Use aar in Android Java demo. (#1616)

android/SherpaOnnxJavaDemo/README.md

@@ -0,0 +1,44 @@
+# Introduction
+
+Please run the following commands to download model files before you run this Android demo:
+
+```bash
+# Assume we are inside
+# /Users/fangjun/open-source/sherpa-onnx/android/SherpaOnnxJavaDemo
+
+cd app/src/main/assets/
+wget https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/sherpa-onnx-streaming-zipformer-bilingual-zh-en-2023-02-20.tar.bz2
+
+tar xvf sherpa-onnx-streaming-zipformer-bilingual-zh-en-2023-02-20.tar.bz2
+rm sherpa-onnx-streaming-zipformer-bilingual-zh-en-2023-02-20.tar.bz2
+
+mv sherpa-onnx-streaming-zipformer-bilingual-zh-en-2023-02-20/encoder-epoch-99-avg-1.int8.onnx ./
+mv sherpa-onnx-streaming-zipformer-bilingual-zh-en-2023-02-20/decoder-epoch-99-avg-1.onnx ./
+mv sherpa-onnx-streaming-zipformer-bilingual-zh-en-2023-02-20/joiner-epoch-99-avg-1.int8.onnx ./
+mv sherpa-onnx-streaming-zipformer-bilingual-zh-en-2023-02-20/tokens.txt ./
+
+rm -rf sherpa-onnx-streaming-zipformer-bilingual-zh-en-2023-02-20/*
+
+mv encoder-epoch-99-avg-1.int8.onnx ./
+mv decoder-epoch-99-avg-1.onnx ./
+mv joiner-epoch-99-avg-1.int8.onnx ./
+mv tokens.txt ./
+```
+
+You should have the following directory structure:
+```
+(py38) fangjuns-MacBook-Pro:assets fangjun$ pwd
+/Users/fangjun/open-source/sherpa-onnx/android/SherpaOnnxJavaDemo/app/src/main/assets
+
+(py38) fangjuns-MacBook-Pro:assets fangjun$ tree .
+.
+└── sherpa-onnx-streaming-zipformer-bilingual-zh-en-2023-02-20
+    ├── decoder-epoch-99-avg-1.onnx
+    ├── encoder-epoch-99-avg-1.int8.onnx
+    ├── joiner-epoch-99-avg-1.int8.onnx
+    └── tokens.txt
+
+1 directory, 4 files
+```
+
+Remember to remove unused files to reduce the file size of the final APK.

android/SherpaOnnxJavaDemo/app/build.gradle

@@ -8,7 +8,7 @@ android {
     defaultConfig {
         applicationId "com.k2fsa.sherpa.onnx"
         minSdk 28
-        targetSdk 32
+        targetSdk 34
         versionCode 1
         versionName "1.0"
 
@@ -25,17 +25,14 @@ android {
         sourceCompatibility JavaVersion.VERSION_1_8
         targetCompatibility JavaVersion.VERSION_1_8
     }
-    sourceSets.main{
-        jniLibs.srcDirs = ['jniLibs']
-    }
 }
 
 dependencies {
-
     implementation 'androidx.appcompat:appcompat:1.3.1'
     implementation 'com.google.android.material:material:1.3.0'
     implementation 'androidx.constraintlayout:constraintlayout:1.1.3'
     implementation 'pub.devrel:easypermissions:3.0.0'
-    implementation project(path: ':sherpa')
+    implementation 'androidx.core:core-ktx:1.7.0'
-
+    // implementation files('/Users/fangjun/open-source/sherpa-onnx/android/SherpaOnnxAar/sherpa_onnx/build/outputs/aar/sherpa_onnx-release.aar')
+    implementation 'com.github.k2-fsa:sherpa-onnx:master-SNAPSHOT'
 }

android/SherpaOnnxJavaDemo/app/src/main/java/com/k2fsa/sherpa/onnx/OnlineRecognizer.java

@@ -1,84 +0,0 @@
-// Copyright 2022-2023 by zhaoming
-// Copyright 2024 Xiaomi Corporation
-
-package com.k2fsa.sherpa.onnx;
-
-import android.content.res.AssetManager;
-
-public class OnlineRecognizer {
-    static {
-        System.loadLibrary("sherpa-onnx-jni");
-    }
-
-    private long ptr = 0;
-
-    public OnlineRecognizer(OnlineRecognizerConfig config) {
-        ptr = newFromFile(config);
-    }
-
-    public OnlineRecognizer(AssetManager assetManager, OnlineRecognizerConfig config) {
-        ptr = newFromAsset(assetManager, config);
-    }
-
-    public void decode(OnlineStream s) {
-        decode(ptr, s.getPtr());
-    }
-
-    public boolean isReady(OnlineStream s) {
-        return isReady(ptr, s.getPtr());
-    }
-
-    public boolean isEndpoint(OnlineStream s) {
-        return isEndpoint(ptr, s.getPtr());
-    }
-
-    public void reset(OnlineStream s) {
-        reset(ptr, s.getPtr());
-    }
-
-    public OnlineStream createStream() {
-        long p = createStream(ptr, "");
-        return new OnlineStream(p);
-    }
-
-    @Override
-    protected void finalize() throws Throwable {
-        release();
-    }
-
-    // You'd better call it manually if it is not used anymore
-    public void release() {
-        if (this.ptr == 0) {
-            return;
-        }
-        delete(this.ptr);
-        this.ptr = 0;
-    }
-
-    public OnlineRecognizerResult getResult(OnlineStream s) {
-        Object[] arr = getResult(ptr, s.getPtr());
-        String text = (String) arr[0];
-        String[] tokens = (String[]) arr[1];
-        float[] timestamps = (float[]) arr[2];
-        return new OnlineRecognizerResult(text, tokens, timestamps);
-    }
-
-
-    private native void delete(long ptr);
-
-    private native long newFromFile(OnlineRecognizerConfig config);
-
-    private native long newFromAsset(AssetManager assetManager, OnlineRecognizerConfig config);
-
-    private native long createStream(long ptr, String hotwords);
-
-    private native void reset(long ptr, long streamPtr);
-
-    private native void decode(long ptr, long streamPtr);
-
-    private native boolean isEndpoint(long ptr, long streamPtr);
-
-    private native boolean isReady(long ptr, long streamPtr);
-
-    private native Object[] getResult(long ptr, long streamPtr);
-}

android/SherpaOnnxJavaDemo/app/src/main/java/com/k2fsa/sherpa/onnx/service/SpeechSherpaRecognitionService.java

@@ -1,6 +1,7 @@
 package com.k2fsa.sherpa.onnx.service;
 
 import android.Manifest;
+import android.annotation.SuppressLint;
 import android.app.Notification;
 import android.app.NotificationChannel;
 import android.app.NotificationManager;
@@ -67,6 +68,16 @@ public class SpeechSherpaRecognitionService extends Service {
         appViewModel = Application.getInstance().getViewModel();
         int numBytes = AudioRecord.getMinBufferSize(sampleRateInHz, channelConfig, audioFormat);
 
+        if (ActivityCompat.checkSelfPermission(this, Manifest.permission.RECORD_AUDIO) != PackageManager.PERMISSION_GRANTED) {
+            // TODO: Consider calling
+            //    ActivityCompat#requestPermissions
+            // here to request the missing permissions, and then overriding
+            //   public void onRequestPermissionsResult(int requestCode, String[] permissions,
+            //                                          int[] grantResults)
+            // to handle the case where the user grants the permission. See the documentation
+            // for ActivityCompat#requestPermissions for more details.
+            return;
+        }
         audioRecord = new AudioRecord(
                 audioSource,
                 sampleRateInHz,
@@ -81,22 +92,21 @@ public class SpeechSherpaRecognitionService extends Service {
 
     private void initializeSherpa() {
         Log.d("Current Directory", System.getProperty("user.dir"));
-        String modelDir = "sherpa-onnx-streaming-zipformer-zh-14M-2023-02-23";
+        String modelDir = "sherpa-onnx-streaming-zipformer-bilingual-zh-en-2023-02-20";
         initializeSherpaDir(modelDir, modelDir);
-        OnlineTransducerModelConfig onlineTransducerModelConfig = OnlineTransducerModelConfig.builder()
+        OnlineTransducerModelConfig onlineTransducerModelConfig = new OnlineTransducerModelConfig();
-                .setEncoder(modelDir + "/encoder-epoch-99-avg-1.int8.onnx")
+        onlineTransducerModelConfig.setEncoder(modelDir + "/encoder-epoch-99-avg-1.int8.onnx");
-                .setDecoder(modelDir + "/decoder-epoch-99-avg-1.onnx")
+        onlineTransducerModelConfig.setDecoder(modelDir + "/decoder-epoch-99-avg-1.onnx");
-                .setJoiner(modelDir + "/joiner-epoch-99-avg-1.int8.onnx")
+        onlineTransducerModelConfig.setJoiner(modelDir + "/joiner-epoch-99-avg-1.int8.onnx");
-                .build();
 
-        OnlineModelConfig onlineModelConfig = OnlineModelConfig.builder()
+        OnlineModelConfig onlineModelConfig = new OnlineModelConfig();
-                .setTransducer(onlineTransducerModelConfig)
+        onlineModelConfig.setTransducer(onlineTransducerModelConfig);
-                .setTokens(modelDir + "/tokens.txt")
+        onlineModelConfig.setTokens(modelDir + "/tokens.txt");
-                .setModelType("zipformer")
+        onlineModelConfig.setModelType("zipformer");
-                .build();
+        onlineModelConfig.setDebug(true);
-        OnlineRecognizerConfig config = OnlineRecognizerConfig.builder()
+
-                .setOnlineModelConfig(onlineModelConfig)
+        OnlineRecognizerConfig config = new OnlineRecognizerConfig();
-                .build();
+        config.setModelConfig(onlineModelConfig);
         recognizer = new OnlineRecognizer(getAssets(), config);
 
         audioRecord.startRecording();
@@ -110,7 +120,7 @@ public class SpeechSherpaRecognitionService extends Service {
     }
 
     private void processSamples() {
-        OnlineStream stream = recognizer.createStream();
+        OnlineStream stream = recognizer.createStream("");
         double interval = 0.1;
         int bufferSize = (int) (interval * sampleRateInHz);
         short[] buffer = new short[bufferSize];
@@ -182,6 +192,7 @@ public class SpeechSherpaRecognitionService extends Service {
     }
 
 
+    @SuppressLint("ForegroundServiceType")
     private void startForegroundService() {
         String channelId = createNotificationChannel();
 

android/SherpaOnnxJavaDemo/settings.gradle

@@ -10,8 +10,8 @@ dependencyResolutionManagement {
     repositories {
         google()
         mavenCentral()
+        maven { url 'https://jitpack.io' }
     }
 }
 rootProject.name = "SherpaOnnxJavaDemo"
 include ':app'
-include ':sherpa'

android/SherpaOnnxJavaDemo/sherpa/.gitignore

@@ -1 +0,0 @@
-/build

android/SherpaOnnxJavaDemo/sherpa/build.gradle

@@ -1,43 +0,0 @@
-plugins {
-    id 'com.android.library'
-}
-
-android {
-    namespace 'com.k2fsa.sherpa'
-    compileSdk 34
-
-    defaultConfig {
-        minSdk 26
-        targetSdk 27
-        versionCode 1
-        versionName "1.0"
-        missingDimensionStrategy 'base', 'feature1'
-        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
-    }
-}
-
-dependencies {
-
-//    implementation "androidx.appcompat"
-//    implementation libs.material
-//    testImplementation libs.junit
-//    androidTestImplementation libs.androidx.test.ext.junit
-//    androidTestImplementation libs.espresso.core
-    implementation 'androidx.appcompat:appcompat:1.6.1'
-    implementation 'com.google.android.material:material:1.9.0'
-    testImplementation 'junit:junit:4.13.2'
-    androidTestImplementation 'androidx.test.ext:junit:1.1.5'
-    androidTestImplementation 'androidx.test.espresso:espresso-core:3.5.1'
-
-}

android/SherpaOnnxJavaDemo/sherpa/proguard-rules.pro

@@ -1,21 +0,0 @@
-# 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

android/SherpaOnnxJavaDemo/sherpa/src/main/AndroidManifest.xml

@@ -1,9 +0,0 @@
-<?xml version="1.0" encoding="utf-8"?>
-<manifest xmlns:android="http://schemas.android.com/apk/res/android">
-
-    <application
-        android:allowBackup="true"
-        android:label="@string/app_name"
-        android:supportsRtl="true" />
-
-</manifest>

android/SherpaOnnxJavaDemo/sherpa/src/main/java/com

@@ -1 +0,0 @@
-../../../../../../../sherpa-onnx/sherpa-onnx/java-api/src/com

android/SherpaOnnxJavaDemo/sherpa/src/main/res/values/strings.xml

@@ -1,3 +0,0 @@
-<resources>
-    <string name="app_name">sherpa</string>
-</resources>

sherpa-onnx/kotlin-api/AudioTagging.kt

@@ -15,8 +15,8 @@ data class AudioTaggingModelConfig(
 )
 
 data class AudioTaggingConfig(
-    var model: AudioTaggingModelConfig,
+    var model: AudioTaggingModelConfig = AudioTaggingModelConfig(),
-    var labels: String,
+    var labels: String = "",
     var topK: Int = 5,
 )
 

sherpa-onnx/kotlin-api/KeywordSpotter.kt

@@ -5,7 +5,7 @@ import android.content.res.AssetManager
 
 data class KeywordSpotterConfig(
     var featConfig: FeatureConfig = FeatureConfig(),
-    var modelConfig: OnlineModelConfig,
+    var modelConfig: OnlineModelConfig = OnlineModelConfig(),
     var maxActivePaths: Int = 4,
     var keywordsFile: String = "keywords.txt",
     var keywordsScore: Float = 1.5f,

sherpa-onnx/kotlin-api/OfflinePunctuation.kt

@@ -3,7 +3,7 @@ package com.k2fsa.sherpa.onnx
 import android.content.res.AssetManager
 
 data class OfflinePunctuationModelConfig(
-    var ctTransformer: String,
+    var ctTransformer: String = "",
     var numThreads: Int = 1,
     var debug: Boolean = false,
     var provider: String = "cpu",

sherpa-onnx/kotlin-api/OfflineRecognizer.kt

@@ -58,14 +58,14 @@ data class OfflineModelConfig(
     var debug: Boolean = false,
     var provider: String = "cpu",
     var modelType: String = "",
-    var tokens: String,
+    var tokens: String = "",
     var modelingUnit: String = "",
     var bpeVocab: String = "",
 )
 
 data class OfflineRecognizerConfig(
     var featConfig: FeatureConfig = FeatureConfig(),
-    var modelConfig: OfflineModelConfig,
+    var modelConfig: OfflineModelConfig = OfflineModelConfig(),
     // var lmConfig: OfflineLMConfig(), // TODO(fangjun): enable it
     var decodingMethod: String = "greedy_search",
     var maxActivePaths: Int = 4,

sherpa-onnx/kotlin-api/OfflineSpeakerDiarization.kt

@@ -3,11 +3,11 @@ package com.k2fsa.sherpa.onnx
 import android.content.res.AssetManager
 
 data class OfflineSpeakerSegmentationPyannoteModelConfig(
-    var model: String,
+    var model: String = "",
 )
 
 data class OfflineSpeakerSegmentationModelConfig(
-    var pyannote: OfflineSpeakerSegmentationPyannoteModelConfig,
+    var pyannote: OfflineSpeakerSegmentationPyannoteModelConfig = OfflineSpeakerSegmentationPyannoteModelConfig(),
     var numThreads: Int = 1,
     var debug: Boolean = false,
     var provider: String = "cpu",
@@ -19,9 +19,9 @@ data class FastClusteringConfig(
 )
 
 data class OfflineSpeakerDiarizationConfig(
-    var segmentation: OfflineSpeakerSegmentationModelConfig,
+    var segmentation: OfflineSpeakerSegmentationModelConfig = OfflineSpeakerSegmentationModelConfig(),
-    var embedding: SpeakerEmbeddingExtractorConfig,
+    var embedding: SpeakerEmbeddingExtractorConfig = SpeakerEmbeddingExtractorConfig(),
-    var clustering: FastClusteringConfig,
+    var clustering: FastClusteringConfig = FastClusteringConfig(),
     var minDurationOn: Float = 0.2f,
     var minDurationOff: Float = 0.5f,
 )

sherpa-onnx/kotlin-api/OnlineRecognizer.kt

@@ -38,7 +38,7 @@ data class OnlineModelConfig(
     var paraformer: OnlineParaformerModelConfig = OnlineParaformerModelConfig(),
     var zipformer2Ctc: OnlineZipformer2CtcModelConfig = OnlineZipformer2CtcModelConfig(),
     var neMoCtc: OnlineNeMoCtcModelConfig = OnlineNeMoCtcModelConfig(),
-    var tokens: String,
+    var tokens: String = "",
     var numThreads: Int = 1,
     var debug: Boolean = false,
     var provider: String = "cpu",
@@ -60,7 +60,7 @@ data class OnlineCtcFstDecoderConfig(
 
 data class OnlineRecognizerConfig(
     var featConfig: FeatureConfig = FeatureConfig(),
-    var modelConfig: OnlineModelConfig,
+    var modelConfig: OnlineModelConfig = OnlineModelConfig(),
     var lmConfig: OnlineLMConfig = OnlineLMConfig(),
     var ctcFstDecoderConfig: OnlineCtcFstDecoderConfig = OnlineCtcFstDecoderConfig(),
     var endpointConfig: EndpointConfig = EndpointConfig(),

sherpa-onnx/kotlin-api/SpeakerEmbeddingExtractorConfig.kt

@@ -1,7 +1,7 @@
 package com.k2fsa.sherpa.onnx
 
 data class SpeakerEmbeddingExtractorConfig(
-    val model: String,
+    val model: String = "",
     var numThreads: Int = 1,
     var debug: Boolean = false,
     var provider: String = "cpu",

sherpa-onnx/kotlin-api/SpokenLanguageIdentification.kt

@@ -3,13 +3,13 @@ package com.k2fsa.sherpa.onnx
 import android.content.res.AssetManager
 
 data class SpokenLanguageIdentificationWhisperConfig(
-    var encoder: String,
+    var encoder: String = "",
-    var decoder: String,
+    var decoder: String = "",
     var tailPaddings: Int = -1,
 )
 
 data class SpokenLanguageIdentificationConfig(
-    var whisper: SpokenLanguageIdentificationWhisperConfig,
+    var whisper: SpokenLanguageIdentificationWhisperConfig = SpokenLanguageIdentificationWhisperConfig(),
     var numThreads: Int = 1,
     var debug: Boolean = false,
     var provider: String = "cpu",

sherpa-onnx/kotlin-api/Tts.kt

@@ -4,9 +4,9 @@ package com.k2fsa.sherpa.onnx
 import android.content.res.AssetManager
 
 data class OfflineTtsVitsModelConfig(
-    var model: String,
+    var model: String = "",
     var lexicon: String = "",
-    var tokens: String,
+    var tokens: String = "",
     var dataDir: String = "",
     var dictDir: String = "",
     var noiseScale: Float = 0.667f,
@@ -15,14 +15,14 @@ data class OfflineTtsVitsModelConfig(
 )
 
 data class OfflineTtsModelConfig(
-    var vits: OfflineTtsVitsModelConfig,
+    var vits: OfflineTtsVitsModelConfig = OfflineTtsVitsModelConfig(),
     var numThreads: Int = 1,
     var debug: Boolean = false,
     var provider: String = "cpu",
 )
 
 data class OfflineTtsConfig(
-    var model: OfflineTtsModelConfig,
+    var model: OfflineTtsModelConfig = OfflineTtsModelConfig(),
     var ruleFsts: String = "",
     var ruleFars: String = "",
     var maxNumSentences: Int = 1,

sherpa-onnx/kotlin-api/Vad.kt

@@ -4,7 +4,7 @@ package com.k2fsa.sherpa.onnx
 import android.content.res.AssetManager
 
 data class SileroVadModelConfig(
-    var model: String,
+    var model: String = "",
     var threshold: Float = 0.5F,
     var minSilenceDuration: Float = 0.25F,
     var minSpeechDuration: Float = 0.25F,
@@ -13,7 +13,7 @@ data class SileroVadModelConfig(
 )
 
 data class VadModelConfig(
-    var sileroVadModelConfig: SileroVadModelConfig,
+    var sileroVadModelConfig: SileroVadModelConfig = SileroVadModelConfig(),
     var sampleRate: Int = 16000,
     var numThreads: Int = 1,
     var provider: String = "cpu",
@@ -112,5 +112,5 @@ fun getVadModelConfig(type: Int): VadModelConfig? {
             )
         }
     }
-    return null;
+    return null
 }