Add Java API for text-to-speech (#811)

2024-04-26 09:26:39 +08:00
parent fa2429920f
commit 15772d2150
15 changed files with 692 additions and 0 deletions
--- a/java-api-examples/NonStreamingTtsCoquiDe.java
+++ b/java-api-examples/NonStreamingTtsCoquiDe.java
@@ -0,0 +1,50 @@
+// Copyright 2024 Xiaomi Corporation
+
+// This file shows how to use a Coqui-ai VITS German TTS model
+// to convert text to speech
+import com.k2fsa.sherpa.onnx.*;
+
+public class NonStreamingTtsCoquiDe {
+  public static void main(String[] args) {
+    // please visit
+    // https://github.com/k2-fsa/sherpa-onnx/releases/tag/tts-models
+    // to download model files
+    String model = "./vits-coqui-de-css10/model.onnx";
+    String tokens = "./vits-coqui-de-css10/tokens.txt";
+    String text = "Alles hat ein Ende, nur die Wurst hat zwei.";
+
+    OfflineTtsVitsModelConfig vitsModelConfig =
+        OfflineTtsVitsModelConfig.builder().setModel(model).setTokens(tokens).build();
+
+    OfflineTtsModelConfig modelConfig =
+        OfflineTtsModelConfig.builder()
+            .setVits(vitsModelConfig)
+            .setNumThreads(1)
+            .setDebug(true)
+            .build();
+
+    OfflineTtsConfig config = OfflineTtsConfig.builder().setModel(modelConfig).build();
+    OfflineTts tts = new OfflineTts(config);
+
+    int sid = 0;
+    float speed = 1.0f;
+    long start = System.currentTimeMillis();
+    GeneratedAudio audio = tts.generate(text, sid, speed);
+    long stop = System.currentTimeMillis();
+
+    float timeElapsedSeconds = (stop - start) / 1000.0f;
+
+    float audioDuration = audio.getSamples().length / (float) audio.getSampleRate();
+    float real_time_factor = timeElapsedSeconds / audioDuration;
+
+    String waveFilename = "tts-coqui-de.wav";
+    audio.save(waveFilename);
+    System.out.printf("-- elapsed : %.3f seconds\n", timeElapsedSeconds);
+    System.out.printf("-- audio duration: %.3f seconds\n", timeElapsedSeconds);
+    System.out.printf("-- real-time factor (RTF): %.3f\n", real_time_factor);
+    System.out.printf("-- text: %s\n", text);
+    System.out.printf("-- Saved to %s\n", waveFilename);
+
+    tts.release();
+  }
+}
--- a/java-api-examples/NonStreamingTtsPiperEn.java
+++ b/java-api-examples/NonStreamingTtsPiperEn.java
@@ -0,0 +1,58 @@
+// Copyright 2024 Xiaomi Corporation
+
+// This file shows how to use a piper VITS English TTS model
+// to convert text to speech
+import com.k2fsa.sherpa.onnx.*;
+
+public class NonStreamingTtsPiperEn {
+  public static void main(String[] args) {
+    // please visit
+    // https://github.com/k2-fsa/sherpa-onnx/releases/tag/tts-models
+    // to download model files
+    String model = "./vits-piper-en_GB-cori-medium/en_GB-cori-medium.onnx";
+    String tokens = "./vits-piper-en_GB-cori-medium/tokens.txt";
+    String dataDir = "./vits-piper-en_GB-cori-medium/espeak-ng-data";
+    String 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.";
+
+    OfflineTtsVitsModelConfig vitsModelConfig =
+        OfflineTtsVitsModelConfig.builder()
+            .setModel(model)
+            .setTokens(tokens)
+            .setDataDir(dataDir)
+            .build();
+
+    OfflineTtsModelConfig modelConfig =
+        OfflineTtsModelConfig.builder()
+            .setVits(vitsModelConfig)
+            .setNumThreads(1)
+            .setDebug(true)
+            .build();
+
+    OfflineTtsConfig config = OfflineTtsConfig.builder().setModel(modelConfig).build();
+    OfflineTts tts = new OfflineTts(config);
+
+    int sid = 0;
+    float speed = 1.0f;
+    long start = System.currentTimeMillis();
+    GeneratedAudio audio = tts.generate(text, sid, speed);
+    long stop = System.currentTimeMillis();
+
+    float timeElapsedSeconds = (stop - start) / 1000.0f;
+
+    float audioDuration = audio.getSamples().length / (float) audio.getSampleRate();
+    float real_time_factor = timeElapsedSeconds / audioDuration;
+
+    String waveFilename = "tts-piper-en.wav";
+    audio.save(waveFilename);
+    System.out.printf("-- elapsed : %.3f seconds\n", timeElapsedSeconds);
+    System.out.printf("-- audio duration: %.3f seconds\n", timeElapsedSeconds);
+    System.out.printf("-- real-time factor (RTF): %.3f\n", real_time_factor);
+    System.out.printf("-- text: %s\n", text);
+    System.out.printf("-- Saved to %s\n", waveFilename);
+
+    tts.release();
+  }
+}
--- a/java-api-examples/NonStreamingTtsVitsZh.java
+++ b/java-api-examples/NonStreamingTtsVitsZh.java
@@ -0,0 +1,64 @@
+// Copyright 2024 Xiaomi Corporation
+
+// This file shows how to use a VITS Chinese TTS model
+// to convert text to speech.
+//
+// You can use https://github.com/Plachtaa/VITS-fast-fine-tuning
+// to train your model
+import com.k2fsa.sherpa.onnx.*;
+
+public class NonStreamingTtsPiperEn {
+  public static void main(String[] args) {
+    // please visit
+    // https://github.com/k2-fsa/sherpa-onnx/releases/tag/tts-models
+    // to download model files
+    String model = "./vits-zh-hf-fanchen-C/vits-zh-hf-fanchen-C.onnx";
+    String tokens = "./vits-zh-hf-fanchen-C/tokens.txt";
+    String lexicon = "./vits-zh-hf-fanchen-C/lexicon.txt";
+    String dictDir = "./vits-zh-hf-fanchen-C/dict";
+    String ruleFsts =
+        "./vits-zh-hf-fanchen-C/phone.fst,./vits-zh-hf-fanchen-C/date.fst,./vits-zh-hf-fanchen-C/number.fst";
+    String text = "有问题，请拨打110或者手机18601239876。我们的价值观是真诚热爱！";
+
+    OfflineTtsVitsModelConfig vitsModelConfig =
+        OfflineTtsVitsModelConfig.builder()
+            .setModel(model)
+            .setTokens(tokens)
+            .setLexicon(lexicon)
+            .setDictDir(dictDir)
+            .build();
+
+    OfflineTtsModelConfig modelConfig =
+        OfflineTtsModelConfig.builder()
+            .setVits(vitsModelConfig)
+            .setNumThreads(1)
+            .setDebug(true)
+            .build();
+
+    OfflineTtsConfig config =
+        OfflineTtsConfig.builder().setModel(modelConfig).setRuleFsts(ruleFsts).build();
+
+    OfflineTts tts = new OfflineTts(config);
+
+    int sid = 100;
+    float speed = 1.0f;
+    long start = System.currentTimeMillis();
+    GeneratedAudio audio = tts.generate(text, sid, speed);
+    long stop = System.currentTimeMillis();
+
+    float timeElapsedSeconds = (stop - start) / 1000.0f;
+
+    float audioDuration = audio.getSamples().length / (float) audio.getSampleRate();
+    float real_time_factor = timeElapsedSeconds / audioDuration;
+
+    String waveFilename = "tts-vits-zh.wav";
+    audio.save(waveFilename);
+    System.out.printf("-- elapsed : %.3f seconds\n", timeElapsedSeconds);
+    System.out.printf("-- audio duration: %.3f seconds\n", timeElapsedSeconds);
+    System.out.printf("-- real-time factor (RTF): %.3f\n", real_time_factor);
+    System.out.printf("-- text: %s\n", text);
+    System.out.printf("-- Saved to %s\n", waveFilename);
+
+    tts.release();
+  }
+}
--- a/java-api-examples/README.md
+++ b/java-api-examples/README.md
@@ -21,3 +21,11 @@ This directory contains examples for the JAVA API of sherpa-onnx.
 ./run-non-streaming-decode-file-whisper.sh
 ./run-non-streaming-decode-file-nemo.sh
 ```
+
+## Non-Streaming text-to-speech
+
+```bash
+./run-non-streaming-tts-piper-en.sh
+./run-non-streaming-tts-coqui-de.sh
+./run-non-streaming-tts-vits-zh.sh
+```
--- a/java-api-examples/run-non-streaming-tts-coqui-de.sh
+++ b/java-api-examples/run-non-streaming-tts-coqui-de.sh
@@ -0,0 +1,54 @@
+#!/usr/bin/env bash
+
+set -ex
+
+if [[ ! -f ../build/lib/libsherpa-onnx-jni.dylib  && ! -f ../build/lib/libsherpa-onnx-jni.so ]]; then
+  mkdir -p ../build
+  pushd ../build
+  cmake \
+    -DSHERPA_ONNX_ENABLE_PYTHON=OFF \
+    -DSHERPA_ONNX_ENABLE_TESTS=OFF \
+    -DSHERPA_ONNX_ENABLE_CHECK=OFF \
+    -DBUILD_SHARED_LIBS=ON \
+    -DSHERPA_ONNX_ENABLE_PORTAUDIO=OFF \
+    -DSHERPA_ONNX_ENABLE_JNI=ON \
+    ..
+
+  make -j4
+  ls -lh lib
+  popd
+fi
+
+if [ ! -f ../sherpa-onnx/java-api/build/sherpa-onnx.jar ]; then
+  pushd ../sherpa-onnx/java-api
+  make
+  popd
+fi
+
+if [[ ! -f ../build/lib/libsherpa-onnx-jni.dylib  && ! -f ../build/lib/libsherpa-onnx-jni.so ]]; then
+  cmake \
+    -DSHERPA_ONNX_ENABLE_PYTHON=OFF \
+    -DSHERPA_ONNX_ENABLE_TESTS=OFF \
+    -DSHERPA_ONNX_ENABLE_CHECK=OFF \
+    -DBUILD_SHARED_LIBS=ON \
+    -DSHERPA_ONNX_ENABLE_PORTAUDIO=OFF \
+    -DSHERPA_ONNX_ENABLE_JNI=ON \
+    ..
+
+  make -j4
+  ls -lh lib
+fi
+
+# please visit
+# https://github.com/k2-fsa/sherpa-onnx/releases/tag/tts-models
+# to download more models
+if [ ! -f ./vits-coqui-de-css10/tokens.txt ]; then
+  curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/tts-models/vits-coqui-de-css10.tar.bz2
+  tar xf vits-coqui-de-css10.tar.bz2
+  rm vits-coqui-de-css10.tar.bz2
+fi
+
+java \
+  -Djava.library.path=$PWD/../build/lib \
+  -cp ../sherpa-onnx/java-api/build/sherpa-onnx.jar \
+  NonStreamingTtsCoquiDe.java
--- a/java-api-examples/run-non-streaming-tts-piper-en.sh
+++ b/java-api-examples/run-non-streaming-tts-piper-en.sh
@@ -0,0 +1,54 @@
+#!/usr/bin/env bash
+
+set -ex
+
+if [[ ! -f ../build/lib/libsherpa-onnx-jni.dylib  && ! -f ../build/lib/libsherpa-onnx-jni.so ]]; then
+  mkdir -p ../build
+  pushd ../build
+  cmake \
+    -DSHERPA_ONNX_ENABLE_PYTHON=OFF \
+    -DSHERPA_ONNX_ENABLE_TESTS=OFF \
+    -DSHERPA_ONNX_ENABLE_CHECK=OFF \
+    -DBUILD_SHARED_LIBS=ON \
+    -DSHERPA_ONNX_ENABLE_PORTAUDIO=OFF \
+    -DSHERPA_ONNX_ENABLE_JNI=ON \
+    ..
+
+  make -j4
+  ls -lh lib
+  popd
+fi
+
+if [ ! -f ../sherpa-onnx/java-api/build/sherpa-onnx.jar ]; then
+  pushd ../sherpa-onnx/java-api
+  make
+  popd
+fi
+
+if [[ ! -f ../build/lib/libsherpa-onnx-jni.dylib  && ! -f ../build/lib/libsherpa-onnx-jni.so ]]; then
+  cmake \
+    -DSHERPA_ONNX_ENABLE_PYTHON=OFF \
+    -DSHERPA_ONNX_ENABLE_TESTS=OFF \
+    -DSHERPA_ONNX_ENABLE_CHECK=OFF \
+    -DBUILD_SHARED_LIBS=ON \
+    -DSHERPA_ONNX_ENABLE_PORTAUDIO=OFF \
+    -DSHERPA_ONNX_ENABLE_JNI=ON \
+    ..
+
+  make -j4
+  ls -lh lib
+fi
+
+# please visit
+# https://github.com/k2-fsa/sherpa-onnx/releases/tag/tts-models
+# to download more models
+if [ ! -f ./vits-piper-en_GB-cori-medium/tokens.txt ]; then
+  curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/tts-models/vits-piper-en_GB-cori-medium.tar.bz2
+  tar xf vits-piper-en_GB-cori-medium.tar.bz2
+  rm vits-piper-en_GB-cori-medium.tar.bz2
+fi
+
+java \
+  -Djava.library.path=$PWD/../build/lib \
+  -cp ../sherpa-onnx/java-api/build/sherpa-onnx.jar \
+  NonStreamingTtsPiperEn.java
--- a/java-api-examples/run-non-streaming-tts-vits-zh.sh
+++ b/java-api-examples/run-non-streaming-tts-vits-zh.sh
@@ -0,0 +1,54 @@
+#!/usr/bin/env bash
+
+set -ex
+
+if [[ ! -f ../build/lib/libsherpa-onnx-jni.dylib  && ! -f ../build/lib/libsherpa-onnx-jni.so ]]; then
+  mkdir -p ../build
+  pushd ../build
+  cmake \
+    -DSHERPA_ONNX_ENABLE_PYTHON=OFF \
+    -DSHERPA_ONNX_ENABLE_TESTS=OFF \
+    -DSHERPA_ONNX_ENABLE_CHECK=OFF \
+    -DBUILD_SHARED_LIBS=ON \
+    -DSHERPA_ONNX_ENABLE_PORTAUDIO=OFF \
+    -DSHERPA_ONNX_ENABLE_JNI=ON \
+    ..
+
+  make -j4
+  ls -lh lib
+  popd
+fi
+
+if [ ! -f ../sherpa-onnx/java-api/build/sherpa-onnx.jar ]; then
+  pushd ../sherpa-onnx/java-api
+  make
+  popd
+fi
+
+if [[ ! -f ../build/lib/libsherpa-onnx-jni.dylib  && ! -f ../build/lib/libsherpa-onnx-jni.so ]]; then
+  cmake \
+    -DSHERPA_ONNX_ENABLE_PYTHON=OFF \
+    -DSHERPA_ONNX_ENABLE_TESTS=OFF \
+    -DSHERPA_ONNX_ENABLE_CHECK=OFF \
+    -DBUILD_SHARED_LIBS=ON \
+    -DSHERPA_ONNX_ENABLE_PORTAUDIO=OFF \
+    -DSHERPA_ONNX_ENABLE_JNI=ON \
+    ..
+
+  make -j4
+  ls -lh lib
+fi
+
+# please visit
+# https://github.com/k2-fsa/sherpa-onnx/releases/tag/tts-models
+# to download more models
+if [ ! -f ./vits-zh-hf-fanchen-C/tokens.txt ]; then
+  curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/tts-models/vits-zh-hf-fanchen-C.tar.bz2
+  tar xf vits-zh-hf-fanchen-C.tar.bz2
+  rm vits-zh-hf-fanchen-C.tar.bz2
+fi
+
+java \
+  -Djava.library.path=$PWD/../build/lib \
+  -cp ../sherpa-onnx/java-api/build/sherpa-onnx.jar \
+  NonStreamingTtsVitsZh.java