Speech Recognition v2 - StartListening/StopListening

Directory.Build.props

@@ -21,14 +21,20 @@
      CS1574: XML comment has cref attribute that could not be resolved 
      CS1580: Invalid type for parameter 'parameter number' in XML comment cref attribute 
      CS1581: Invalid return type in XML comment cref attribute 
-     CS1584: XML comment has syntactically incorrect cref attribute 
+     CS1584: XML comment has syntactically incorrect cref attribute
+     CS1587: XML comment is not placed on a valid language element 
      CS1589: The syntax of a tag which referenced a file was incorrect 
      CS1590: Invalid XML include element Missing file attribute 
+     CS1591: Missing XML comment for publicly visible type or member
      CS1592: Badly formed XML in included comments file 
      CS1598: XML parser could not be loaded. The XML documentation file will not be generated. 
-     CS1658: Identifier expected; 'true' is a keyword 
+     CS1658: Identifier expected; 'true' is a keyword
+     CS1710: XML comment on 'type' has a duplicate typeparam tag for 'parameter'
+     CS1711: XML comment has a typeparam tag, but there is no type parameter by that name 
+     CS1712: Type parameter has no matching typeparam tag in the XML comment
+     CS1723: XML comment has cref attribute that refers to a type parameter
      CS1734: XML comment has a paramref tag, but there is no parameter by that name -->
-     <WarningsAsErrors>nullable,CS0419,CS1570,CS1571,CS1572,CS1573,CS1574,CS1580,CS1581,CS1584,CS1589,CS1590,CS1592,CS1598,CS1658,CS1734</WarningsAsErrors>
+     <WarningsAsErrors>nullable,CS0419,CS1570,CS1571,CS1572,CS1573,CS1574,CS1580,CS1581,CS1584,CS1587,CS1589,CS1590,CS1591,CS1592,CS1598,CS1658,CS1710,CS1711,CS1712,CS1723,CS1734</WarningsAsErrors>
   </PropertyGroup>
 
   <ItemGroup>

samples/CommunityToolkit.Maui.Sample/Pages/Essentials/SpeechToTextPage.xaml

@@ -14,62 +14,115 @@
     </ContentPage.Resources>
 
     <ScrollView>
-        <Grid
-            RowDefinitions="64, 24, 64, 24, auto, 60, 60, 60"
+        <VerticalStackLayout
+            Spacing="20"
             Padding="30,0">
 
             <Label
-                Grid.Row="0"
-                Text="SpeechToText allows the user to convert speech to text in real time"/>
+                Text="SpeechToText allows the user to convert speech to text in real time"
+                HorizontalTextAlignment="Center"/>
 
             <Label
-                Grid.Row="1"
                 Text="Locale"
                 FontAttributes="Bold"/>
 
             <Picker
-                Grid.Row="2"
                 ItemsSource="{Binding Locales}"
                 SelectedItem="{Binding CurrentLocale}"
-                ItemDisplayBinding="{Binding ., Converter={StaticResource PickerLocaleDisplayConverter}}"
-                Margin="0,0,0,20">
-            </Picker>
+                ItemDisplayBinding="{Binding ., Converter={StaticResource PickerLocaleDisplayConverter}}"/>
 
             <Label
-                Grid.Row="3"
                 Text="Language Output"
                 FontAttributes="Bold"/>
 
             <Label
-                Grid.Row="4"
                 Text="{Binding RecognitionText}"
                 FontSize="18"
                 HorizontalOptions="Center"
                 HorizontalTextAlignment="Center"
-                MinimumHeightRequest="100"
-                Margin="0,0,0,20" />
+                MinimumHeightRequest="100" />
 
             <Button
-                Grid.Row="5"
                 Text="Play"
                 Command="{Binding PlayCommand}"
-                HorizontalOptions="Center"
-                Margin="0,0,0,20"/>
+                HorizontalOptions="Center" />
 
-            <Button
-                Grid.Row="6"
-                Text="Listen"
-                Command="{Binding ListenCommand}"
-                HorizontalOptions="Center"
-                Margin="0,0,0,20"/>
+            <Border
+                StrokeThickness="2"
+                Stroke="#808080"
+                StrokeShape="RoundRectangle 8,8,8,8"
+                Padding="12">
+                <Border.Content>
 
-            <Button
-                Grid.Row="7"
-                Text="Stop Listening"
-                Command="{Binding ListenCancelCommand}"
-                HorizontalOptions="Center"/>
+                    <Grid RowDefinitions="*,60"
+                          ColumnDefinitions="*,*"
+                          RowSpacing="12"
+                          ColumnSpacing="12">
+
+                        <Button
+                            Grid.Row="0"
+                            Grid.Column="0"
+                            Text="ListenAsync"
+                            Command="{Binding ListenCommand}"
+                            HorizontalOptions="End" />
+
+                        <Button
+                            Grid.Row="0"
+                            Grid.Column="1"
+                            Text="Cancel Token"
+                            Command="{Binding ListenCancelCommand}"
+                            HorizontalOptions="Start" />
+
+                        <Label
+                            Grid.Row="1"
+                            Grid.ColumnSpan="2"
+                            Text="The `ListenAsync` API allows you to await the final speech recognition results using async/await. `ListenAsync` is cancelled via CancellationToken."
+                            HorizontalOptions="Center"
+                            HorizontalTextAlignment="Center"
+                            FontSize="12"/>
 
-        </Grid>
+                    </Grid>
+                </Border.Content>
+            </Border>
+
+            <Border
+                StrokeThickness="2"
+                Stroke="#808080"
+                StrokeShape="RoundRectangle 8,8,8,8"
+                Padding="12">
+                <Border.Content>
+                    <Grid RowDefinitions="*,60"
+                          ColumnDefinitions="*,*"
+                          RowSpacing="12"
+                          ColumnSpacing="12">
+
+                        <Button
+                            Grid.Row="0"
+                            Grid.Column="0"
+                            Text="StartListenAsync"
+                            Command="{Binding StartListenCommand}"
+                            HorizontalOptions="End" />
+
+                        <Button
+                            Grid.Row="0"
+                            Grid.Column="1"
+                            Text="StopListenAsync"
+                            Command="{Binding StopListenCommand}"
+                            HorizontalOptions="Start" />
+
+                        <Label
+                            Grid.Row="1"
+                            Grid.ColumnSpan="2"
+                            Text="The `StartListenAsync` API starts the speech-to-text service and shares the results using `RecognitionResultUpdated` event and `RecognitionResultCompleted` event."
+                            HorizontalOptions="Center"
+                            HorizontalTextAlignment="Center"
+                            FontSize="12"/>
+
+                    </Grid>
+                </Border.Content>
+            </Border>
+
+        </VerticalStackLayout>
     </ScrollView>
-    
+
 </pages:BasePage>

samples/CommunityToolkit.Maui.Sample/ViewModels/Converters/ByteArrayToImageSourceConverterViewModel.cs

@@ -1,7 +1,6 @@
 using System.Diagnostics;
 using CommunityToolkit.Mvvm.ComponentModel;
 using CommunityToolkit.Mvvm.Input;
-using Microsoft.Maui.Dispatching;
 
 namespace CommunityToolkit.Maui.Sample.ViewModels.Converters;
 

samples/CommunityToolkit.Maui.Sample/ViewModels/Essentials/SpeechToTextViewModel.cs

@@ -23,12 +23,23 @@ public partial class SpeechToTextViewModel : BaseViewModel
 	[ObservableProperty]
 	string? recognitionText = "Welcome to .NET MAUI Community Toolkit!";
 
+	[ObservableProperty, NotifyCanExecuteChangedFor(nameof(ListenCommand))]
+	bool canListenExecute = true;
+
+	[ObservableProperty, NotifyCanExecuteChangedFor(nameof(StartListenCommand))]
+	bool canStartListenExecute = true;
+
+	[ObservableProperty, NotifyCanExecuteChangedFor(nameof(StopListenCommand))]
+	bool canStopListenExecute = false;
+
 	public SpeechToTextViewModel(ITextToSpeech textToSpeech, ISpeechToText speechToText)
 	{
 		this.textToSpeech = textToSpeech;
 		this.speechToText = speechToText;
 
 		Locales.CollectionChanged += HandleLocalesCollectionChanged;
+		this.speechToText.StateChanged += HandleSpeechToTextStateChanged;
+		this.speechToText.RecognitionResultCompleted += HandleRecognitionResultCompleted;
 	}
 
 	public ObservableCollection<Locale> Locales { get; } = new();
@@ -59,9 +70,63 @@ await textToSpeech.SpeakAsync(RecognitionText ?? "Welcome to .NET MAUI Community
 		}, cancellationToken);
 	}
 
-	[RelayCommand(IncludeCancelCommand = true)]
+	[RelayCommand(IncludeCancelCommand = true, CanExecute = nameof(CanListenExecute))]
 	async Task Listen(CancellationToken cancellationToken)
 	{
+		CanStartListenExecute = false;
+
+		try
+		{
+			var isGranted = await speechToText.RequestPermissions(cancellationToken);
+			if (!isGranted)
+			{
+				await Toast.Make("Permission not granted").Show(CancellationToken.None);
+				return;
+			}
+
+			const string beginSpeakingPrompt = "Begin speaking...";
+
+			RecognitionText = beginSpeakingPrompt;
+
+			var recognitionResult = await speechToText.ListenAsync(
+												CultureInfo.GetCultureInfo(CurrentLocale?.Language ?? defaultLanguage),
+												new Progress<string>(partialText =>
+												{
+													if (RecognitionText is beginSpeakingPrompt)
+													{
+														RecognitionText = string.Empty;
+													}
+
+													RecognitionText += partialText + " ";
+												}), cancellationToken);
+
+			if (recognitionResult.IsSuccessful)
+			{
+				RecognitionText = recognitionResult.Text;
+			}
+			else
+			{
+				await Toast.Make(recognitionResult.Exception?.Message ?? "Unable to recognize speech").Show(CancellationToken.None);
+			}
+
+			if (RecognitionText is beginSpeakingPrompt)
+			{
+				RecognitionText = string.Empty;
+			}
+		}
+		finally
+		{
+			CanStartListenExecute = true;
+		}
+	}
+
+	[RelayCommand(CanExecute = nameof(CanStartListenExecute))]
+	async Task StartListen(CancellationToken cancellationToken)
+	{
+		CanListenExecute = false;
+		CanStartListenExecute = false;
+		CanStopListenExecute = true;
+
 		var isGranted = await speechToText.RequestPermissions(cancellationToken);
 		if (!isGranted)
 		{
@@ -73,33 +138,43 @@ async Task Listen(CancellationToken cancellationToken)
 
 		RecognitionText = beginSpeakingPrompt;
 
-		var recognitionResult = await speechToText.ListenAsync(
-											CultureInfo.GetCultureInfo(CurrentLocale?.Language ?? defaultLanguage),
-											new Progress<string>(partialText =>
-											{
-												if (RecognitionText is beginSpeakingPrompt)
-												{
-													RecognitionText = string.Empty;
-												}
-
-												RecognitionText += partialText + " ";
-											}), cancellationToken);
+		await speechToText.StartListenAsync(CultureInfo.GetCultureInfo(CurrentLocale?.Language ?? defaultLanguage), cancellationToken);
 
-		if (recognitionResult.IsSuccessful)
-		{
-			RecognitionText = recognitionResult.Text;
-		}
-		else
-		{
-			await Toast.Make(recognitionResult.Exception?.Message ?? "Unable to recognize speech").Show(CancellationToken.None);
-		}
+		speechToText.RecognitionResultUpdated += HandleRecognitionResultUpdated;
 
 		if (RecognitionText is beginSpeakingPrompt)
 		{
 			RecognitionText = string.Empty;
 		}
 	}
 
+	[RelayCommand(CanExecute = nameof(CanStopListenExecute))]
+	Task StopListen(CancellationToken cancellationToken)
+	{
+		CanListenExecute = true;
+		CanStartListenExecute = true;
+		CanStopListenExecute = false;
+
+		speechToText.RecognitionResultUpdated -= HandleRecognitionResultUpdated;
+
+		return speechToText.StopListenAsync(cancellationToken);
+	}
+
+	void HandleRecognitionResultUpdated(object? sender, SpeechToTextRecognitionResultUpdatedEventArgs e)
+	{
+		RecognitionText += e.RecognitionResult;
+	}
+
+	void HandleRecognitionResultCompleted(object? sender, SpeechToTextRecognitionResultCompletedEventArgs e)
+	{
+		RecognitionText = e.RecognitionResult;
+	}
+
+	async void HandleSpeechToTextStateChanged(object? sender, SpeechToTextStateChangedEventArgs e)
+	{
+		await Toast.Make($"State Changed: {e.State}").Show(CancellationToken.None);
+	}
+
 	void HandleLocalesCollectionChanged(object? sender, NotifyCollectionChangedEventArgs e)
 	{
 		OnPropertyChanged(nameof(CurrentLocale));

src/CommunityToolkit.Maui.Core/Essentials/SpeechToText/Events/SpeechToTextRecognitionResultCompletedEventArgs.cs

@@ -0,0 +1,20 @@
+namespace CommunityToolkit.Maui.Media;
+
+/// <summary>
+/// <see cref="EventArgs"/> for <see cref="ISpeechToText.RecognitionResultCompleted"/>
+/// </summary>
+public class SpeechToTextRecognitionResultCompletedEventArgs : EventArgs
+{
+	/// <summary>
+	/// Initialize a new instance of <see cref="SpeechToTextRecognitionResultCompletedEventArgs"/>
+	/// </summary>
+	public SpeechToTextRecognitionResultCompletedEventArgs(string recognitionResult)
+	{
+		RecognitionResult = recognitionResult;
+	}
+
+	/// <summary>
+	/// Speech recognition result
+	/// </summary>
+	public string RecognitionResult { get; }
+}

src/CommunityToolkit.Maui.Core/Essentials/SpeechToText/Events/SpeechToTextRecognitionResultUpdatedEventArgs.cs

@@ -0,0 +1,20 @@
+namespace CommunityToolkit.Maui.Media;
+
+/// <summary>
+/// <see cref="EventArgs"/> for <see cref="ISpeechToText.RecognitionResultUpdated"/>
+/// </summary>
+public class SpeechToTextRecognitionResultUpdatedEventArgs : EventArgs
+{
+	/// <summary>
+	/// Initialize a new instance of <see cref="SpeechToTextRecognitionResultUpdatedEventArgs"/>
+	/// </summary>
+	public SpeechToTextRecognitionResultUpdatedEventArgs(string recognitionResult)
+	{
+		RecognitionResult = recognitionResult;
+	}
+
+	/// <summary>
+	/// Speech recognition result
+	/// </summary>
+	public string RecognitionResult { get; }
+}

src/CommunityToolkit.Maui.Core/Essentials/SpeechToText/Events/SpeechToTextStateChangedEventArgs.cs

@@ -0,0 +1,20 @@
+namespace CommunityToolkit.Maui.Media;
+
+/// <summary>
+/// <see cref="EventArgs"/> for <see cref="ISpeechToText.StateChanged"/>
+/// </summary>
+public class SpeechToTextStateChangedEventArgs : EventArgs
+{
+	/// <summary>
+	/// Initialize a new instance of <see cref="SpeechToTextStateChangedEventArgs"/>
+	/// </summary>
+	public SpeechToTextStateChangedEventArgs(SpeechToTextState state)
+	{
+		State = state;
+	}
+
+	/// <summary>
+	/// Speech To Text State
+	/// </summary>
+	public SpeechToTextState State { get; }
+}