[Batch Avatar] Add sample of how to enable voice sync for avatar in Batch Synthesis (#2946)

zhifzhan · web-flow · commit cc230ecaa743 · 2025-10-28T17:36:45.000+08:00
* Update batch avatar sample (C#) to include built-in voice

* Update python sample

* Merge

* Add photo avatar sample in batch C#

* Add more samples
diff --git a/samples/batch-avatar/csharp/BatchAvatarSample/BatchAvatarSample.sln b/samples/batch-avatar/csharp/BatchAvatarSample/BatchAvatarSample.sln
@@ -5,20 +5,20 @@ MinimumVisualStudioVersion = 10.0.40219.1
 Project("{9A19103F-16F7-4668-BE54-9A1E7A4F7556}") = "BatchAvatarSample", "BatchAvatarSample\BatchAvatarSample.csproj", "{CF5ED996-4313-480A-8A3D-0BB8C3D0B73A}"
 EndProject
 Global
-    GlobalSection(SolutionConfigurationPlatforms) = preSolution
-        Debug|Any CPU = Debug|Any CPU
-        Release|Any CPU = Release|Any CPU
-    EndGlobalSection
-    GlobalSection(ProjectConfigurationPlatforms) = postSolution
-        {CF5ED996-4313-480A-8A3D-0BB8C3D0B73A}.Debug|Any CPU.ActiveCfg = Debug|Any CPU
-        {CF5ED996-4313-480A-8A3D-0BB8C3D0B73A}.Debug|Any CPU.Build.0 = Debug|Any CPU
-        {CF5ED996-4313-480A-8A3D-0BB8C3D0B73A}.Release|Any CPU.ActiveCfg = Release|Any CPU
-        {CF5ED996-4313-480A-8A3D-0BB8C3D0B73A}.Release|Any CPU.Build.0 = Release|Any CPU
-    EndGlobalSection
-    GlobalSection(SolutionProperties) = preSolution
-        HideSolutionNode = FALSE
-    EndGlobalSection
-    GlobalSection(ExtensibilityGlobals) = postSolution
-                        SolutionGuid = {B0A9639E-340F-46FF-B5AE-92569CFF2A72}
-    EndGlobalSection
+	GlobalSection(SolutionConfigurationPlatforms) = preSolution
+		Debug|Any CPU = Debug|Any CPU
+		Release|Any CPU = Release|Any CPU
+	EndGlobalSection
+	GlobalSection(ProjectConfigurationPlatforms) = postSolution
+		{CF5ED996-4313-480A-8A3D-0BB8C3D0B73A}.Debug|Any CPU.ActiveCfg = Debug|Any CPU
+		{CF5ED996-4313-480A-8A3D-0BB8C3D0B73A}.Debug|Any CPU.Build.0 = Debug|Any CPU
+		{CF5ED996-4313-480A-8A3D-0BB8C3D0B73A}.Release|Any CPU.ActiveCfg = Release|Any CPU
+		{CF5ED996-4313-480A-8A3D-0BB8C3D0B73A}.Release|Any CPU.Build.0 = Release|Any CPU
+	EndGlobalSection
+	GlobalSection(SolutionProperties) = preSolution
+		HideSolutionNode = FALSE
+	EndGlobalSection
+	GlobalSection(ExtensibilityGlobals) = postSolution
+		                        SolutionGuid = {B0A9639E-340F-46FF-B5AE-92569CFF2A72}
+	EndGlobalSection
 EndGlobal
diff --git a/samples/batch-avatar/csharp/BatchAvatarSample/BatchAvatarSample/Program.cs b/samples/batch-avatar/csharp/BatchAvatarSample/BatchAvatarSample/Program.cs
@@ -73,6 +73,9 @@ private static async Task Main(string[] args)
             Console.WriteLine("Summary file can be downloaded from:");
             Console.WriteLine(job.Outputs!.Summary);
 
+            Console.WriteLine("Press Enter to delete the job and exit.");
+            var input = Console.ReadLine();
+
             //
             // Delete a job
             //
@@ -97,8 +100,7 @@ private static async Task Main(string[] args)
 
     private static async Task<BatchAvatarJob> CreateBatchAvatarJob(HttpClient httpClient, string jobUri)
     {
-        // To use SSML as input, please refer to RequestExamples.SsmlRequest
-        // To use your custom neural voice, please refer to RequestExamples.CustomVoiceRequest
+        // Please refer to RequestExamples.cs for more request examples.
         var requestBody = new BatchAvatarRequest
         {
             InputKind = "PlainText",
@@ -117,11 +119,12 @@ private static async Task<BatchAvatarJob> CreateBatchAvatarJob(HttpClient httpCl
             {
                 TalkingAvatarCharacter = "lisa",            // Avatar character
                 TalkingAvatarStyle = "graceful-sitting",    // Avatar style, required for prebuilt avatar, optional for custom avatar
+                Customized = false,                         // Set to true if you want to use custom avatar
                 VideoFormat = "mp4",                        // mp4 or webm, webm is required for transparent background
                 VideoCodec = "h264",                        // hevc, h264 or vp9, vp9 is required for transparent background; default is hevc
                 SubtitleType = "soft_embedded",
                 BackgroundColor = "#FFFFFFFF",              // background color in RGBA format, default is white; can be set to 'transparent' for transparent background
-                Customized = false,                         // Set to true if you want to use custom avatar
+                UseBuiltInVoice = false,                    // Set to true to use voice sync for avatar for custom avatar
             },
         };
 
diff --git a/samples/batch-avatar/csharp/BatchAvatarSample/BatchAvatarSample/RequestExamples.cs b/samples/batch-avatar/csharp/BatchAvatarSample/BatchAvatarSample/RequestExamples.cs
@@ -49,7 +49,7 @@ internal static class RequestExamples
         // Invalid voice name or deployment ID will be rejected.
         CustomVoices = new Dictionary<string, Guid>
         {
-            ["YOUR_CUSTOM_VOICE_NAME"]  = Guid.Parse("YOUR_CUSTOM_VOICE_DEPLOYMENT_ID"),
+            ["YOUR_CUSTOM_VOICE_NAME"]  = Guid.Parse("12345678-0000-0000-0000-000000000000"),
         },
         SynthesisConfig = new()
         {
@@ -65,4 +65,71 @@ internal static class RequestExamples
             BackgroundColor = "#FFFFFFFF",
         },
     };
+
+    public static BatchAvatarRequest VoiceSyncForAvatar = new()
+    {
+        InputKind = "PlainText",
+        Inputs =
+            [
+                new BatchAvatarInput
+                {
+                    Content = "Hi, I'm a virtual assistant created by Microsoft.",
+                },
+            ],
+        AvatarConfig = new()
+        {
+            TalkingAvatarCharacter = "my-custom-avatar",
+            Customized = true,
+            UseBuiltInVoice = true,
+            VideoFormat = "mp4",
+            VideoCodec = "h264",
+        },
+    };
+
+    public static BatchAvatarRequest PublicPhotoAvatar = new()
+    {
+        InputKind = "PlainText",
+        Inputs =
+            [
+                new BatchAvatarInput
+                {
+                    Content = "Hi, I'm a virtual assistant created by Microsoft.",
+                },
+            ],
+        SynthesisConfig = new()
+        {
+            Voice = "en-US-AndrewNeural",
+        },
+        AvatarConfig = new()
+        {
+            TalkingAvatarCharacter = "Matteo",
+            PhotoAvatarBaseModel = "vasa-1",
+            VideoFormat = "mp4",
+            VideoCodec = "h264",
+        },
+    };
+
+    public static BatchAvatarRequest CustomPhotoAvatar = new()
+    {
+        InputKind = "PlainText",
+        Inputs =
+            [
+                new BatchAvatarInput
+                {
+                    Content = "Hi, I'm a virtual assistant created by Microsoft.",
+                },
+            ],
+        SynthesisConfig = new()
+        {
+            Voice = "en-US-AndrewNeural",
+        },
+        AvatarConfig = new()
+        {
+            TalkingAvatarCharacter = "my-photo-avatar",
+            Customized = true,
+            PhotoAvatarBaseModel = "vasa-1",
+            VideoFormat = "mp4",
+            VideoCodec = "h264",
+        },
+    };
 }
diff --git a/samples/batch-avatar/csharp/BatchAvatarSample/BatchAvatarSample/dto/BatchAvatarJob.cs b/samples/batch-avatar/csharp/BatchAvatarSample/BatchAvatarSample/dto/BatchAvatarJob.cs
@@ -21,7 +21,7 @@ public class BatchAvatarRequest
 
     public BatchSynthesisProperties? Properties { get; set; }
 
-    public required BatchAvatarconfig AvatarConfig { get; set; }
+    public required BatchAvatarConfig AvatarConfig { get; set; }
 }
 
 public class BatchAvatarJob
@@ -42,7 +42,7 @@ public class BatchAvatarJob
 
     public BatchSynthesisProperties? Properties { get; set; }
 
-    public required BatchAvatarconfig AvatarConfig { get; set; }
+    public required BatchAvatarConfig AvatarConfig { get; set; }
 
     public BatchSynthesisOutputs? Outputs { get; set; }
 }
@@ -62,23 +62,23 @@ public class BatchSynthesisProperties
 
     public string? DestinationPath { get; set; }
 
-    public int? sizeInBytes { get; set; }
+    public int? SizeInBytes { get; set; }
 
-    public int? succeededCount { get; set; }
+    public int? SucceededCount { get; set; }
 
-    public int? failedCount { get; set; }
+    public int? FailedCount { get; set; }
 
-    public int? durationInMilliseconds { get; set; }
+    public int? DurationInMilliseconds { get; set; }
 
-    public Billingdetails? billingDetails { get; set; }
+    public Billingdetails? BillingDetails { get; set; }
 
     public Error? Error { get; set; }
 }
 
 public class Billingdetails
 {
-    public int neuralCharacters { get; set; }
-    public int talkingAvatarDurationSeconds { get; set; }
+    public int NeuralCharacters { get; set; }
+    public int TalkingAvatarDurationSeconds { get; set; }
 }
 
 public class BatchSynthesisconfig
@@ -94,7 +94,7 @@ public class BatchSynthesisconfig
     public string? Volume { get; set; }
 }
 
-public class BatchAvatarconfig
+public class BatchAvatarConfig
 {
     public required string TalkingAvatarCharacter { get; set; }
 
@@ -119,6 +119,10 @@ public class BatchAvatarconfig
     public int? BitrateKbps { get; set; }
 
     public bool? Customized { get; set; }
+
+    public bool? UseBuiltInVoice { get; set; }
+
+    public string? PhotoAvatarBaseModel { get; set; }
 }
 
 public class Videocrop
@@ -143,7 +147,7 @@ public class BillingDetails
 
     public long CustomNeuralCharacters { get; set; }
 
-    public long talkingAvatarDurationSeconds { get; set; }
+    public long TalkingAvatarDurationSeconds { get; set; }
 
     public long AoaiCharacters { get; set; }
 
diff --git a/samples/batch-avatar/python/synthesis.py b/samples/batch-avatar/python/synthesis.py
@@ -89,9 +89,11 @@ def submit_synthesis(job_id: str):
             "videoCodec": "h264",  # hevc, h264 or vp9, vp9 is required for transparent background; default is hevc
             "subtitleType": "soft_embedded",
             "backgroundColor": "#FFFFFFFF",  # background color in RGBA format, default is white;
-                                             # can be set to 'transparent' for transparent background
+                                               # can be set to 'transparent' for transparent background
             # "backgroundImage": "https://samples-files.com/samples/Images/jpg/1920-1080-sample.jpg",
             # background image URL, only support https, either backgroundImage or backgroundColor can be set
+            "useBuiltInVoice": False,  # whether to use voice sync of custom avatar.
+                                       # Only valid if customized is True and the custom avatar is trained with voice sync enabled.
         }
     }
 

Original file line number	Diff line number	Diff line change
`@@ -89,9 +89,11 @@ def submit_synthesis(job_id: str):`
`89`	`89`	`"videoCodec": "h264", # hevc, h264 or vp9, vp9 is required for transparent background; default is hevc`
`90`	`90`	`"subtitleType": "soft_embedded",`
`91`	`91`	`"backgroundColor": "#FFFFFFFF", # background color in RGBA format, default is white;`
`92`		`- # can be set to 'transparent' for transparent background`
	`92`	`+ # can be set to 'transparent' for transparent background`
`93`	`93`	`# "backgroundImage": "https://samples-files.com/samples/Images/jpg/1920-1080-sample.jpg",`
`94`	`94`	`# background image URL, only support https, either backgroundImage or backgroundColor can be set`
	`95`	`+ "useBuiltInVoice": False, # whether to use voice sync of custom avatar.`
	`96`	`+ # Only valid if customized is True and the custom avatar is trained with voice sync enabled.`
`95`	`97`	`}`
`96`	`98`	`}`
`97`	`99`