using System.Speech
using System.Speech.Recognition
using System.Speech.AudioFormat

SpeechRecognitionEngine sre = new SpeechRecognitionEngine();

sre.SpeechHypothesized += new EventHandler<SpeechHypothesizedEventArgs>(sre_SpeechHypothesized);
sre.SpeechRecognized += new EventHandler<SpeechRecognizedEventArgs>(sre_SpeechRecognized);
sre.RecognizeCompleted += new EventHandler<RecognizeCompletedEventArgs>(sre_RecognizeCompleted);
sre.AudioSignalProblemOccurred += new EventHandler<AudioSignalProblemOccurredEventArgs>(sre_AudioSignalProblemOccurred);
sre.SpeechDetected += new EventHandler<SpeechDetectedEventArgs>(sre_SpeechDetected);
sre.SpeechRecognitionRejected += new EventHandler<SpeechRecognitionRejectedEventArgs>(sre_SpeechRecognitionRejected);

string[] words = {"power", "on", "off"};
Choices choices = new Choices(words);
GrammarBuilder gb = new GrammarBuilder(choices);
Grammar grammar = new Grammar(gb);
sre.LoadGrammar(grammar);

DictationGrammar dg = new DictationGrammar();
sre.LoadGrammar(dg);

sre.SetInputToDefaultAudioDevice();

sre.SetInputToWaveFile("c:/audio.wav");

sre.EndSilenceTimeout = new TimeSpan(0, 0, 2);

sre.Recognize();

sre.RecognizeAsync();

void sre_AudioSignalProblemOccurred(object sender, AudioSignalProblemOccurredEventArgs e)
{
	Debug.Print(e.AudioSignalProblem.ToString());
}

void sre_SpeechHypothesized(object sender, SpeechHypothesizedEventArgs e)
{
	Debug.Print(e.Result.Text);
}

void sre_SpeechRecognitionRejected(object sender, SpeechRecognitionRejectedEventArgs e)
{
	Debug.Print("Rejected!");
}

void sre_RecognizeCompleted(object sender, RecognizeCompletedEventArgs e)
{
	Debug.Print("Recognition Complete!");
}

void sre_SpeechRecognized(object sender, SpeechRecognizedEventArgs e)
{
	Debug.Print(e.Result.Text);
}

void sre_SpeechDetected(object sender, SpeechDetectedEventArgs e)
{
	Debug.Print("Speech Detected!");
}

10 Responses to “Getting Started With Windows Voice Recognition”

1. Divey says:

   January 7, 2011 at 5:54 am

   Hi,
   This webpage has given me a great start with my Speech Recognition project. I just had one question, once i entered the code above in Visual c# 2010 and tried to compile it, i am getting errors like
   “Error 1 An object reference is required for the non-static field, method, or property ‘Hello.sre_SpeechHypothesized(object, System.Speech.Recognition.SpeechHypothesizedEventArgs)’ C:\Users\Owner\Documents\Visual Studio 2010\Projects\Learning\Learning\Class1.cs 16 35 Learning”

   This is just one of the errors which i am showing, i am not sure how to solve this problem. Can you please provide me some assistance with this.

   Thanks
   Divey

   Reply
   • Tom Sherman says:

     January 9, 2011 at 10:18 pm

     Tough to know without looking at your actual code file, but the error is that you are missing a reference. This basically means that you are calling something, but the program doesn’t have a definition for it. In your example many things may not be defined, I can’t know what one.

     1. Hello isn’t defined
     2. sre_SpeechHypothesized is not defined
     3. You are missing the ‘using’ references

     Reply
2. Divey says:

   January 10, 2011 at 12:00 am

   My code right is exactly the same as the example that has been given above. I just need to make it work first then i will implement my own application with it. I am pasting the code with this, please try to help me, i have tried alot to fix the errors, but it is not working.

   using System;
   using System.IO;
   using System.Speech;
   using System.Speech.Recognition;
   using System.Speech.AudioFormat;
   
   public class Hello
   {
   
       public static void Main()
       {
   
           SpeechRecognitionEngine sre = new SpeechRecognitionEngine();
           sre.SpeechHypothesized += new EventHandler<SpeechHypothesizedEventArgs>(sre_SpeechHypothesized);
           sre.SpeechRecognized += new EventHandler<SpeechRecognizedEventArgs>(sre_SpeechRecognized);
           sre.RecognizeCompleted += new EventHandler<RecognizeCompletedEventArgs>(sre_RecognizeCompleted);
           sre.AudioSignalProblemOccurred += new EventHandler<AudioSignalProblemOccurredEventArgs>(sre_AudioSignalProblemOccurred);
           sre.SpeechDetected += new EventHandler<SpeechDetectedEventArgs>(sre_SpeechDetected);
           sre.SpeechRecognitionRejected += new EventHandler<SpeechRecognitionRejectedEventArgs>(sre_SpeechRecognitionRejected);
   
           DictationGrammar dg = new DictationGrammar();
           sre.LoadGrammar(dg);
   
           sre.SetInputToWaveFile("hello.wav");
           sre.EndSilenceTimeout = new TimeSpan(0, 0, 2);
   
           sre.Recognize();
       }
   
       void sre_SpeechHypothesized(object sender, SpeechHypothesizedEventArgs e)
       {
           Console.WriteLine(e.Result.Text);
       }
       void sre_AudioSignalProblemOccurred(object sender, AudioSignalProblemOccurredEventArgs e)
       {
           Console.WriteLine(e.AudioSignalProblem.ToString());
       }
   
   void sre_SpeechRecognitionRejected(object sender, SpeechRecognitionRejectedEventArgs e)
   {
       Console.WriteLine("Rejected!");
   }
   
   void sre_RecognizeCompleted(object sender, RecognizeCompletedEventArgs e)
   {
       Console.WriteLine("Recognition Complete!");
   }
   
   void sre_SpeechRecognized(object sender, SpeechRecognizedEventArgs e)
   {
       Console.WriteLine(e.Result.Text);
   }
   
   void sre_SpeechDetected(object sender, SpeechDetectedEventArgs e)
   {
       Console.WriteLine("Speech Detected!");
   }
   
       }
   

   Reply
3. Tom Sherman says:

   January 10, 2011 at 3:12 am

   Edit Note: Looks like cut and paste of EventArg casts didn’t make it through WordPress. In my initial response I thought that he had failed to cut and paste the proper event handler code, but that doesn’t appear to be the case. I am going to go back and fix up his initial post so that nobody else gets confused copy and pasting it.

   You need to decide how you want your program structured. You would benefit from reading some web articles on constructing a program. In this case you have declared all your methods in Main which is a static entry point. Thus all the methods have to be static. You could add “static” in front of each of the methods and get your program to compile, but this may not be what you want. Instead you may want to create an instance of your class and have it call an entry point method. Then you wouldn’t have to add static before each of the methods. Read up on static and and classes too. For example:

   public static void Main() {
        Hello hello = new Hello();
        hello.doStuff();
   }
   
   private void doStuff() {
      //all the stuff from my post
      SpeechRecognitionEngine sre = new SpeechRecognitionEngine();
      //...
   }
   

   Reply
4. Divey says:

   January 10, 2011 at 4:34 am

   Thanks for that help, now the errors are finally gone. What i need to know is how can i tell if the recognition is being done, like i don’t know how to set an output text document, where the audio will be written as text.

   Reply
   • Tom Sherman says:

     January 10, 2011 at 9:50 am

     If you used the methods I provided the output will be written to the console in visual studio. Just google how to bring up the console. If you want to write to a file then take a look into fileinfo and filestream. I would also recommend that you sign up for a stackoverflow account. Here you can ask questions and get answers very quickly. They have rules about how quickly you can post as a new user, but it is a fantastic place for programmers of all skill levels.

     Reply
5. Divey says:

   January 11, 2011 at 5:20 am

   Hey, thanks a lot for the help and i will definitely look into getting a stackoverflow account, (i think i have one of these). But, i am still stuck with the code that i have right now. For some reason, when i run the code, it crashes and gives an error saying the input file is not found. I have put my audio file in the correct destination path, but still its not recognising the input. I would greatly appreciate some more help with this topic.
   Thanks
   Divey

   Reply
   • Tom Sherman says:

     January 23, 2011 at 11:30 am

     Your path must be formatted incorrectly then. I would double check that everything is correct.

     Reply
6. Shoaib says:

   March 11, 2012 at 6:35 am

   Hello and thank you for taking time to publish this very helpful document. It is the best and most helpful document I have ever seen on speech recognition using C#. I have implemented the code, however, it is not giving me accurate results. I have tried the code with different audio files. Can you please give me any tips for increasing accuracy? Thank you again!

   Reply
7. crisppy says:

   March 29, 2012 at 11:23 am

   oh this is great..

   Reply