<</speak> in the WSR command window, it returns the text between <listenFor> and </listenFor>. Thus in your example, if I say "hello," I get "hello" in the WSR window, and no "hello Master!" in audio. I created your macro in WSRMacros and signed it properly. Also tried copying your code in directly, with the same result. |
Posted By: Gherkin
Date Posted: 09/Oct/2010 at 5:21am
|
Hmmmm that is very odd indeed...
Have you gotten it to say anything at all? Ever?
If not there must be somthing set up worong somewhere. If you have heard it speak before, Im not sure what the problem could be. You see "Hello" in the text because thats the command you gave, but you should be getting a hello back in voice...
|
Posted By: edo
Date Posted: 09/Oct/2010 at 6:34pm
Not to complicate matters, here's the other interesting thing I've noticed: Suppose I output something via voice input, for example the line "Here's a sentence." Then if I immediately say "select sentence" and then say "speak", I get audio output "Here's a sentence," along with the correction pane with suggestions for "speak." If however I select a previously inputted string, or a typed-in string, saying "speak" has no effect. FWIW I can't find anywhere in Rob's Rhapsody list a command "speak." Before working with the <speak> tag but after I'd loaded WSRMacros I'd been messing around with a registry cleaner (Registry Booster by Uniblue). I'm wondering that maybe a key value was deleted in the course of registry cleanup. Haven't found anything yet, but will continue making my way through WSRMacros keys and the Registry Booster logs. <sigh> - edo |
Gherkin wrote:Posted By: Gherkin
Date Posted: 18/Oct/2010 at 1:36am
|
At this point I would try a complete uninstall and reinstall of the program and the sdk. I would think that would clear up any issues I would think. Maybe redownload the SDK just to be sure you have everything clean.
Sorry for the frustration hope you stick with it
|
Posted By: edo
Date Posted: 18/Oct/2010 at 9:46am
|
Posted By: ed
Date Posted: 08/Nov/2010 at 6:55pm
| I thought I'd close out this month-old query. A few days ago I had to considerably repair my vista x64 OS due to an unrelated problem. And now I find that the <speak> tag works fine. Unfortunately I can't tell what driver caused the original problem with WSRMacros. - edo |