When I load the grammar and use:
SpeechRecognitionEngine _sre = new SpeechRecognitionEngine();
...
_sre.RecognizeAsync(RecognizeMode.Multiple);
And I pout or cough or say any other word (like doh!) On the microphone, I get a false positive. For instance:
If I say to the microphone: "To!" (e.g. Homer ...)
RecognizeText: "yes" Confidence: 0.8639852
About grxml ... Weight does not work as I expected for my "WakeUpWord".
Can I add trust to my grxml file?
Is there a way to prevent false triggering with SpeechRecognitionEngine in C #?
I tried the following grammar:
<?xml version="1.0" encoding="UTF-8"?>
<grammar xmlns="http://www.w3.org/2001/06/grammar" version="1.0" root="yesOrNo">
<rule id="yesOrNo">
<one-of>
<item>yes</item>
<item>no</item>
</one-of>
</rule>
</grammar>
source
share