onnx 加载本地模型 tts基础功能完成

config.gradle

@@ -5,7 +5,7 @@ ext {
 
             compileSdkVersion: 34,
             buildToolsVersion: "29.0.2",
-            minSdkVersion    : 19,
+            minSdkVersion    : 21,
             targetSdkVersion : 29,
     ]
     dependencies = [

libraries/mogo-tts/app/build.gradle

@@ -68,6 +68,7 @@ dependencies {
 
     implementation(project(":libraries:mogo-tts:tts-iflytek-offline"))
     implementation(project(":libraries:mogo-tts:tts-base"))
+    implementation(project(":libraries:mogo-tts:tts-mogo"))
 
     // ======================集成 Bugly================================================================
     implementation 'com.tencent.bugly:crashreport:4.1.9.2'

libraries/mogo-tts/app/src/main/java/com/mogo/tts/MoGoApplication.kt

@@ -9,7 +9,6 @@ import com.blankj.utilcode.util.PathUtils
 import com.blankj.utilcode.util.TimeUtils
 import com.bytedance.boost_multidex.BoostMultiDex
 import com.mogo.tts.common.TtsManager
-import com.mogo.tts.iflytek.offline.IFlyTekOfflineTts
 import com.mogo.tts.utils.PhoneUtilsExtend
 import com.tencent.bugly.crashreport.CrashReport
 import com.tencent.bugly.crashreport.CrashReport.CrashHandleCallback

libraries/mogo-tts/tts-base/src/main/java/com/mogo/tts/common/TtsManager.kt

@@ -14,7 +14,8 @@ object TtsManager {
             // 暂时换成反射，解决死锁问题
             var clazz1: Class<*>? = null
             try {
-                clazz1 = Class.forName("com.mogo.tts.iflytek.offline.IFlyTekOfflineTts")
+                //clazz1 = Class.forName("com.mogo.tts.iflytek.offline.IFlyTekOfflineTts")
+                clazz1 = Class.forName("com.k2fsa.sherpa.onnx.MogoOfflineTTS")
             } catch (ignored: Exception) {
             }
 

libraries/mogo-tts/tts-base/src/main/java/com/mogo/tts/common/impl/BaseMogoTTS.java

@@ -0,0 +1,103 @@
+package com.mogo.tts.common.impl;
+
+import android.content.Context;
+import android.os.Looper;
+
+import com.elegant.utils.UiThreadHandler;
+import com.mogo.tts.common.IGlobalTtsCallback;
+import com.mogo.tts.common.IMogoTTS;
+import com.mogo.tts.common.IMogoTTSCallback;
+import com.mogo.tts.common.LangTtsEntity;
+import com.mogo.tts.common.LanguageType;
+import com.mogo.tts.common.log.TtsLogManager;
+
+import java.util.HashMap;
+
+public abstract class BaseMogoTTS implements IMogoTTS {
+    protected Context context;
+
+    // 由于主动打断不会有回调事件，所以主动打断时清掉map中被打断的text和callback
+    protected volatile String curTtsContent = "";
+    protected volatile LangTtsEntity curTtsEntity = null;
+
+    protected HashMap<String, IMogoTTSCallback> speakVoiceMap = new HashMap<>();
+
+    protected IGlobalTtsCallback mGlobalTtsCallback = null;
+
+    protected String getTAG() {
+        return "BaseMogoTTS";
+    }
+
+    @Override
+    public void initTts(Context context) {
+        this.context = context;
+    }
+
+    @Override
+    public void speakTTSVoice(String tts) {
+        if (tts.isEmpty()) return;
+        speakTTSVoice(tts,null);
+    }
+
+    @Override
+    public void speakTTSVoice(String tts, IMogoTTSCallback callBack) {
+        if (tts.isEmpty()) return;
+        speakTTSVoice(new LangTtsEntity(tts, LanguageType.CHINESE), callBack);
+    }
+
+    @Override
+    public void speakTTSVoice(LangTtsEntity ttsEntity, IMogoTTSCallback callBack) {
+        if (Thread.currentThread() != Looper.getMainLooper().getThread()) {
+            UiThreadHandler.post(new Runnable() {
+                @Override
+                public void run() {
+                    if (callBack != null) {
+                        speakVoiceMap.put(ttsEntity.toString(),callBack);
+                    }
+                    speakMultiLangTTS(ttsEntity);
+                }
+            });
+        }else {
+            if (callBack != null) {
+                speakVoiceMap.put(ttsEntity.toString(),callBack);
+            }
+            speakMultiLangTTS(ttsEntity);
+        }
+    }
+
+    protected void speakMultiLangTTS(LangTtsEntity ttsEntity){
+        this.curTtsEntity = ttsEntity;
+        // 合成并播放
+        TtsLogManager.d(getTAG(), "tts准备合成："+ttsEntity);
+    }
+
+    @Override
+    public void stopTts() {
+        if (Thread.currentThread() != Looper.getMainLooper().getThread()) {
+            UiThreadHandler.post(new Runnable() {
+                @Override
+                public void run() {
+                    realStop();
+                }
+            });
+        } else {
+            realStop();
+        }
+    }
+
+    protected void realStop() {
+        TtsLogManager.d(getTAG(),"停止tts");
+        if(curTtsEntity!=null){
+            String key = curTtsEntity.toString();
+            if (speakVoiceMap.containsKey(key)) {
+                speakVoiceMap.remove(key).onStopTts(key);
+            }
+            curTtsEntity = null;
+        }
+        this.curTtsContent = "";
+    }
+
+    public void registerTtsListener(IGlobalTtsCallback callback) {
+        this.mGlobalTtsCallback = callback;
+    }
+}

libraries/mogo-tts/tts-iflytek-offline/src/main/java/com/mogo/tts/iflytek/offline/IFlyTekOfflineTts.kt

@@ -18,21 +18,14 @@ import com.iflytek.aikit.core.BaseLibrary
 import com.iflytek.aikit.core.CoreListener
 import com.iflytek.aikit.core.ErrType
 import com.mogo.tts.common.IGlobalTtsCallback
-import com.mogo.tts.common.IMogoTTS
-import com.mogo.tts.common.IMogoTTSCallback
 import com.mogo.tts.common.LangTtsEntity
 import com.mogo.tts.common.log.TtsLogManager
-import com.mogo.tts.common.LanguageType
+import com.mogo.tts.common.impl.BaseMogoTTS
 import com.mogo.tts.common.utils.FileUtils
 import java.io.File
 
 @Keep
-class IFlyTekOfflineTts : IMogoTTS {
+class IFlyTekOfflineTts : BaseMogoTTS() {
-    companion object {
-        const val TAG = "IFlyTekTts"
-    }
-
-    private var context: Context? = null
 
     private var aiHandle: AiHandle? = null
     private var OUTPUT_DIR :String = ""
@@ -40,20 +33,12 @@ class IFlyTekOfflineTts : IMogoTTS {
         "e2e44feff"
     }
 
-    // 由于主动打断不会有回调事件，所以主动打断时清掉map中被打断的text和callback
+    override fun getTAG(): String {
-    @Volatile
+        return "IFlyTekOfflineTts"
-    private var curTtsContent = ""
-    @Volatile
-    private var curTtsEntity: LangTtsEntity? = null
-
-    private val speakVoiceMap by lazy {
-        HashMap<String, IMogoTTSCallback>()
     }
 
-    private var mGlobalTtsCallback: IGlobalTtsCallback? = null
-
     override fun initTts(context: Context) {
-        this.context = context
+        super.initTts(context)
         initSDK()
     }
 
@@ -96,16 +81,16 @@ class IFlyTekOfflineTts : IMogoTTS {
             ErrType.AUTH -> {
                 if (code == 0) {
                     // SDK授权成功
-                    TtsLogManager.d(TAG, "科大讯飞离线语音合成授权成功!")
+                    TtsLogManager.d(tag, "科大讯飞离线语音合成授权成功!")
                 } else {
                     // SDK授权失败，授权码为：code
-                    TtsLogManager.d(TAG, "科大讯飞离线语音合成授权失败码：$code")
+                    TtsLogManager.d(tag, "科大讯飞离线语音合成授权失败码：$code")
                 }
             }
 
             else -> {
                 // SDK状态为：type, code
-                TtsLogManager.d(TAG, "type:$type, code:$code")
+                TtsLogManager.d(tag, "type:$type, code:$code")
             }
         }
     }
@@ -126,69 +111,19 @@ class IFlyTekOfflineTts : IMogoTTS {
         }
     }
 
-    override fun speakTTSVoice(tts: String?) {
+    override fun realStop() {
-        speakTTSVoice(tts, null)
+        super.realStop()
-    }
-
-    override fun speakTTSVoice(tts: String?, callBack: IMogoTTSCallback?) {
-        if (tts.isNullOrEmpty()) return
-        speakTTSVoice(LangTtsEntity(tts, LanguageType.CHINESE), callBack)
-    }
-
-    override fun speakTTSVoice(ttsEntity: LangTtsEntity, callBack: IMogoTTSCallback?) {
-        if (Thread.currentThread() != Looper.getMainLooper().thread) {
-            UiThreadHandler.post {
-                if (callBack != null) {
-                    speakVoiceMap[ttsEntity.toString()] = callBack
-                }
-                speakMultiLangTTS(ttsEntity)
-            }
-        } else {
-            if (callBack != null) {
-                speakVoiceMap[ttsEntity.toString()] = callBack
-            }
-            speakMultiLangTTS(ttsEntity)
-        }
-    }
-
-    override fun stopTts() {
-        if (Thread.currentThread() != Looper.getMainLooper().thread) {
-            UiThreadHandler.post {
-                realStop()
-            }
-        } else {
-            realStop()
-        }
-    }
-
-    private fun realStop() {
-        TtsLogManager.d(TAG,"停止tts")
-        curTtsEntity?.let {
-            val string = it.toString()
-            if (speakVoiceMap.containsKey(string)) {
-                speakVoiceMap.remove(string)?.onStopTts(string)
-            }
-            curTtsEntity = null
-        }
-        curTtsContent = ""
         AudioTrackManager.instance?.stopPlay()
         if (aiHandle == null || aiHandle?.isSuccess == false) {
             return
         }
         val end = AiHelper.getInst().end(aiHandle)
         aiHandle = null
-        TtsLogManager.d(TAG,"停止tts:${end}")
+        TtsLogManager.d(tag,"停止tts:${end}")
     }
 
-
+    override fun speakMultiLangTTS(ttsEntity: LangTtsEntity) {
-    override fun registerTtsListener(callback: IGlobalTtsCallback?) {
+        super.speakMultiLangTTS(ttsEntity)
-        mGlobalTtsCallback = callback
-    }
-
-    private fun speakMultiLangTTS(ttsEntity: LangTtsEntity) {
-        curTtsEntity = ttsEntity
-        // 合成并播放
-        TtsLogManager.d(TAG, "tts准备合成：$ttsEntity")
         stopTts()
         startSpeak(ttsEntity)
     }
@@ -278,7 +213,7 @@ class IFlyTekOfflineTts : IMogoTTS {
                 var bytes: ByteArray?
                 for (i in list.indices) {
                     bytes = list[i].value ?: continue
-                    TtsLogManager.d(TAG, "onResult:handleID:" + handleID + ":" + list[i].key)
+                    TtsLogManager.d(tag, "onResult:handleID:" + handleID + ":" + list[i].key)
                     if (!dir.exists()) {
                         dir.mkdirs()
                     }
@@ -295,23 +230,23 @@ class IFlyTekOfflineTts : IMogoTTS {
         ) {
             when (event) {
                 AeeEvent.AEE_EVENT_UNKNOWN.value -> {
-                    TtsLogManager.d(TAG, "未知错误")
+                    TtsLogManager.d(tag, "未知错误")
 //                    handleErrorEvent("未知错误")
                     aiHandle?.let {
                         val ret = AiHelper.getInst().end(it)
                         aiHandle = null
-                        TtsLogManager.d(TAG, "AIKit_End：$ret")
+                        TtsLogManager.d(tag, "AIKit_End：$ret")
                     }
                     val errorInfo = ResourcesHelper.getString(context, R.string.module_tts_unknown_error)
                     handleErrorEvent(errorInfo)
                 }
                 AeeEvent.AEE_EVENT_TIMEOUT.value->{
-                    TtsLogManager.d(TAG, "超时错误")
+                    TtsLogManager.d(tag, "超时错误")
 //                    handleErrorEvent("未知错误")
                     aiHandle?.let {
                         val ret = AiHelper.getInst().end(it)
                         aiHandle = null
-                        TtsLogManager.d(TAG, "AIKit_End：$ret")
+                        TtsLogManager.d(tag, "AIKit_End：$ret")
                     }
                     val errorInfo = ResourcesHelper.getString(context, R.string.module_tts_unknown_error)
                     handleErrorEvent(errorInfo)
@@ -324,7 +259,7 @@ class IFlyTekOfflineTts : IMogoTTS {
                     aiHandle?.let {
                         val ret = AiHelper.getInst().end(it)
                         aiHandle = null
-                        TtsLogManager.d(TAG, "AIKit_End：$ret")
+                        TtsLogManager.d(tag, "AIKit_End：$ret")
                     }
                     onSpeakBegin()
                     AudioTrackManager.instance?.startPlay("${OUTPUT_DIR}/OutPut_mogo.pcm")
@@ -333,7 +268,7 @@ class IFlyTekOfflineTts : IMogoTTS {
         }
 
         override fun onError(handleID: Int, err: Int, msg: String?, usrCxt: Any?) {
-            TtsLogManager.d(TAG, "错误码：$err,错误信息:$msg")
+            TtsLogManager.d(tag, "错误码：$err,错误信息:$msg")
 //            handleErrorEvent("错误码：$err,错误信息:$msg")
             val errorInfo = ResourcesHelper.getResources(context)
                 .getString(R.string.module_tts_ai_handle_error_code, err, msg)

libraries/mogo-tts/tts-mogo/build.gradle

@@ -32,6 +32,8 @@ dependencies {
     implementation fileTree(dir: "libs", include: ["*.jar"])
     implementation rootProject.ext.dependencies.androidxappcompat
 
+    implementation(project(":libraries:mogo-tts:tts-base"))
+
 }
 
 apply from: new File(rootProject.rootDir, "gradle/upload.gradle").toString()

libraries/mogo-tts/tts-mogo/src/main/java/com/k2fsa/sherpa/onnx/MogoOfflineTTS.kt

@@ -0,0 +1,228 @@
+package com.k2fsa.sherpa.onnx
+
+import android.content.Context
+import android.content.res.AssetManager
+import android.media.AudioAttributes
+import android.media.AudioFormat
+import android.media.AudioManager
+import android.media.AudioTrack
+import android.util.Log
+import com.elegant.utils.ThreadPoolService
+import com.elegant.utils.UiThreadHandler
+import com.mogo.tts.common.LangTtsEntity
+import com.mogo.tts.common.impl.BaseMogoTTS
+import com.mogo.tts.common.log.TtsLogManager
+import com.mogo.tts.common.utils.FileUtils
+import java.io.File
+import java.util.concurrent.locks.ReentrantLock
+
+class MogoOfflineTTS : BaseMogoTTS() {
+
+    private var tts: OfflineTts? = null
+    private var track: AudioTrack? = null
+    @Volatile
+    private var stopped: Boolean = false
+    private var OUTPUT_DIR :String = ""
+    private var waitPlayInfo:LangTtsEntity?=null
+
+    val lock = ReentrantLock()
+
+    override fun getTAG(): String {
+        return "MogoOfflineTTS"
+    }
+
+    override fun initTts(context: Context?) {
+        super.initTts(context)
+        initSdk()
+    }
+
+    private fun initSdk() {
+
+        context?.let {
+            val workPath = it.filesDir.absolutePath+ File.separator+"mogotts"+ File.separator+"xtts"
+            OUTPUT_DIR = workPath+File.separator+"output"
+            val file = File("$workPath/yue_dict")
+            if (!file.exists() || file.length() == 0L) {
+                ThreadPoolService.execute {
+                    try {
+                        FileUtils.copyAssetsToLocal(context, "matcha-icefall-zh-en/espeak-ng-data", workPath)
+                        UiThreadHandler.post {
+                            initEngine(workPath)
+                            initAudioTrack()
+                        }
+                    }catch (e:Exception){
+                        e.printStackTrace()
+                    }
+
+                }
+            } else {
+                try {
+                    initEngine(workPath)
+                    initAudioTrack()
+                }catch (e:Exception){
+                    e.printStackTrace()
+                }
+            }
+        }
+    }
+
+    private fun initEngine(workPath: String) {
+
+        // The purpose of such a design is to make the CI test easier
+        // Please see
+        // https://github.com/k2-fsa/sherpa-onnx/blob/master/scripts/apk/generate-tts-apk-script.py
+
+
+        val modelDir = "matcha-icefall-zh-en"
+        val acousticModelName  = "model-steps-3.onnx"
+        val vocoder = "vocos-16khz-univ.onnx"
+        val lexicon = "lexicon.txt"
+        val assets: AssetManager? = context.assets
+        val isKitten = false
+
+        val config = getOfflineTtsConfig(
+            modelDir = modelDir,
+            modelName = "",
+            acousticModelName = acousticModelName ?: "",
+            vocoder = vocoder ?: "",
+            voices = "",
+            lexicon = lexicon ?: "",
+            dataDir = workPath,
+            dictDir = "",
+            ruleFsts = "",
+            ruleFars = "",
+            isKitten = isKitten,
+        )
+
+        tts = OfflineTts(assetManager = assets, config = config)
+
+        TtsLogManager.d(tag,"初始化tts:${tts?.config}")
+    }
+
+    private fun initAudioTrack() {
+        tts?.let {
+            val sampleRate = it.sampleRate()
+            val bufLength = AudioTrack.getMinBufferSize(
+                sampleRate,
+                AudioFormat.CHANNEL_OUT_MONO,
+                AudioFormat.ENCODING_PCM_FLOAT
+            )
+            Log.i(tag, "sampleRate: $sampleRate, buffLength: $bufLength")
+
+            val attr = AudioAttributes.Builder().setContentType(AudioAttributes.CONTENT_TYPE_SPEECH)
+                .setUsage(AudioAttributes.USAGE_MEDIA)
+                .build()
+
+            val format = AudioFormat.Builder()
+                .setEncoding(AudioFormat.ENCODING_PCM_FLOAT)
+                .setChannelMask(AudioFormat.CHANNEL_OUT_MONO)
+                .setSampleRate(sampleRate)
+                .build()
+
+            track = AudioTrack(
+                attr, format, bufLength, AudioTrack.MODE_STREAM,
+                AudioManager.AUDIO_SESSION_ID_GENERATE
+            )
+            track?.play()
+        }
+    }
+
+    override fun release() {
+        tts?.release()
+    }
+
+    override fun speakMultiLangTTS(ttsEntity: LangTtsEntity?) {
+        super.speakMultiLangTTS(ttsEntity)
+        TtsLogManager.d(tag,"1 播放语言:${ttsEntity}")
+        if (lock.isLocked) {
+            waitPlayInfo = ttsEntity
+            TtsLogManager.d(tag,"speakMultiLangTTS 正在生成 锁定中")
+            return
+        }else{
+            if(waitPlayInfo!=null) {
+                TtsLogManager.d(tag,"speakMultiLangTTS 没有被锁 等待播放有值 ${waitPlayInfo}")
+                waitPlayInfo = null
+            }
+        }
+        startSpeak(ttsEntity)
+    }
+
+    private fun startSpeak(langTtsEntity: LangTtsEntity?) {
+        stopTts()
+        langTtsEntity?.let {
+            curTtsContent = it.ttsContent
+            realSpeak(it.ttsContent)
+        }
+    }
+
+    private fun realSpeak(content: String) {
+        TtsLogManager.d(tag,"2 realSpeak-开始进行生成播放--${Thread.currentThread().name}")
+        track?.pause()
+        track?.flush()
+        track?.play()
+        ThreadPoolService.execute {
+            var audio:GeneratedAudio?=null
+            stopped = false
+            try {
+                lock.lock()
+                TtsLogManager.d(tag,"3 开始生成 去加锁")
+                audio = tts?.generateWithConfigAndCallback(
+                    text = content,
+                    config = GenerationConfig(sid = 0, speed = 1.0f),
+                    callback =  object : Function1<FloatArray, Int> {
+                        override fun invoke(samples: FloatArray): Int {
+                            if(lock.isLocked){
+                                TtsLogManager.d(tag,"4 生成成功 去解锁")
+                                lock.unlock()
+                            }
+                            if(waitPlayInfo!=null){
+                                TtsLogManager.d(tag,"生成成功 解锁后 等待播放有值 停止播放去生成")
+                                track?.stop()
+                                UiThreadHandler.post {
+                                    waitPlayInfo?.let {
+                                        startSpeak(it.copy())
+                                        waitPlayInfo = null
+                                    }
+
+                                }
+                                return 0
+                            }
+                            if (!stopped) {
+                                track?.write(samples, 0, samples.size, AudioTrack.WRITE_BLOCKING)
+                                return 1
+                            } else {
+                                track?.stop()
+                                return 0
+                            }
+                        }
+                    }
+                )
+                TtsLogManager.d(tag,"5 realSpeak-结束播放语音")
+            }finally {
+                if(lock.isLocked) {
+                    TtsLogManager.d(tag,"realSpeak-finally 中发现还在锁定解锁")
+                    lock.unlock()
+                }
+            }
+            val filename = "$OUTPUT_DIR/generated.wav"
+            audio?.let {
+                val ok = it.samples.isNotEmpty() && it.save(filename)
+                if(ok){
+                    UiThreadHandler.post {
+                        track?.stop()
+                    }
+                }
+            }
+
+        }
+    }
+
+    override fun realStop() {
+        super.realStop()
+        TtsLogManager.d(tag,"mogooffline realStop")
+        stopped = true
+        track?.pause()
+        track?.flush()
+    }
+
+}

libraries/mogo-tts/tts-mogo/src/main/java/com/k2fsa/sherpa/onnx/Tts.kt

@@ -0,0 +1,382 @@
+// Copyright (c)  2023  Xiaomi Corporation
+package com.k2fsa.sherpa.onnx
+
+import android.content.res.AssetManager
+
+data class OfflineTtsVitsModelConfig(
+    var model: String = "",
+    var lexicon: String = "",
+    var tokens: String = "",
+    var dataDir: String = "",
+    var dictDir: String = "", // unused
+    var noiseScale: Float = 0.667f,
+    var noiseScaleW: Float = 0.8f,
+    var lengthScale: Float = 1.0f,
+)
+
+data class OfflineTtsMatchaModelConfig(
+    var acousticModel: String = "",
+    var vocoder: String = "",
+    var lexicon: String = "",
+    var tokens: String = "",
+    var dataDir: String = "",
+    var dictDir: String = "", // unused
+    var noiseScale: Float = 1.0f,
+    var lengthScale: Float = 1.0f,
+)
+
+data class OfflineTtsKokoroModelConfig(
+    var model: String = "",
+    var voices: String = "",
+    var tokens: String = "",
+    var dataDir: String = "",
+    var lexicon: String = "",
+    var lang: String = "",
+    var dictDir: String = "", // unused
+    var lengthScale: Float = 1.0f,
+)
+
+data class OfflineTtsZipVoiceModelConfig(
+    var tokens: String = "",
+    var encoder: String = "",
+    var decoder: String = "",
+    var vocoder: String = "",
+    var dataDir: String = "",
+    var lexicon: String = "",
+    var featScale: Float = 0.1f,
+    var tShift: Float = 0.5f,
+    var targetRms: Float = 0.1f,
+    var guidanceScale: Float = 1.0f,
+)
+
+data class OfflineTtsKittenModelConfig(
+    var model: String = "",
+    var voices: String = "",
+    var tokens: String = "",
+    var dataDir: String = "",
+    var lengthScale: Float = 1.0f,
+)
+
+/**
+ * Configuration for Pocket TTS models.
+ *
+ * See https://k2-fsa.github.io/sherpa/onnx/tts/pocket/index.html for details.
+ *
+ * @property lmFlow Path to the LM flow model (.onnx)
+ * @property lmMain Path to the LM main model (.onnx)
+ * @property encoder Path to the encoder model (.onnx)
+ * @property decoder Path to the decoder model (.onnx)
+ * @property textConditioner Path to the text conditioner model (.onnx)
+ * @property vocabJson Path to vocabulary JSON file
+ * @property tokenScoresJson Path to token scores JSON file
+ */
+data class OfflineTtsPocketModelConfig(
+  var lmFlow: String = "",
+  var lmMain: String = "",
+  var encoder: String = "",
+  var decoder: String = "",
+  var textConditioner: String = "",
+  var vocabJson: String = "",
+  var tokenScoresJson: String = "",
+  var voiceEmbeddingCacheCapacity: Int = 50,
+)
+
+data class OfflineTtsSupertonicModelConfig(
+  var durationPredictor: String = "",
+  var textEncoder: String = "",
+  var vectorEstimator: String = "",
+  var vocoder: String = "",
+  var ttsJson: String = "",
+  var unicodeIndexer: String = "",
+  var voiceStyle: String = "",
+)
+
+data class OfflineTtsModelConfig(
+    var vits: OfflineTtsVitsModelConfig = OfflineTtsVitsModelConfig(),
+    var matcha: OfflineTtsMatchaModelConfig = OfflineTtsMatchaModelConfig(),
+    var kokoro: OfflineTtsKokoroModelConfig = OfflineTtsKokoroModelConfig(),
+    var zipvoice: OfflineTtsZipVoiceModelConfig = OfflineTtsZipVoiceModelConfig(),
+    var kitten: OfflineTtsKittenModelConfig = OfflineTtsKittenModelConfig(),
+    var pocket: OfflineTtsPocketModelConfig = OfflineTtsPocketModelConfig(),
+    var supertonic: OfflineTtsSupertonicModelConfig = OfflineTtsSupertonicModelConfig(),
+
+    var numThreads: Int = 1,
+    var debug: Boolean = false,
+    var provider: String = "cpu",
+)
+
+data class OfflineTtsConfig(
+    var model: OfflineTtsModelConfig = OfflineTtsModelConfig(),
+    var ruleFsts: String = "",
+    var ruleFars: String = "",
+    var maxNumSentences: Int = 1,
+    var silenceScale: Float = 0.2f,
+)
+
+class GeneratedAudio(
+    val samples: FloatArray,
+    val sampleRate: Int,
+) {
+    fun save(filename: String) = saveImpl(filename = filename, samples = samples, sampleRate = sampleRate)
+
+    private external fun saveImpl(
+        filename: String,
+        samples: FloatArray,
+        sampleRate: Int
+    ): Boolean
+}
+
+data class GenerationConfig(
+    var silenceScale: Float = 0.2f,
+    var speed: Float = 1.0f,
+    var sid: Int = 0,
+    var referenceAudio: FloatArray? = null,
+    var referenceSampleRate: Int = 0,
+    var referenceText: String? = null,
+    var numSteps: Int = 5,
+    var extra: Map<String, String>? = null
+)
+
+class OfflineTts(
+    assetManager: AssetManager? = null,
+    var config: OfflineTtsConfig,
+) {
+    private var ptr: Long
+
+    init {
+        ptr = if (assetManager != null) {
+            newFromAsset(assetManager, config)
+        } else {
+            newFromFile(config)
+        }
+    }
+
+    fun sampleRate() = getSampleRate(ptr)
+
+    fun numSpeakers() = getNumSpeakers(ptr)
+
+    fun generate(
+        text: String,
+        sid: Int = 0,
+        speed: Float = 1.0f
+    ): GeneratedAudio {
+        return generateImpl(ptr, text = text, sid = sid, speed = speed)
+    }
+
+    fun generateWithCallback(
+        text: String,
+        sid: Int = 0,
+        speed: Float = 1.0f,
+        callback: (samples: FloatArray) -> Int
+    ): GeneratedAudio {
+        return generateWithCallbackImpl(
+            ptr,
+            text = text,
+            sid = sid,
+            speed = speed,
+            callback = callback
+        )
+    }
+
+    fun generateWithConfig(
+      text: String,
+      config: GenerationConfig
+    ): GeneratedAudio {
+        return generateWithConfigImpl(ptr, text, config, null)
+    }
+
+    fun generateWithConfigAndCallback(
+        text: String,
+        config: GenerationConfig,
+        callback: (samples: FloatArray) -> Int
+    ): GeneratedAudio {
+        return generateWithConfigImpl(ptr, text, config, callback)
+    }
+
+    fun allocate(assetManager: AssetManager? = null) {
+        if (ptr == 0L) {
+            ptr = if (assetManager != null) {
+                newFromAsset(assetManager, config)
+            } else {
+                newFromFile(config)
+            }
+        }
+    }
+
+    fun free() {
+        if (ptr != 0L) {
+            delete(ptr)
+            ptr = 0
+        }
+    }
+
+    protected fun finalize() {
+        if (ptr != 0L) {
+            delete(ptr)
+            ptr = 0
+        }
+    }
+
+    fun release() = finalize()
+
+    private external fun newFromAsset(
+        assetManager: AssetManager,
+        config: OfflineTtsConfig,
+    ): Long
+
+    private external fun newFromFile(
+        config: OfflineTtsConfig,
+    ): Long
+
+    private external fun delete(ptr: Long)
+    private external fun getSampleRate(ptr: Long): Int
+    private external fun getNumSpeakers(ptr: Long): Int
+
+    // The returned array has two entries:
+    //  - the first entry is an 1-D float array containing audio samples.
+    //    Each sample is normalized to the range [-1, 1]
+    //  - the second entry is the sample rate
+    private external fun generateImpl(
+        ptr: Long,
+        text: String,
+        sid: Int = 0,
+        speed: Float = 1.0f
+    ): GeneratedAudio
+
+    private external fun generateWithCallbackImpl(
+        ptr: Long,
+        text: String,
+        sid: Int = 0,
+        speed: Float = 1.0f,
+        callback: (samples: FloatArray) -> Int
+    ): GeneratedAudio
+
+
+    private external fun generateWithConfigImpl(
+        ptr: Long,
+        text: String,
+        config: GenerationConfig,
+        callback: ((samples: FloatArray) -> Int)?
+    ): GeneratedAudio
+
+    companion object {
+        init {
+            System.loadLibrary("sherpa-onnx-jni")
+        }
+    }
+}
+
+// please refer to
+// https://k2-fsa.github.io/sherpa/onnx/tts/pretrained_models/index.html
+// to download models
+fun getOfflineTtsConfig(
+    modelDir: String,
+    modelName: String, // for VITS
+    acousticModelName: String, // for Matcha
+    vocoder: String, // for Matcha
+    voices: String, // for Kokoro or kitten
+    lexicon: String,
+    dataDir: String,
+    dictDir: String, // unused
+    ruleFsts: String,
+    ruleFars: String,
+    numThreads: Int? = null,
+    isKitten: Boolean = false
+): OfflineTtsConfig {
+    // For Matcha TTS, please set
+    // acousticModelName, vocoder
+
+    // For Kokoro TTS, please set
+    // modelName, voices
+
+    // For Kitten TTS, please set
+    // modelName, voices, isKitten
+
+    // For VITS, please set
+    // modelName
+
+    val numberOfThreads = if (numThreads != null) {
+        numThreads
+    } else if (voices.isNotEmpty()) {
+        // for Kokoro and Kitten TTS models, we use more threads
+        4
+    } else {
+        2
+    }
+
+    if (modelName.isEmpty() && acousticModelName.isEmpty()) {
+        throw IllegalArgumentException("Please specify a TTS model")
+    }
+
+    if (modelName.isNotEmpty() && acousticModelName.isNotEmpty()) {
+        throw IllegalArgumentException("Please specify either a VITS or a Matcha model, but not both")
+    }
+
+    if (acousticModelName.isNotEmpty() && vocoder.isEmpty()) {
+        throw IllegalArgumentException("Please provide vocoder for Matcha TTS")
+    }
+
+    val vits = if (modelName.isNotEmpty() && voices.isEmpty()) {
+        OfflineTtsVitsModelConfig(
+            model = "$modelDir/$modelName",
+            lexicon = "$modelDir/$lexicon",
+            tokens = "$modelDir/tokens.txt",
+            dataDir = dataDir,
+        )
+    } else {
+        OfflineTtsVitsModelConfig()
+    }
+
+    val matcha = if (acousticModelName.isNotEmpty()) {
+        OfflineTtsMatchaModelConfig(
+            acousticModel = "$modelDir/$acousticModelName",
+            vocoder = vocoder,
+            lexicon = "$modelDir/$lexicon",
+            tokens = "$modelDir/tokens.txt",
+            dataDir = dataDir,
+        )
+    } else {
+        OfflineTtsMatchaModelConfig()
+    }
+
+    val kokoro = if (voices.isNotEmpty() && !isKitten) {
+        OfflineTtsKokoroModelConfig(
+            model = "$modelDir/$modelName",
+            voices = "$modelDir/$voices",
+            tokens = "$modelDir/tokens.txt",
+            dataDir = dataDir,
+            lexicon = when {
+                lexicon == "" -> lexicon
+                "," in lexicon -> lexicon
+                else -> "$modelDir/$lexicon"
+            },
+        )
+    } else {
+        OfflineTtsKokoroModelConfig()
+    }
+
+    val kitten = if (isKitten) {
+        OfflineTtsKittenModelConfig(
+            model = "$modelDir/$modelName",
+            voices = "$modelDir/$voices",
+            tokens = "$modelDir/tokens.txt",
+            dataDir = dataDir,
+        )
+    } else {
+        OfflineTtsKittenModelConfig()
+    }
+
+    return OfflineTtsConfig(
+        model = OfflineTtsModelConfig(
+            vits = vits,
+            matcha = matcha,
+            kokoro = kokoro,
+            kitten = kitten,
+            numThreads = numberOfThreads,
+            debug = true,
+            provider = "cpu",
+        ),
+        ruleFsts = ruleFsts,
+        ruleFars = ruleFars,
+    )
+}

libraries/mogo-tts/tts-mogo/src/main/jniLibs/arm64-v8a/README.md

@@ -0,0 +1,11 @@
+# Introduction
+
+Note that if you use Android Studio, then you only need to
+copy libonnxruntime.so and libsherpa-onnx-jni.so
+to your jniLibs, and you don't need libsherpa-onnx-c-api.so or
+libsherpa-onnx-cxx-api.so.
+
+libsherpa-onnx-c-api.so and libsherpa-onnx-cxx-api.so are for users
+who don't use JNI. In that case, libsherpa-onnx-jni.so is not needed.
+
+In any case, libonnxruntime.so is always needed.

libraries/mogo-tts/tts-mogo/src/main/jniLibs/armeabi-v7a/README.md

@@ -0,0 +1,11 @@
+# Introduction
+
+Note that if you use Android Studio, then you only need to
+copy libonnxruntime.so and libsherpa-onnx-jni.so
+to your jniLibs, and you don't need libsherpa-onnx-c-api.so or
+libsherpa-onnx-cxx-api.so.
+
+libsherpa-onnx-c-api.so and libsherpa-onnx-cxx-api.so are for users
+who don't use JNI. In that case, libsherpa-onnx-jni.so is not needed.
+
+In any case, libonnxruntime.so is always needed.

libraries/mogo-tts/tts-mogo/src/main/jniLibs/x86/README.md

@@ -0,0 +1,11 @@
+# Introduction
+
+Note that if you use Android Studio, then you only need to
+copy libonnxruntime.so and libsherpa-onnx-jni.so
+to your jniLibs, and you don't need libsherpa-onnx-c-api.so or
+libsherpa-onnx-cxx-api.so.
+
+libsherpa-onnx-c-api.so and libsherpa-onnx-cxx-api.so are for users
+who don't use JNI. In that case, libsherpa-onnx-jni.so is not needed.
+
+In any case, libonnxruntime.so is always needed.

libraries/mogo-tts/tts-mogo/src/main/jniLibs/x86_64/README.md

@@ -0,0 +1,11 @@
+# Introduction
+
+Note that if you use Android Studio, then you only need to
+copy libonnxruntime.so and libsherpa-onnx-jni.so
+to your jniLibs, and you don't need libsherpa-onnx-c-api.so or
+libsherpa-onnx-cxx-api.so.
+
+libsherpa-onnx-c-api.so and libsherpa-onnx-cxx-api.so are for users
+who don't use JNI. In that case, libsherpa-onnx-jni.so is not needed.
+
+In any case, libonnxruntime.so is always needed.