VoiceLive SDK updates

sdk/ai/Azure.AI.VoiceLive/CHANGELOG.md

@@ -3,8 +3,15 @@
 ## 1.0.0-beta.2 (Unreleased)
 
 ### Features Added
+Added overloads for MessageItem creation to accept a single content part.
 
 ### Breaking Changes
+AudioFormat was split into InputAudioFormat and OutputAudioFormat.
+Emotion classes / options dropped.
+Eou and TurnDetection classes renamed.
+API properties that were duration based are now TimeSpans
+Methods to configure session collapsed to ConfigureSession
+Renamed ToolChoiceFunctionObjectFunction to ToolChoiceFunctionObject
 
 ### Bugs Fixed
 

sdk/ai/Azure.AI.VoiceLive/README.md

@@ -151,22 +151,22 @@ VoiceLiveSessionOptions sessionOptions = new()
     Model = model,
     Instructions = "You are a helpful AI assistant. Respond naturally and conversationally.",
     Voice = new AzureStandardVoice("en-US-AvaNeural"),
-    TurnDetection = new ServerVad()
+    TurnDetection = new AzureSemanticVadTurnDetection()
     {
         Threshold = 0.5f,
-        PrefixPaddingMs = 300,
-        SilenceDurationMs = 500
+        PrefixPadding = TimeSpan.FromMilliseconds(300),
+        SilenceDuration = TimeSpan.FromMilliseconds(500)
     },
-    InputAudioFormat = AudioFormat.Pcm16,
-    OutputAudioFormat = AudioFormat.Pcm16
+    InputAudioFormat = InputAudioFormat.Pcm16,
+    OutputAudioFormat = OutputAudioFormat.Pcm16
 };
 
 // Ensure modalities include audio
 sessionOptions.Modalities.Clear();
 sessionOptions.Modalities.Add(InputModality.Text);
 sessionOptions.Modalities.Add(InputModality.Audio);
 
-await session.ConfigureConversationSessionAsync(sessionOptions).ConfigureAwait(false);
+await session.ConfigureSessionAsync(sessionOptions).ConfigureAwait(false);
 
 // Process events from the session
 await foreach (SessionUpdate serverEvent in session.GetUpdatesAsync().ConfigureAwait(false))
@@ -196,22 +196,20 @@ VoiceLiveSessionOptions sessionOptions = new()
     {
         Temperature = 0.8f
     },
-    TurnDetection = new AzureSemanticVad()
+    TurnDetection = new AzureSemanticVadTurnDetection()
     {
-        NegThreshold = 0.3f,
-        WindowSize = 300,
         RemoveFillerWords = true
     },
-    InputAudioFormat = AudioFormat.Pcm16,
-    OutputAudioFormat = AudioFormat.Pcm16
+    InputAudioFormat = InputAudioFormat.Pcm16,
+    OutputAudioFormat = OutputAudioFormat.Pcm16
 };
 
 // Ensure modalities include audio
 sessionOptions.Modalities.Clear();
 sessionOptions.Modalities.Add(InputModality.Text);
 sessionOptions.Modalities.Add(InputModality.Audio);
 
-await session.ConfigureConversationSessionAsync(sessionOptions).ConfigureAwait(false);
+await session.ConfigureSessionAsync(sessionOptions).ConfigureAwait(false);
 ```
 
 ### Function calling example
@@ -240,8 +238,8 @@ VoiceLiveSessionOptions sessionOptions = new()
     Model = model,
     Instructions = "You are a weather assistant. Use the get_current_weather function to help users with weather information.",
     Voice = new AzureStandardVoice("en-US-AvaNeural"),
-    InputAudioFormat = AudioFormat.Pcm16,
-    OutputAudioFormat = AudioFormat.Pcm16
+    InputAudioFormat = InputAudioFormat.Pcm16,
+    OutputAudioFormat = OutputAudioFormat.Pcm16
 };
 
 // Add the function tool
@@ -252,7 +250,7 @@ sessionOptions.Modalities.Clear();
 sessionOptions.Modalities.Add(InputModality.Text);
 sessionOptions.Modalities.Add(InputModality.Audio);
 
-await session.ConfigureConversationSessionAsync(sessionOptions).ConfigureAwait(false);
+await session.ConfigureSessionAsync(sessionOptions).ConfigureAwait(false);
 ```
 
 ## Troubleshooting

sdk/ai/Azure.AI.VoiceLive/samples/BasicVoiceAssistant/BasicVoiceAssistant.cs

@@ -12,7 +12,7 @@ namespace Azure.AI.VoiceLive.Samples;
 /// This sample now demonstrates some of the new convenience methods added to the VoiceLive SDK:
 /// - ClearStreamingAudioAsync() - Clears all input audio currently being streamed
 /// - CancelResponseAsync() - Cancels the current response generation (existing method)
-/// - ConfigureConversationSessionAsync() - Configures session options (existing method)
+/// - ConfigureSessionAsync() - Configures session options (existing method)
 ///
 /// Additional convenience methods available but not shown in this sample:
 /// - StartAudioTurnAsync() / EndAudioTurnAsync() / CancelAudioTurnAsync() - Audio turn management
@@ -123,11 +123,11 @@ private async Task SetupSessionAsync(CancellationToken cancellationToken)
         var azureVoice = new AzureStandardVoice(_voice);
 
         // Create strongly typed turn detection configuration
-        var turnDetectionConfig = new ServerVad
+        var turnDetectionConfig = new ServerVadTurnDetection
         {
             Threshold = 0.5f,
-            PrefixPaddingMs = 300,
-            SilenceDurationMs = 500
+            PrefixPadding = TimeSpan.FromMilliseconds(300),
+            SilenceDuration = TimeSpan.FromMilliseconds(500)
         };
 
         // Create conversation session options
@@ -137,8 +137,8 @@ private async Task SetupSessionAsync(CancellationToken cancellationToken)
             Model = _model,
             Instructions = _instructions,
             Voice = azureVoice,
-            InputAudioFormat = AudioFormat.Pcm16,
-            OutputAudioFormat = AudioFormat.Pcm16,
+            InputAudioFormat = InputAudioFormat.Pcm16,
+            OutputAudioFormat = OutputAudioFormat.Pcm16,
             TurnDetection = turnDetectionConfig
         };
 
@@ -147,7 +147,7 @@ private async Task SetupSessionAsync(CancellationToken cancellationToken)
         sessionOptions.Modalities.Add(InputModality.Text);
         sessionOptions.Modalities.Add(InputModality.Audio);
 
-        await _session!.ConfigureConversationSessionAsync(sessionOptions, cancellationToken).ConfigureAwait(false);
+        await _session!.ConfigureSessionAsync(sessionOptions, cancellationToken).ConfigureAwait(false);
 
         _logger.LogInformation("Session configuration sent");
     }

sdk/ai/Azure.AI.VoiceLive/samples/BasicVoiceAssistant/README.md

@@ -8,7 +8,7 @@ This sample now demonstrates some of the new convenience methods added to the Vo
 
 **Used in this sample:**
 - `ClearStreamingAudioAsync()` - Clears all input audio currently being streamed
-- `ConfigureConversationSessionAsync()` - Configures conversation session options
+- `ConfigureSessionAsync()` - Configures conversation session options
 - `CancelResponseAsync()` - Cancels the current response generation
 - `SendInputAudioAsync()` - Sends audio data to the service
 

sdk/ai/Azure.AI.VoiceLive/samples/CustomerServiceBot/CustomerServiceBot.cs

@@ -126,11 +126,11 @@ private async Task SetupSessionAsync(CancellationToken cancellationToken)
         var azureVoice = new AzureStandardVoice(_voice);
 
         // Create strongly typed turn detection configuration
-        var turnDetectionConfig = new ServerVad
+        var turnDetectionConfig = new ServerVadTurnDetection
         {
             Threshold = 0.5f,
-            PrefixPaddingMs = 300,
-            SilenceDurationMs = 500
+            PrefixPadding = TimeSpan.FromMilliseconds(300),
+            SilenceDuration = TimeSpan.FromMilliseconds(500)
         };
 
         // Create conversation session options with function tools
@@ -139,8 +139,8 @@ private async Task SetupSessionAsync(CancellationToken cancellationToken)
             Model = _model,
             Instructions = _instructions,
             Voice = azureVoice,
-            InputAudioFormat = AudioFormat.Pcm16,
-            OutputAudioFormat = AudioFormat.Pcm16,
+            InputAudioFormat = InputAudioFormat.Pcm16,
+            OutputAudioFormat = OutputAudioFormat.Pcm16,
             TurnDetection = turnDetectionConfig
         };
 
@@ -157,7 +157,7 @@ private async Task SetupSessionAsync(CancellationToken cancellationToken)
         sessionOptions.Tools.Add(CreateUpdateShippingAddressTool());
 
 
-        await _session!.ConfigureConversationSessionAsync(sessionOptions, cancellationToken).ConfigureAwait(false);
+        await _session!.ConfigureSessionAsync(sessionOptions, cancellationToken).ConfigureAwait(false);
 
         _logger.LogInformation("Session configuration sent with {ToolCount} customer service tools", sessionOptions.Tools.Count);
     }

sdk/ai/Azure.AI.VoiceLive/samples/snippets/BasicUsageSnippets.cs

@@ -35,22 +35,22 @@ public async Task BasicVoiceAssistantExample()
                 Model = model,
                 Instructions = "You are a helpful AI assistant. Respond naturally and conversationally.",
                 Voice = new AzureStandardVoice("en-US-AvaNeural"),
-                TurnDetection = new ServerVad()
+                TurnDetection = new AzureSemanticVadTurnDetection()
                 {
                     Threshold = 0.5f,
-                    PrefixPaddingMs = 300,
-                    SilenceDurationMs = 500
+                    PrefixPadding = TimeSpan.FromMilliseconds(300),
+                    SilenceDuration = TimeSpan.FromMilliseconds(500)
                 },
-                InputAudioFormat = AudioFormat.Pcm16,
-                OutputAudioFormat = AudioFormat.Pcm16
+                InputAudioFormat = InputAudioFormat.Pcm16,
+                OutputAudioFormat = OutputAudioFormat.Pcm16
             };
 
             // Ensure modalities include audio
             sessionOptions.Modalities.Clear();
             sessionOptions.Modalities.Add(InputModality.Text);
             sessionOptions.Modalities.Add(InputModality.Audio);
 
-            await session.ConfigureConversationSessionAsync(sessionOptions).ConfigureAwait(false);
+            await session.ConfigureSessionAsync(sessionOptions).ConfigureAwait(false);
 
             // Process events from the session
             await foreach (SessionUpdate serverEvent in session.GetUpdatesAsync().ConfigureAwait(false))
@@ -92,22 +92,20 @@ public async Task AdvancedVoiceConfiguration()
                 {
                     Temperature = 0.8f
                 },
-                TurnDetection = new AzureSemanticVad()
+                TurnDetection = new AzureSemanticVadTurnDetection()
                 {
-                    NegThreshold = 0.3f,
-                    WindowSize = 300,
                     RemoveFillerWords = true
                 },
-                InputAudioFormat = AudioFormat.Pcm16,
-                OutputAudioFormat = AudioFormat.Pcm16
+                InputAudioFormat = InputAudioFormat.Pcm16,
+                OutputAudioFormat = OutputAudioFormat.Pcm16
             };
 
             // Ensure modalities include audio
             sessionOptions.Modalities.Clear();
             sessionOptions.Modalities.Add(InputModality.Text);
             sessionOptions.Modalities.Add(InputModality.Audio);
 
-            await session.ConfigureConversationSessionAsync(sessionOptions).ConfigureAwait(false);
+            await session.ConfigureSessionAsync(sessionOptions).ConfigureAwait(false);
             #endregion
         }
 
@@ -148,8 +146,8 @@ public async Task FunctionCallingExample()
                 Model = model,
                 Instructions = "You are a weather assistant. Use the get_current_weather function to help users with weather information.",
                 Voice = new AzureStandardVoice("en-US-AvaNeural"),
-                InputAudioFormat = AudioFormat.Pcm16,
-                OutputAudioFormat = AudioFormat.Pcm16
+                InputAudioFormat = InputAudioFormat.Pcm16,
+                OutputAudioFormat = OutputAudioFormat.Pcm16
             };
 
             // Add the function tool
@@ -160,7 +158,7 @@ public async Task FunctionCallingExample()
             sessionOptions.Modalities.Add(InputModality.Text);
             sessionOptions.Modalities.Add(InputModality.Audio);
 
-            await session.ConfigureConversationSessionAsync(sessionOptions).ConfigureAwait(false);
+            await session.ConfigureSessionAsync(sessionOptions).ConfigureAwait(false);
             #endregion
         }
     }

sdk/ai/Azure.AI.VoiceLive/src/Customizations/AnimationOptions.cs

@@ -0,0 +1,24 @@
+// Copyright (c) Microsoft Corporation. All rights reserved.
+// Licensed under the MIT License.
+
+#nullable disable
+
+using System;
+using System.Collections.Generic;
+
+namespace Azure.AI.VoiceLive
+{
+    /// <summary> Configuration for animation outputs including blendshapes, visemes, and emotion metadata. </summary>
+    public partial class AnimationOptions
+    {
+        /// <summary> Interval for emotion detection in milliseconds. If not set, emotion detection is disabled. </summary>
+        public int? EmotionDetectionIntervalMs { get; set; }
+
+        /// <summary> Interval for emotion detection. If not set, emotion detection is disabled. </summary>
+        public TimeSpan? EmotionDetectionInterval
+        {
+            get => EmotionDetectionIntervalMs.HasValue ? TimeSpan.FromMilliseconds(EmotionDetectionIntervalMs.Value) : (TimeSpan?)null;
+            set => EmotionDetectionIntervalMs = value.HasValue ? (int?)value.Value.TotalMilliseconds : null;
+        }
+    }
+}

sdk/ai/Azure.AI.VoiceLive/src/Customizations/AssistantMessageItem.cs

@@ -0,0 +1,20 @@
+// Copyright (c) Microsoft Corporation. All rights reserved.
+// Licensed under the MIT License.
+
+#nullable disable
+
+using System;
+using System.Collections.Generic;
+using System.Linq;
+
+namespace Azure.AI.VoiceLive
+{
+    /// <summary> The AssistantMessageItem. </summary>
+    public partial class AssistantMessageItem : MessageItem
+    {
+        /// <summary> Initializes a new instance of <see cref="AssistantMessageItem"/>. </summary>
+        /// <param name="content"></param>
+        /// <exception cref="ArgumentNullException"> <paramref name="content"/> is null. </exception>
+        public AssistantMessageItem(OutputTextContentPart content) : this(new[] { content }) { }
+    }
+}

sdk/ai/Azure.AI.VoiceLive/src/Customizations/AzureSemanticEouDetection.cs

@@ -0,0 +1,23 @@
+// Copyright (c) Microsoft Corporation. All rights reserved.
+// Licensed under the MIT License.
+
+#nullable disable
+
+using System;
+
+namespace Azure.AI.VoiceLive
+{
+    /// <summary> Azure semantic end-of-utterance detection (default). </summary>
+    public partial class AzureSemanticEouDetection
+    {
+        /// <summary> Gets or sets the Timeout. </summary>
+        internal float? TimeoutMs { get; set; }
+
+        /// <summary> Gets or sets the Timeout. </summary>
+        public TimeSpan Timeout
+        {
+            get => TimeSpan.FromMilliseconds(TimeoutMs ?? 0);
+            set => TimeoutMs = (float)value.TotalMilliseconds;
+        }
+    }
+}

sdk/ai/Azure.AI.VoiceLive/src/Customizations/AzureSemanticEouDetectionEn.cs

@@ -0,0 +1,23 @@
+// Copyright (c) Microsoft Corporation. All rights reserved.
+// Licensed under the MIT License.
+
+#nullable disable
+
+using System;
+
+namespace Azure.AI.VoiceLive
+{
+    /// <summary> Azure semantic end-of-utterance detection (default). </summary>
+    public partial class AzureSemanticEouDetectionEn
+    {
+        /// <summary> Gets or sets the Timeout. </summary>
+        internal float? TimeoutMs { get; set; }
+
+        /// <summary> Gets or sets the Timeout. </summary>
+        public TimeSpan Timeout
+        {
+            get => TimeSpan.FromMilliseconds(TimeoutMs ?? 0);
+            set => TimeoutMs = (float)value.TotalMilliseconds;
+        }
+    }
+}

sdk/ai/Azure.AI.VoiceLive/src/Customizations/AzureSemanticEouDetectionMultilingual.cs

@@ -0,0 +1,23 @@
+// Copyright (c) Microsoft Corporation. All rights reserved.
+// Licensed under the MIT License.
+
+#nullable disable
+
+using System;
+
+namespace Azure.AI.VoiceLive
+{
+    /// <summary> Azure semantic end-of-utterance detection (default). </summary>
+    public partial class AzureSemanticEouDetectionMultilingual
+    {
+        /// <summary> Gets or sets the Timeout. </summary>
+        internal float? TimeoutMs { get; set; }
+
+        /// <summary> Gets or sets the Timeout. </summary>
+        public TimeSpan Timeout
+        {
+            get => TimeSpan.FromMilliseconds(TimeoutMs ?? 0);
+            set => TimeoutMs = (float)value.TotalMilliseconds;
+        }
+    }
+}