Add GenerateSpeechStreamingAsync() for streaming TTS responses.

global.json

@@ -1,6 +1,6 @@
 {
   "sdk": {
-    "version": "8.0.100",
+    //"version": "8.0.100",
     "rollForward": "feature"
   }
 }

src/Custom/Audio/AudioClient.cs

@@ -1,7 +1,10 @@
 using System;
+using System.Buffers;
 using System.ClientModel;
 using System.ClientModel.Primitives;
+using System.Collections.Generic;
 using System.IO;
+using System.Runtime.CompilerServices;
 using System.Threading;
 using System.Threading.Tasks;
 
@@ -108,9 +111,47 @@ public virtual async Task<ClientResult<BinaryData>> GenerateSpeechAsync(string t
 
         using BinaryContent content = options;
         ClientResult result = await GenerateSpeechAsync(content, cancellationToken.ToRequestOptions()).ConfigureAwait(false);
+        await result.GetRawResponse().BufferContentAsync(cancellationToken).ConfigureAwait(false);
         return ClientResult.FromValue(result.GetRawResponse().Content, result.GetRawResponse());
     }
 
+    /// <summary> Generates a life-like, spoken audio recording of the input text. </summary>
+    /// <remarks>
+    ///     The default format of the generated audio is <see cref="GeneratedSpeechFormat.Mp3"/> unless otherwise specified
+    ///     via <see cref="SpeechGenerationOptions.ResponseFormat"/>.
+    /// </remarks>
+    /// <param name="text"> The text to generate audio for. </param>
+    /// <param name="voice"> The voice to use in the generated audio. </param>
+    /// <param name="options"> The options to configure the audio generation. </param>
+    /// <param name="cancellationToken"> A token that can be used to cancel this method call. </param>
+    /// <exception cref="ArgumentNullException"> <paramref name="text"/> is null. </exception>
+    /// <returns> Streaming chunks of the generated audio in the specified output format. </returns>
+    public virtual async IAsyncEnumerable<BinaryData> GenerateSpeechStreamingAsync(string text, GeneratedSpeechVoice voice, SpeechGenerationOptions options = null, [EnumeratorCancellation] CancellationToken cancellationToken = default)
+    {
+        Argument.AssertNotNull(text, nameof(text));
+
+        options ??= new();
+        CreateSpeechGenerationOptions(text, voice, ref options);
+
+        using BinaryContent content = options;
+        ClientResult result = await GenerateSpeechAsync(content, cancellationToken.ToRequestOptions()).ConfigureAwait(false);
+        var stream = result.GetRawResponse().ContentStream;
+
+        var buffer = ArrayPool<byte>.Shared.Rent(1024 * 16);
+        try
+        {
+            int bytesRead;
+            while ((bytesRead = await stream.ReadAsync(buffer, 0, buffer.Length, cancellationToken).ConfigureAwait(false)) != 0)
+            {
+                yield return BinaryData.FromBytes(buffer.AsMemory(0, bytesRead));
+            }
+        }
+        finally
+        {
+            ArrayPool<byte>.Shared.Return(buffer);
+        }
+    }
+
     /// <summary> Generates a life-like, spoken audio recording of the input text. </summary>
     /// <remarks>
     ///     The default format of the generated audio is <see cref="GeneratedSpeechFormat.Mp3"/> unless otherwise specified
@@ -130,7 +171,8 @@ public virtual ClientResult<BinaryData> GenerateSpeech(string text, GeneratedSpe
         CreateSpeechGenerationOptions(text, voice, ref options);
 
         using BinaryContent content = options;
-        ClientResult result = GenerateSpeech(content, cancellationToken.ToRequestOptions()); ;
+        ClientResult result = GenerateSpeech(content, cancellationToken.ToRequestOptions());
+        result.GetRawResponse().BufferContent(cancellationToken);
         return ClientResult.FromValue(result.GetRawResponse().Content, result.GetRawResponse());
     }
 

src/Custom/Audio/SpeechGenerationOptions.cs

@@ -37,4 +37,11 @@ public SpeechGenerationOptions()
     [CodeGenMember("Speed")]
 
     public float? SpeedRatio { get; set; }
+
+    // CUSTOM:
+    /// <summary>
+    ///     Control the voice of your generated audio with additional instructions. Does not work with tts-1 or tts-1-hd.
+    /// </summary>
+    [CodeGenMember("Instructions")]
+    public string Instructions { get; set; }
 }

src/Generated/AudioClient.RestClient.cs

@@ -18,6 +18,7 @@ internal virtual PipelineMessage CreateCreateSpeechRequest(BinaryContent content
         {
             PipelineMessage message = Pipeline.CreateMessage();
             message.ResponseClassifier = PipelineMessageClassifier200;
+            message.BufferResponse = false;
             PipelineRequest request = message.Request;
             request.Method = "POST";
             ClientUriBuilder uri = new ClientUriBuilder();

src/Generated/Models/SpeechGenerationOptions.Serialization.cs

@@ -52,6 +52,11 @@ protected virtual void JsonModelWriteCore(Utf8JsonWriter writer, ModelReaderWrit
                 writer.WritePropertyName("speed"u8);
                 writer.WriteNumberValue(SpeedRatio.Value);
             }
+            if (Optional.IsDefined(Instructions) && _additionalBinaryDataProperties?.ContainsKey("instructions") != true)
+            {
+                writer.WritePropertyName("instructions"u8);
+                writer.WriteStringValue(Instructions);
+            }
             if (_additionalBinaryDataProperties != null)
             {
                 foreach (var item in _additionalBinaryDataProperties)
@@ -97,6 +102,7 @@ internal static SpeechGenerationOptions DeserializeSpeechGenerationOptions(JsonE
             string input = default;
             GeneratedSpeechVoice voice = default;
             float? speedRatio = default;
+            string instructions = default;
             IDictionary<string, BinaryData> additionalBinaryDataProperties = new ChangeTrackingDictionary<string, BinaryData>();
             foreach (var prop in element.EnumerateObject())
             {
@@ -133,6 +139,11 @@ internal static SpeechGenerationOptions DeserializeSpeechGenerationOptions(JsonE
                     speedRatio = prop.Value.GetSingle();
                     continue;
                 }
+                if (prop.NameEquals("instructions"u8))
+                {
+                    instructions = prop.Value.GetString();
+                    continue;
+                }
                 additionalBinaryDataProperties.Add(prop.Name, BinaryData.FromString(prop.Value.GetRawText()));
             }
             return new SpeechGenerationOptions(
@@ -141,6 +152,7 @@ internal static SpeechGenerationOptions DeserializeSpeechGenerationOptions(JsonE
                 input,
                 voice,
                 speedRatio,
+                instructions,
                 additionalBinaryDataProperties);
         }
 

src/Generated/Models/SpeechGenerationOptions.cs

@@ -11,13 +11,14 @@ public partial class SpeechGenerationOptions
     {
         private protected IDictionary<string, BinaryData> _additionalBinaryDataProperties;
 
-        internal SpeechGenerationOptions(GeneratedSpeechFormat? responseFormat, InternalCreateSpeechRequestModel model, string input, GeneratedSpeechVoice voice, float? speedRatio, IDictionary<string, BinaryData> additionalBinaryDataProperties)
+        internal SpeechGenerationOptions(GeneratedSpeechFormat? responseFormat, InternalCreateSpeechRequestModel model, string input, GeneratedSpeechVoice voice, float? speedRatio, string instructions, IDictionary<string, BinaryData> additionalBinaryDataProperties)
         {
             ResponseFormat = responseFormat;
             Model = model;
             Input = input;
             Voice = voice;
             SpeedRatio = speedRatio;
+            Instructions = instructions;
             _additionalBinaryDataProperties = additionalBinaryDataProperties;
         }
 

src/OpenAI.csproj

@@ -11,7 +11,7 @@
     <VersionPrefix>2.2.0</VersionPrefix>
     <VersionSuffix>beta.4</VersionSuffix>
 
-    <TargetFrameworks>net8.0;net6.0;netstandard2.0</TargetFrameworks>
+    <TargetFrameworks>net9.0;net8.0;net6.0;netstandard2.0</TargetFrameworks>
     <LangVersion>latest</LangVersion>
 
     <!-- Generate an XML documentation file for the project. -->