Add C# API for Kokoro TTS 1.0 (#1805)

k2-fsa · Feb 7, 2025 · ae32dfa · ae32dfa
1 parent e1a88a7
commit ae32dfa
Show file tree

Hide file tree

Showing 4 changed files with 74 additions and 2 deletions.
diff --git a/.github/scripts/test-dot-net.sh b/.github/scripts/test-dot-net.sh
@@ -3,7 +3,7 @@
 cd dotnet-examples/
 
 cd ./kokoro-tts
-./run-kokoro-en.sh
+./run-kokoro.sh
 ls -lh
 
 cd ../offline-tts

diff --git a/dotnet-examples/kokoro-tts/Program.cs b/dotnet-examples/kokoro-tts/Program.cs
@@ -13,6 +13,63 @@
 class OfflineTtsDemo
 {
   static void Main(string[] args)
+  {
+
+    TestZhEn();
+    TestEn();
+  }
+
+  static void TestZhEn()
+  {
+    var config = new OfflineTtsConfig();
+    config.Model.Kokoro.Model = "./kokoro-multi-lang-v1_0/model.onnx";
+    config.Model.Kokoro.Voices = "./kokoro-multi-lang-v1_0/voices.bin";
+    config.Model.Kokoro.Tokens = "./kokoro-multi-lang-v1_0/tokens.txt";
+    config.Model.Kokoro.DataDir = "./kokoro-multi-lang-v1_0/espeak-ng-data";
+    config.Model.Kokoro.DictDir = "./kokoro-multi-lang-v1_0/dict";
+    config.Model.Kokoro.Lexicon = "./kokoro-multi-lang-v1_0/lexicon-us-en.txt,./kokoro-multi-lang-v1_0/lexicon-zh.txt";
+
+    config.Model.NumThreads = 2;
+    config.Model.Debug = 1;
+    config.Model.Provider = "cpu";
+
+    var tts = new OfflineTts(config);
+    var speed = 1.0f;
+    var text = "中英文语音合成测试。This is generated by next generation Kaldi using Kokoro without Misaki. 你觉得中英文说的如何呢？";
+
+    var sid = 50;
+
+    var MyCallback = (IntPtr samples, int n, float progress) =>
+    {
+      float[] data = new float[n];
+      Marshal.Copy(samples, data, 0, n);
+      // You can process samples here, e.g., play them.
+      // See ../kokoro-tts-playback for how to play them
+      Console.WriteLine($"Progress {progress*100}%");
+
+      // 1 means to keep generating
+      // 0 means to stop generating
+      return 1;
+    };
+
+    var callback = new OfflineTtsCallbackProgress(MyCallback);
+
+    var audio = tts.GenerateWithCallbackProgress(text, speed, sid, callback);
+
+    var outputFilename = "./generated-kokoro-zh-en.wav";
+    var ok = audio.SaveToWaveFile(outputFilename);
+
+    if (ok)
+    {
+      Console.WriteLine($"Wrote to {outputFilename} succeeded!");
+    }
+    else
+    {
+      Console.WriteLine($"Failed to write {outputFilename}");
+    }
+  }
+
+  static void TestEn()
   {
     var config = new OfflineTtsConfig();
     config.Model.Kokoro.Model = "./kokoro-en-v0_19/model.onnx";
@@ -54,7 +111,7 @@ static void Main(string[] args)
 
     var audio = tts.GenerateWithCallbackProgress(text, speed, sid, callback);
 
-    var outputFilename = "./generated-kokoro-0.wav";
+    var outputFilename = "./generated-kokoro-en.wav";
     var ok = audio.SaveToWaveFile(outputFilename);
 
     if (ok)

diff --git a/dotnet-examples/kokoro-tts/run-kokoro-en.sh → dotnet-examples/kokoro-tts/run-kokoro.sh b/dotnet-examples/kokoro-tts/run-kokoro-en.sh → dotnet-examples/kokoro-tts/run-kokoro.sh
@@ -1,6 +1,12 @@
 #!/usr/bin/env bash
 set -ex
 
+if [ ! -f ./kokoro-multi-lang-v1_0/model.onnx ]; then
+  curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/tts-models/kokoro-multi-lang-v1_0.tar.bz2
+  tar xf kokoro-multi-lang-v1_0.tar.bz2
+  rm kokoro-multi-lang-v1_0.tar.bz2
+fi
+
 if [ ! -f ./kokoro-en-v0_19/model.onnx ]; then
   curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/tts-models/kokoro-en-v0_19.tar.bz2
   tar xf kokoro-en-v0_19.tar.bz2

diff --git a/scripts/dotnet/OfflineTtsKokoroModelConfig.cs b/scripts/dotnet/OfflineTtsKokoroModelConfig.cs
@@ -15,6 +15,9 @@ public OfflineTtsKokoroModelConfig()
             DataDir = "";
 
             LengthScale = 1.0F;
+
+            DictDir = "";
+            Lexicon = "";
         }
         [MarshalAs(UnmanagedType.LPStr)]
         public string Model;
@@ -29,5 +32,11 @@ public OfflineTtsKokoroModelConfig()
         public string DataDir;
 
         public float LengthScale;
+
+        [MarshalAs(UnmanagedType.LPStr)]
+        public string DictDir;
+
+        [MarshalAs(UnmanagedType.LPStr)]
+        public string Lexicon;
     }
 }