Add JavaScript API (node-addon) for Kokoro TTS models (#1725)

Author: csukuangfj

.github/scripts/test-nodejs-addon-npm.sh

@@ -85,6 +85,13 @@ fi
 
 echo "----------tts----------"
 
+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
+rm kokoro-en-v0_19.tar.bz2
+
+node ./test_tts_non_streaming_kokoro_en.js
+ls -lh *.wav
+
 curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/tts-models/matcha-icefall-en_US-ljspeech.tar.bz2
 tar xvf matcha-icefall-en_US-ljspeech.tar.bz2
 rm matcha-icefall-en_US-ljspeech.tar.bz2

harmony-os/SherpaOnnxHar/sherpa_onnx/src/main/cpp/non-streaming-tts.cc

@@ -53,6 +53,25 @@ static SherpaOnnxOfflineTtsMatchaModelConfig GetOfflineTtsMatchaModelConfig(
   return c;
 }
 
+static SherpaOnnxOfflineTtsKokoroModelConfig GetOfflineTtsKokoroModelConfig(
+    Napi::Object obj) {
+  SherpaOnnxOfflineTtsKokoroModelConfig c;
+  memset(&c, 0, sizeof(c));
+
+  if (!obj.Has("kokoro") || !obj.Get("kokoro").IsObject()) {
+    return c;
+  }
+
+  Napi::Object o = obj.Get("kokoro").As<Napi::Object>();
+  SHERPA_ONNX_ASSIGN_ATTR_STR(model, model);
+  SHERPA_ONNX_ASSIGN_ATTR_STR(voices, voices);
+  SHERPA_ONNX_ASSIGN_ATTR_STR(tokens, tokens);
+  SHERPA_ONNX_ASSIGN_ATTR_STR(data_dir, dataDir);
+  SHERPA_ONNX_ASSIGN_ATTR_FLOAT(length_scale, lengthScale);
+
+  return c;
+}
+
 static SherpaOnnxOfflineTtsModelConfig GetOfflineTtsModelConfig(
     Napi::Object obj) {
   SherpaOnnxOfflineTtsModelConfig c;
@@ -66,6 +85,7 @@ static SherpaOnnxOfflineTtsModelConfig GetOfflineTtsModelConfig(
 
   c.vits = GetOfflineTtsVitsModelConfig(o);
   c.matcha = GetOfflineTtsMatchaModelConfig(o);
+  c.kokoro = GetOfflineTtsKokoroModelConfig(o);
 
   SHERPA_ONNX_ASSIGN_ATTR_INT32(num_threads, numThreads);
 
@@ -180,6 +200,22 @@ static Napi::External<SherpaOnnxOfflineTts> CreateOfflineTtsWrapper(
     delete[] c.model.matcha.dict_dir;
   }
 
+  if (c.model.kokoro.model) {
+    delete[] c.model.kokoro.model;
+  }
+
+  if (c.model.kokoro.voices) {
+    delete[] c.model.kokoro.voices;
+  }
+
+  if (c.model.kokoro.tokens) {
+    delete[] c.model.kokoro.tokens;
+  }
+
+  if (c.model.kokoro.data_dir) {
+    delete[] c.model.kokoro.data_dir;
+  }
+
   if (c.model.provider) {
     delete[] c.model.provider;
   }

nodejs-addon-examples/README.md

@@ -133,6 +133,7 @@ The following tables list the examples in this folder.
 
 |File| Description|
 |---|---|
+|[./test_tts_non_streaming_kokoro_en.js](./test_tts_non_streaming_kokoro_en.js)| Text-to-speech with a Kokoro English Model|
 |[./test_tts_non_streaming_matcha_icefall_en.js](./test_tts_non_streaming_matcha_icefall_en.js)| Text-to-speech with a [MatchaTTS English Model](https://k2-fsa.github.io/sherpa/onnx/tts/pretrained_models/matcha.html#matcha-icefall-en-us-ljspeech-american-english-1-female-speaker)|
 |[./test_tts_non_streaming_matcha_icefall_zhjs](./test_tts_non_streaming_matcha_icefall_zh.js)| Text-to-speech with a [MatchaTTS Chinese Model](https://k2-fsa.github.io/sherpa/onnx/tts/pretrained_models/matcha.html#matcha-icefall-zh-baker-chinese-1-female-speaker)|
 |[./test_tts_non_streaming_vits_piper_en.js](./test_tts_non_streaming_vits_piper_en.js)| Text-to-speech with a [piper](https://github.com/rhasspy/piper) English model|
@@ -347,6 +348,16 @@ npm install naudiodon2
 node ./test_vad_asr_non_streaming_sense_voice_microphone.js
 ```
 
+### Text-to-speech with Kokoro TTS models (English TTS)
+
+```bash
+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
+rm kokoro-en-v0_19.tar.bz2
+
+node ./test_tts_non_streaming_kokoro_en.js
+```
+
 ### Text-to-speech with MatchaTTS models (English TTS)
 ```bash
 wget https://github.com/k2-fsa/sherpa-onnx/releases/download/tts-models/matcha-icefall-en_US-ljspeech.tar.bz2

nodejs-addon-examples/test_tts_non_streaming_kokoro_en.js

@@ -0,0 +1,47 @@
+// Copyright (c)  2025  Xiaomi Corporation
+const sherpa_onnx = require('sherpa-onnx-node');
+
+// please refer to
+// https://k2-fsa.github.io/sherpa/onnx/tts/pretrained_models/kokoro.html
+// to download model files
+function createOfflineTts() {
+  const config = {
+    model: {
+      kokoro: {
+        model: './kokoro-en-v0_19/model.onnx',
+        voices: './kokoro-en-v0_19/voices.bin',
+        tokens: './kokoro-en-v0_19/tokens.txt',
+        dataDir: './kokoro-en-v0_19/espeak-ng-data',
+      },
+      debug: true,
+      numThreads: 1,
+      provider: 'cpu',
+    },
+    maxNumSentences: 1,
+  };
+  return new sherpa_onnx.OfflineTts(config);
+}
+
+const tts = createOfflineTts();
+
+const text =
+    'Today as always, men fall into two groups: slaves and free men. Whoever does not have two-thirds of his day for himself, is a slave, whatever he may be: a statesman, a businessman, an official, or a scholar.'
+
+
+let start = Date.now();
+const audio = tts.generate({text: text, sid: 6, speed: 1.0});
+let stop = Date.now();
+const elapsed_seconds = (stop - start) / 1000;
+const duration = audio.samples.length / audio.sampleRate;
+const real_time_factor = elapsed_seconds / duration;
+console.log('Wave duration', duration.toFixed(3), 'secodns')
+console.log('Elapsed', elapsed_seconds.toFixed(3), 'secodns')
+console.log(
+    `RTF = ${elapsed_seconds.toFixed(3)}/${duration.toFixed(3)} =`,
+    real_time_factor.toFixed(3))
+
+const filename = 'test-kokoro-en-6.wav';
+sherpa_onnx.writeWave(
+    filename, {samples: audio.samples, sampleRate: audio.sampleRate});
+
+console.log(`Saved to ${filename}`);