SpeechRecognizer Class

Definition

public ref class SpeechRecognizer : IDisposable

public class SpeechRecognizer : IDisposable

type SpeechRecognizer = class
    interface IDisposable

Public Class SpeechRecognizer
Implements IDisposable

Inheritance

Object

SpeechRecognizer

Implements

IDisposable

Examples

The following example is part of a console application that loads a speech recognition grammar and demonstrates asynchronous emulated input, the associated recognition results, and the associated events raised by the speech recognizer. If Windows Speech Recognition is not running, then starting this application will also start Windows Speech Recognition. If Windows Speech Recognition is in the Sleeping state, then EmulateRecognizeAsync always returns null.

using System;  
using System.Speech.Recognition;  
using System.Threading;  

namespace SharedRecognizer  
{  
  class Program  
  {  

    // Indicate whether the asynchronous emulate recognition  
    // operation has completed.  
    static bool completed;  

    static void Main(string[] args)  
    {  

      // Initialize an instance of the shared recognizer.  
      using (SpeechRecognizer recognizer = new SpeechRecognizer())  
      {  

        // Create and load a sample grammar.  
        Grammar testGrammar =  
          new Grammar(new GrammarBuilder("testing testing"));  
        testGrammar.Name = "Test Grammar";  
        recognizer.LoadGrammar(testGrammar);  

        // Attach event handlers for recognition events.  
        recognizer.SpeechRecognized +=  
          new EventHandler<SpeechRecognizedEventArgs>(  
            SpeechRecognizedHandler);  
        recognizer.EmulateRecognizeCompleted +=  
          new EventHandler<EmulateRecognizeCompletedEventArgs>(  
            EmulateRecognizeCompletedHandler);  

        completed = false;  

        // Start asynchronous emulated recognition.   
        // This matches the grammar and generates a SpeechRecognized event.  
        recognizer.EmulateRecognizeAsync("testing testing");  

        // Wait for the asynchronous operation to complete.  
        while (!completed)  
        {  
          Thread.Sleep(333);  
        }  

        completed = false;  

        // Start asynchronous emulated recognition.  
        // This does not match the grammar or generate a SpeechRecognized event.  
        recognizer.EmulateRecognizeAsync("testing one two three");  

        // Wait for the asynchronous operation to complete.  
        while (!completed)  
        {  
          Thread.Sleep(333);  
        }  
      }  

      Console.WriteLine();  
      Console.WriteLine("Press any key to exit...");  
      Console.ReadKey();  
    }  

    // Handle the SpeechRecognized event.  
    static void SpeechRecognizedHandler(  
      object sender, SpeechRecognizedEventArgs e)  
    {  
      if (e.Result != null)  
      {  
        Console.WriteLine("Recognition result = {0}",  
          e.Result.Text ?? "<no text>");  
      }  
      else  
      {  
        Console.WriteLine("No recognition result");  
      }  
    }  

    // Handle the SpeechRecognizeCompleted event.  
    static void EmulateRecognizeCompletedHandler(  
      object sender, EmulateRecognizeCompletedEventArgs e)  
    {  
      if (e.Result == null)  
      {  
        Console.WriteLine("No result generated.");  
      }  

      // Indicate the asynchronous operation is complete.  
      completed = true;  
    }  
  }  
}  

Remarks

Applications use the shared recognizer to access Windows Speech Recognition. Use the SpeechRecognizer object to add to the Windows speech user experience.

This class provides control over various aspects of the speech recognition process:

• To manage speech recognition grammars, use the LoadGrammar, LoadGrammarAsync, UnloadGrammar, UnloadAllGrammars, and Grammars.

• To get information about current speech recognition operations, subscribe to the SpeechRecognizer's SpeechDetected, SpeechHypothesized, SpeechRecognitionRejected, and SpeechRecognized events.

• To view or modify the number of alternate results the recognizer returns, use the MaxAlternates property. The recognizer returns recognition results in a RecognitionResult object.

• To access or monitor the state of the shared recognizer, use the AudioLevel, AudioPosition, AudioState, Enabled, PauseRecognizerOnRecognition, RecognizerAudioPosition, and State properties and the AudioLevelUpdated, AudioSignalProblemOccurred, AudioStateChanged, and StateChanged events.

• To synchronize changes to the recognizer, use the RequestRecognizerUpdate method. The shared recognizer uses more than one thread to perform tasks.

• To emulate input to the shared recognizer, use the EmulateRecognize and EmulateRecognizeAsync methods.

The configuration of Windows Speech Recognition is managed by the use of the Speech Properties dialog in the Control Panel. This interface is used to select the default desktop speech recognition engine and language, the audio input device, and the sleep behavior of speech recognition. If the configuration of Windows Speech Recognition is changed while the application is running, (for instance, if speech recognition is disabled or the input language is changed), the change affects all SpeechRecognizer objects.

To create an in-process speech recognizer that is independent of Windows Speech Recognition, use the SpeechRecognitionEngine class.

Constructors

Properties

Methods

Events

Applies to

See also

• SpeechRecognitionEngine
• Grammar
• RecognitionResult