diff --git a/.github/workflows/apk.yaml b/.github/workflows/apk.yaml index a9d01a10..4f01bd62 100644 --- a/.github/workflows/apk.yaml +++ b/.github/workflows/apk.yaml @@ -38,6 +38,7 @@ jobs: shell: bash run: | export ANDROID_NDK=$ANDROID_NDK_LATEST_HOME + ./build-apk-tts.sh ./build-apk-vad.sh ./build-apk-two-pass.sh ./build-apk.sh diff --git a/android/SherpaOnnxTts/.gitignore b/android/SherpaOnnxTts/.gitignore new file mode 100644 index 00000000..aa724b77 --- /dev/null +++ b/android/SherpaOnnxTts/.gitignore @@ -0,0 +1,15 @@ +*.iml +.gradle +/local.properties +/.idea/caches +/.idea/libraries +/.idea/modules.xml +/.idea/workspace.xml +/.idea/navEditor.xml +/.idea/assetWizardSettings.xml +.DS_Store +/build +/captures +.externalNativeBuild +.cxx +local.properties diff --git a/android/SherpaOnnxTts/app/.gitignore b/android/SherpaOnnxTts/app/.gitignore new file mode 100644 index 00000000..42afabfd --- /dev/null +++ b/android/SherpaOnnxTts/app/.gitignore @@ -0,0 +1 @@ +/build \ No newline at end of file diff --git a/android/SherpaOnnxTts/app/build.gradle b/android/SherpaOnnxTts/app/build.gradle new file mode 100644 index 00000000..aa92a739 --- /dev/null +++ b/android/SherpaOnnxTts/app/build.gradle @@ -0,0 +1,44 @@ +plugins { + id 'com.android.application' + id 'org.jetbrains.kotlin.android' +} + +android { + namespace 'com.k2fsa.sherpa.onnx' + compileSdk 32 + + defaultConfig { + applicationId "com.k2fsa.sherpa.onnx" + minSdk 21 + targetSdk 32 + versionCode 1 + versionName "1.0" + + testInstrumentationRunner "androidx.test.runner.AndroidJUnitRunner" + } + + buildTypes { + release { + minifyEnabled false + proguardFiles getDefaultProguardFile('proguard-android-optimize.txt'), 'proguard-rules.pro' + } + } + compileOptions { + sourceCompatibility JavaVersion.VERSION_1_8 + targetCompatibility JavaVersion.VERSION_1_8 + } + kotlinOptions { + jvmTarget = '1.8' + } +} + +dependencies { + + implementation 'com.android.support.constraint:constraint-layout:1.1.3' + implementation 'androidx.core:core-ktx:1.7.0' + implementation 'com.google.android.material:material:1.9.0' + implementation 'androidx.constraintlayout:constraintlayout:2.1.4' + testImplementation 'junit:junit:4.13.2' + androidTestImplementation 'androidx.test.ext:junit:1.1.5' + androidTestImplementation 'androidx.test.espresso:espresso-core:3.5.1' +} \ No newline at end of file diff --git a/android/SherpaOnnxTts/app/proguard-rules.pro b/android/SherpaOnnxTts/app/proguard-rules.pro new file mode 100644 index 00000000..481bb434 --- /dev/null +++ b/android/SherpaOnnxTts/app/proguard-rules.pro @@ -0,0 +1,21 @@ +# Add project specific ProGuard rules here. +# You can control the set of applied configuration files using the +# proguardFiles setting in build.gradle. +# +# For more details, see +# http://developer.android.com/guide/developing/tools/proguard.html + +# If your project uses WebView with JS, uncomment the following +# and specify the fully qualified class name to the JavaScript interface +# class: +#-keepclassmembers class fqcn.of.javascript.interface.for.webview { +# public *; +#} + +# Uncomment this to preserve the line number information for +# debugging stack traces. +#-keepattributes SourceFile,LineNumberTable + +# If you keep the line number information, uncomment this to +# hide the original source file name. +#-renamesourcefileattribute SourceFile \ No newline at end of file diff --git a/android/SherpaOnnxTts/app/src/androidTest/java/com/k2fsa/sherpa/onnx/ExampleInstrumentedTest.kt b/android/SherpaOnnxTts/app/src/androidTest/java/com/k2fsa/sherpa/onnx/ExampleInstrumentedTest.kt new file mode 100644 index 00000000..18338320 --- /dev/null +++ b/android/SherpaOnnxTts/app/src/androidTest/java/com/k2fsa/sherpa/onnx/ExampleInstrumentedTest.kt @@ -0,0 +1,24 @@ +package com.k2fsa.sherpa.onnx + +import androidx.test.platform.app.InstrumentationRegistry +import androidx.test.ext.junit.runners.AndroidJUnit4 + +import org.junit.Test +import org.junit.runner.RunWith + +import org.junit.Assert.* + +/** + * Instrumented test, which will execute on an Android device. + * + * See [testing documentation](http://d.android.com/tools/testing). + */ +@RunWith(AndroidJUnit4::class) +class ExampleInstrumentedTest { + @Test + fun useAppContext() { + // Context of the app under test. + val appContext = InstrumentationRegistry.getInstrumentation().targetContext + assertEquals("com.k2fsa.sherpa.onnx", appContext.packageName) + } +} \ No newline at end of file diff --git a/android/SherpaOnnxTts/app/src/main/.gitignore b/android/SherpaOnnxTts/app/src/main/.gitignore new file mode 100644 index 00000000..3bd9e832 --- /dev/null +++ b/android/SherpaOnnxTts/app/src/main/.gitignore @@ -0,0 +1,2 @@ +vits-zh-aishell3 +vits-vctk diff --git a/android/SherpaOnnxTts/app/src/main/AndroidManifest.xml b/android/SherpaOnnxTts/app/src/main/AndroidManifest.xml new file mode 100644 index 00000000..584bbec5 --- /dev/null +++ b/android/SherpaOnnxTts/app/src/main/AndroidManifest.xml @@ -0,0 +1,30 @@ + + + + + + + + + + + + + + + + \ No newline at end of file diff --git a/android/SherpaOnnxTts/app/src/main/assets/.gitkeep b/android/SherpaOnnxTts/app/src/main/assets/.gitkeep new file mode 100644 index 00000000..e69de29b diff --git a/android/SherpaOnnxTts/app/src/main/java/com/k2fsa/sherpa/onnx/MainActivity.kt b/android/SherpaOnnxTts/app/src/main/java/com/k2fsa/sherpa/onnx/MainActivity.kt new file mode 100644 index 00000000..88f7e220 --- /dev/null +++ b/android/SherpaOnnxTts/app/src/main/java/com/k2fsa/sherpa/onnx/MainActivity.kt @@ -0,0 +1,113 @@ +package com.k2fsa.sherpa.onnx + +import android.media.MediaPlayer +import android.net.Uri +import android.os.Bundle +import android.util.Log +import android.widget.Button +import android.widget.EditText +import android.widget.Toast +import androidx.appcompat.app.AppCompatActivity +import java.io.File + +const val TAG = "sherpa-onnx" + +class MainActivity : AppCompatActivity() { + private lateinit var tts: OfflineTts + private lateinit var text: EditText + private lateinit var sid: EditText + private lateinit var speed: EditText + private lateinit var generate: Button + private lateinit var play: Button + private var hasFile: Boolean = false + + override fun onCreate(savedInstanceState: Bundle?) { + super.onCreate(savedInstanceState) + setContentView(R.layout.activity_main) + + Log.i(TAG, "Start to initialize TTS") + initTts() + Log.i(TAG, "Finish initializing TTS") + + text = findViewById(R.id.text) + sid = findViewById(R.id.sid) + speed = findViewById(R.id.speed) + + generate = findViewById(R.id.generate) + play = findViewById(R.id.play) + + generate.setOnClickListener { onClickGenerate() } + play.setOnClickListener { onClickPlay() } + + sid.setText("0") + speed.setText("1.0") + + // we will change sampleText here in the CI + val sampleText = "" + text.setText(sampleText) + + play.isEnabled = false; + } + + fun onClickGenerate() { + val sidInt = sid.text.toString().toIntOrNull() + if (sidInt == null || sidInt < 0) { + Toast.makeText( + applicationContext, + "Please input a non-negative integer for speaker ID!", + Toast.LENGTH_SHORT + ).show() + return + } + + val speedFloat = speed.text.toString().toFloatOrNull() + if (speedFloat == null || speedFloat <= 0) { + Toast.makeText( + applicationContext, + "Please input a positive number for speech speed!", + Toast.LENGTH_SHORT + ).show() + return + } + + val textStr = text.text.toString().trim() + if (textStr.isBlank() || textStr.isEmpty()) { + Toast.makeText(applicationContext, "Please input a non-empty text!", Toast.LENGTH_SHORT) + .show() + return + } + + Toast.makeText(applicationContext, "Generating...Please wait", Toast.LENGTH_LONG).show() + val audio = tts.generate(text = textStr, sid = sidInt, speed = speedFloat) + + val filename = application.filesDir.absolutePath + "/generated.wav" + val ok = audio.samples.size > 0 && audio.save(filename) + if (ok) { + play.isEnabled = true + Toast.makeText( + applicationContext, + "Generated! Please click play to listen to it", + Toast.LENGTH_LONG + ).show() + } else { + play.isEnabled = false + } + } + + fun onClickPlay() { + val filename = application.filesDir.absolutePath + "/generated.wav" + val mediaPlayer = MediaPlayer.create( + applicationContext, + Uri.fromFile(File(filename)) + ) + mediaPlayer.start() + } + + fun initTts() { + // 0 - vits-vctk (multi-speaker, English) + // 1 - vits-zh-aishell3 (multi-speaker, Chinese) + val type = 0 + val config = getOfflineTtsConfig(type = type, debug = true)!! + tts = OfflineTts(assetManager = application.assets, config = config) + } +} \ No newline at end of file diff --git a/android/SherpaOnnxTts/app/src/main/java/com/k2fsa/sherpa/onnx/Tts.kt b/android/SherpaOnnxTts/app/src/main/java/com/k2fsa/sherpa/onnx/Tts.kt new file mode 100644 index 00000000..748e7b73 --- /dev/null +++ b/android/SherpaOnnxTts/app/src/main/java/com/k2fsa/sherpa/onnx/Tts.kt @@ -0,0 +1,160 @@ +// 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 noiseScale: Float = 0.667f, + var noiseScaleW: Float = 0.8f, + var lengthScale: Float = 1.0f, +) + +data class OfflineTtsModelConfig( + var vits: OfflineTtsVitsModelConfig, + var numThreads: Int = 1, + var debug: Boolean = false, + var provider: String = "cpu", +) + +data class OfflineTtsConfig( + var model: OfflineTtsModelConfig, +) + +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 +} + +class OfflineTts( + assetManager: AssetManager? = null, + var config: OfflineTtsConfig, +) { + private var ptr: Long + + init { + if (assetManager != null) { + ptr = new(assetManager, config) + } else { + ptr = newFromFile(config) + } + } + + fun generate( + text: String, + sid: Int = 0, + speed: Float = 1.0f + ): GeneratedAudio { + var objArray = generateImpl(ptr, text = text, sid = sid, speed = speed) + return GeneratedAudio( + samples = objArray[0] as FloatArray, + sampleRate = objArray[1] as Int + ) + } + + fun allocate(assetManager: AssetManager? = null) { + if (ptr == 0L) { + if (assetManager != null) { + ptr = new(assetManager, config) + } else { + ptr = newFromFile(config) + } + } + } + + fun free() { + if (ptr != 0L) { + delete(ptr) + ptr = 0 + } + } + + protected fun finalize() { + delete(ptr) + } + + private external fun new( + assetManager: AssetManager, + config: OfflineTtsConfig, + ): Long + + private external fun newFromFile( + config: OfflineTtsConfig, + ): Long + + private external fun delete(ptr: Long) + + // 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 + external fun generateImpl( + ptr: Long, + text: String, + sid: Int = 0, + speed: Float = 1.0f + ): Array + + 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 +// +// You can change the type as you wish +fun getOfflineTtsConfig(type: Int, debug: Boolean = false): OfflineTtsConfig? { + when (type) { + 0 -> { + val modelDir = "vits-vctk" + return OfflineTtsConfig( + model = OfflineTtsModelConfig( + vits = OfflineTtsVitsModelConfig( + model = "$modelDir/vits-vctk.onnx", + lexicon = "$modelDir/lexicon.txt", + tokens = "$modelDir/tokens.txt" + ), + numThreads = 2, + debug = debug, + provider = "cpu", + ) + ) + } + + 1 -> { + val modelDir = "vits-zh-aishell3" + return OfflineTtsConfig( + model = OfflineTtsModelConfig( + vits = OfflineTtsVitsModelConfig( + model = "$modelDir/vits-aishell3.onnx", + lexicon = "$modelDir/lexicon.txt", + tokens = "$modelDir/tokens.txt" + ), + numThreads = 2, + debug = debug, + provider = "cpu", + ) + ) + } + } + + println("Unsupported type $type") + + return null + +} diff --git a/android/SherpaOnnxTts/app/src/main/jniLibs/arm64-v8a/.gitignore b/android/SherpaOnnxTts/app/src/main/jniLibs/arm64-v8a/.gitignore new file mode 100644 index 00000000..e69de29b diff --git a/android/SherpaOnnxTts/app/src/main/jniLibs/armeabi-v7a/.gitignore b/android/SherpaOnnxTts/app/src/main/jniLibs/armeabi-v7a/.gitignore new file mode 100644 index 00000000..e69de29b diff --git a/android/SherpaOnnxTts/app/src/main/jniLibs/x86/.gitignore b/android/SherpaOnnxTts/app/src/main/jniLibs/x86/.gitignore new file mode 100644 index 00000000..e69de29b diff --git a/android/SherpaOnnxTts/app/src/main/jniLibs/x86_64/.gitignore b/android/SherpaOnnxTts/app/src/main/jniLibs/x86_64/.gitignore new file mode 100644 index 00000000..e69de29b diff --git a/android/SherpaOnnxTts/app/src/main/res/drawable-v24/ic_launcher_foreground.xml b/android/SherpaOnnxTts/app/src/main/res/drawable-v24/ic_launcher_foreground.xml new file mode 100644 index 00000000..2b068d11 --- /dev/null +++ b/android/SherpaOnnxTts/app/src/main/res/drawable-v24/ic_launcher_foreground.xml @@ -0,0 +1,30 @@ + + + + + + + + + + + \ No newline at end of file diff --git a/android/SherpaOnnxTts/app/src/main/res/drawable/ic_launcher_background.xml b/android/SherpaOnnxTts/app/src/main/res/drawable/ic_launcher_background.xml new file mode 100644 index 00000000..07d5da9c --- /dev/null +++ b/android/SherpaOnnxTts/app/src/main/res/drawable/ic_launcher_background.xml @@ -0,0 +1,170 @@ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + diff --git a/android/SherpaOnnxTts/app/src/main/res/layout/activity_main.xml b/android/SherpaOnnxTts/app/src/main/res/layout/activity_main.xml new file mode 100644 index 00000000..3547de87 --- /dev/null +++ b/android/SherpaOnnxTts/app/src/main/res/layout/activity_main.xml @@ -0,0 +1,87 @@ + + + + + + + + + + + +