1

Few questions,

  1. What is the difference between the SpeechRecognizer and the SpeechRecognitionEngine classes? why use one over the other for speech recognition?

  2. Is the speech recognition widget that I see in Windows 10 when I start my program has to be shown?

  3. I loaded the SpeechRecognizer object with simple grammar such as "a", "b", "a r". it recognizes it perfectly but the time it takes is not ideal for my program, I would like it to be faster, any way to do that?

mountSin
  • 31
  • 4
  • Have you read documentations of those classes especially remarks sections? There is information to answer your 1. question. For 2. try and you will see. 3. Buy a better pc? – Renatas M. Nov 06 '18 at 14:12
  • 1. I will, thanks. 2. What do you mean try and see? the widget opens everytime I start my program 3. I have a macbookpro i5, I don't think that's the problem – mountSin Nov 06 '18 at 14:20
  • 2. I don't understand what do you mean by saying is it has to be shown if it is shown when you start. Maybe you want to know can you control or switch it off? 3. Then what is slow and fast in thins case? – Renatas M. Nov 06 '18 at 14:27
  • 2. I'm talking about this widget: https://imgur.com/a/3IR2QqS , is there a way Windows can do the speech recognition in the background and not show it to me? 3. How exactly do I measure? I think it's taking approx 0.1 seconds to recoginze and I want it to take less – mountSin Nov 06 '18 at 15:06
  • 2. I guess you need to use SpeechRecognizerEngine to be able to control this. 3. `I want` is not a measurement unit, you need to do speed comparison. Maybe 0.1s for your grammar is blazing fast. Also it might depend on million reasons. Do you run it in debug mode or is it release build? Have you compared speed with other grammars and applications? – Renatas M. Nov 06 '18 at 15:32
  • In the future, please only ask [one question per question](https://meta.stackexchange.com/questions/222735/can-i-ask-only-one-question-per-post). It makes things easier on everybody. –  Nov 07 '18 at 22:21

1 Answers1

1

I think this has been answered in the past. See Using System.Speech.Recognition opens Windows Speech Recognition, does this help?

In general, you can use System.Speech as inproc or shared. When shared, you see a recognizer "widget" on the screen. If you use an inproc recognizer, you control the recognizer and windows does not add a UI. See good Speech recognition API for some more background.

Michael Levy
  • 13,097
  • 15
  • 66
  • 100