Add Java API for non-streaming ASR (#807)

.github/workflows/run-java-test.yaml

@@ -92,13 +92,9 @@ jobs:
             make -j4
             ls -lh lib
 
-      - name:  Run java test
+      - name:  Run java test (Streaming ASR)
         shell: bash
         run: |
-          export CMAKE_CXX_COMPILER_LAUNCHER=ccache
-          export PATH="/usr/lib/ccache:/usr/local/opt/ccache/libexec:$PATH"
-          cmake --version
-
           cd ./java-api-examples
           ./run-streaming-decode-file-ctc.sh
           # Delete model files to save space
@@ -109,3 +105,16 @@ jobs:
 
           ./run-streaming-decode-file-transducer.sh
           rm -rf sherpa-onnx-streaming-*
+
+      - name:  Run java test (Non-Streaming ASR)
+        shell: bash
+        run: |
+          cd ./java-api-examples
+          ./run-non-streaming-decode-file-paraformer.sh
+          rm -rf sherpa-onnx-paraformer-zh-*
+
+          ./run-non-streaming-decode-file-transducer.sh
+          rm -rf sherpa-onnx-zipformer-*
+
+          ./run-non-streaming-decode-file-whisper.sh
+          rm -rf sherpa-onnx-whisper-*

java-api-examples/.gitignore

@@ -1,3 +1,4 @@
 lib
 hs_err*
 !run-streaming*.sh
+!run-non-streaming*.sh

java-api-examples/NonStreamingDecodeFileParaformer.java

@@ -0,0 +1,49 @@
+// Copyright 2024 Xiaomi Corporation
+
+// This file shows how to use an offline paraformer, i.e., non-streaming paraformer,
+// to decode files.
+import com.k2fsa.sherpa.onnx.*;
+
+public class NonStreamingDecodeFileTransducer {
+  public static void main(String[] args) {
+    // please refer to
+    // https://k2-fsa.github.io/sherpa/onnx/pretrained_models/offline-paraformer/paraformer-models.html#csukuangfj-sherpa-onnx-paraformer-zh-2023-03-28-chinese-english
+    // to download model files
+    String model = "./sherpa-onnx-paraformer-zh-2023-03-28/model.int8.onnx";
+    String tokens = "./sherpa-onnx-paraformer-zh-2023-03-28/tokens.txt";
+
+    String waveFilename = "./sherpa-onnx-paraformer-zh-2023-03-28/test_wavs/3-sichuan.wav";
+
+    WaveReader reader = new WaveReader(waveFilename);
+
+    OfflineParaformerModelConfig paraformer =
+        OfflineParaformerModelConfig.builder().setModel(model).build();
+
+    OfflineModelConfig modelConfig =
+        OfflineModelConfig.builder()
+            .setParaformer(paraformer)
+            .setTokens(tokens)
+            .setNumThreads(1)
+            .setDebug(true)
+            .build();
+
+    OfflineRecognizerConfig config =
+        OfflineRecognizerConfig.builder()
+            .setOfflineModelConfig(modelConfig)
+            .setDecodingMethod("greedy_search")
+            .build();
+
+    OfflineRecognizer recognizer = new OfflineRecognizer(config);
+    OfflineStream stream = recognizer.createStream();
+    stream.acceptWaveform(reader.getSamples(), reader.getSampleRate());
+
+    recognizer.decode(stream);
+
+    String text = recognizer.getResult(stream).getText();
+
+    System.out.printf("filename:%s\nresult:%s\n", waveFilename, text);
+
+    stream.release();
+    recognizer.release();
+  }
+}

java-api-examples/NonStreamingDecodeFileTransducer.java

@@ -0,0 +1,57 @@
+// Copyright 2024 Xiaomi Corporation
+
+// This file shows how to use an offline transducer, i.e., non-streaming transducer,
+// to decode files.
+import com.k2fsa.sherpa.onnx.*;
+
+public class NonStreamingDecodeFileTransducer {
+  public static void main(String[] args) {
+    // please refer to
+    // https://k2-fsa.github.io/sherpa/onnx/pretrained_models/offline-transducer/zipformer-transducer-models.html#sherpa-onnx-zipformer-gigaspeech-2023-12-12-english
+    // to download model files
+    String encoder =
+        "./sherpa-onnx-zipformer-gigaspeech-2023-12-12/encoder-epoch-30-avg-1.int8.onnx";
+    String decoder = "./sherpa-onnx-zipformer-gigaspeech-2023-12-12/decoder-epoch-30-avg-1.onnx";
+    String joiner = "./sherpa-onnx-zipformer-gigaspeech-2023-12-12/joiner-epoch-30-avg-1.onnx";
+    String tokens = "./sherpa-onnx-zipformer-gigaspeech-2023-12-12/tokens.txt";
+
+    String waveFilename =
+        "./sherpa-onnx-zipformer-gigaspeech-2023-12-12/test_wavs/1089-134686-0001.wav";
+
+    WaveReader reader = new WaveReader(waveFilename);
+
+    OfflineTransducerModelConfig transducer =
+        OfflineTransducerModelConfig.builder()
+            .setEncoder(encoder)
+            .setDecoder(decoder)
+            .setJoiner(joiner)
+            .build();
+
+    OfflineModelConfig modelConfig =
+        OfflineModelConfig.builder()
+            .setTransducer(transducer)
+            .setTokens(tokens)
+            .setNumThreads(1)
+            .setDebug(true)
+            .build();
+
+    OfflineRecognizerConfig config =
+        OfflineRecognizerConfig.builder()
+            .setOfflineModelConfig(modelConfig)
+            .setDecodingMethod("greedy_search")
+            .build();
+
+    OfflineRecognizer recognizer = new OfflineRecognizer(config);
+    OfflineStream stream = recognizer.createStream();
+    stream.acceptWaveform(reader.getSamples(), reader.getSampleRate());
+
+    recognizer.decode(stream);
+
+    String text = recognizer.getResult(stream).getText();
+
+    System.out.printf("filename:%s\nresult:%s\n", waveFilename, text);
+
+    stream.release();
+    recognizer.release();
+  }
+}

java-api-examples/NonStreamingDecodeFileWhisper.java

@@ -0,0 +1,50 @@
+// Copyright 2024 Xiaomi Corporation
+
+// This file shows how to use an offline whisper, i.e., non-streaming whisper,
+// to decode files.
+import com.k2fsa.sherpa.onnx.*;
+
+public class NonStreamingDecodeFileWhisper {
+  public static void main(String[] args) {
+    // please refer to
+    // https://k2-fsa.github.io/sherpa/onnx/pretrained_models/whisper/tiny.en.html
+    // to download model files
+    String encoder = "./sherpa-onnx-whisper-tiny.en/tiny.en-encoder.int8.onnx";
+    String decoder = "./sherpa-onnx-whisper-tiny.en/tiny.en-decoder.int8.onnx";
+    String tokens = "./sherpa-onnx-whisper-tiny.en/tiny.en-tokens.txt";
+
+    String waveFilename = "./sherpa-onnx-whisper-tiny.en/test_wavs/1.wav";
+
+    WaveReader reader = new WaveReader(waveFilename);
+
+    OfflineWhisperModelConfig whisper =
+        OfflineWhisperModelConfig.builder().setEncoder(encoder).setDecoder(decoder).build();
+
+    OfflineModelConfig modelConfig =
+        OfflineModelConfig.builder()
+            .setWhisper(whisper)
+            .setTokens(tokens)
+            .setNumThreads(1)
+            .setDebug(true)
+            .build();
+
+    OfflineRecognizerConfig config =
+        OfflineRecognizerConfig.builder()
+            .setOfflineModelConfig(modelConfig)
+            .setDecodingMethod("greedy_search")
+            .build();
+
+    OfflineRecognizer recognizer = new OfflineRecognizer(config);
+    OfflineStream stream = recognizer.createStream();
+    stream.acceptWaveform(reader.getSamples(), reader.getSampleRate());
+
+    recognizer.decode(stream);
+
+    String text = recognizer.getResult(stream).getText();
+
+    System.out.printf("filename:%s\nresult:%s\n", waveFilename, text);
+
+    stream.release();
+    recognizer.release();
+  }
+}

java-api-examples/README.md

@@ -4,8 +4,18 @@ This directory contains examples for the JAVA API of sherpa-onnx.
 
 # Usage
 
+## Streaming Speech recognition
+
 ```
 ./run-streaming-decode-file-ctc.sh
 ./run-streaming-decode-file-paraformer.sh
 ./run-streaming-decode-file-transducer.sh
 ```
+
+## Non-Streaming Speech recognition
+
+```bash
+./run-non-streaming-decode-file-paraformer.sh
+./run-non-streaming-decode-file-transducer.sh
+./run-non-streaming-decode-file-whisper.sh
+```

java-api-examples/StreamingDecodeFileCtc.java

@@ -1,4 +1,3 @@
-// Copyright 2022-2023 by zhaoming
 // Copyright 2024 Xiaomi Corporation
 
 // This file shows how to use an online CTC model, i.e., streaming CTC model,
@@ -16,8 +15,6 @@ public static void main(String[] args) {
     String waveFilename = "./sherpa-onnx-streaming-zipformer-ctc-small-2024-03-18/test_wavs/8k.wav";
 
     WaveReader reader = new WaveReader(waveFilename);
-    System.out.println(reader.getSampleRate());
-    System.out.println(reader.getSamples().length);
 
     OnlineZipformer2CtcModelConfig ctc =
         OnlineZipformer2CtcModelConfig.builder().setModel(model).build();

java-api-examples/StreamingDecodeFileParaformer.java

@@ -1,4 +1,3 @@
-// Copyright 2022-2023 by zhaoming
 // Copyright 2024 Xiaomi Corporation
 
 // This file shows how to use an online paraformer, i.e., streaming paraformer,
@@ -16,8 +15,6 @@ public static void main(String[] args) {
     String waveFilename = "./sherpa-onnx-streaming-paraformer-bilingual-zh-en/test_wavs/2.wav";
 
     WaveReader reader = new WaveReader(waveFilename);
-    System.out.println(reader.getSampleRate());
-    System.out.println(reader.getSamples().length);
 
     OnlineParaformerModelConfig paraformer =
         OnlineParaformerModelConfig.builder().setEncoder(encoder).setDecoder(decoder).build();

java-api-examples/StreamingDecodeFileTransducer.java

@@ -22,8 +22,6 @@ public static void main(String[] args) {
         "./sherpa-onnx-streaming-zipformer-bilingual-zh-en-2023-02-20/test_wavs/0.wav";
 
     WaveReader reader = new WaveReader(waveFilename);
-    System.out.println(reader.getSampleRate());
-    System.out.println(reader.getSamples().length);
 
     OnlineTransducerModelConfig transducer =
         OnlineTransducerModelConfig.builder()

java-api-examples/run-non-streaming-decode-file-paraformer.sh

@@ -0,0 +1,52 @@
+#!/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
+
+if [ ! -f ./sherpa-onnx-paraformer-zh-2023-03-28/tokens.txt ]; then
+  curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/sherpa-onnx-paraformer-zh-2023-03-28.tar.bz2
+
+  tar xvf sherpa-onnx-paraformer-zh-2023-03-28.tar.bz2
+  rm sherpa-onnx-paraformer-zh-2023-03-28.tar.bz2
+fi
+
+java \
+  -Djava.library.path=$PWD/../build/lib \
+  -cp ../sherpa-onnx/java-api/build/sherpa-onnx.jar \
+  NonStreamingDecodeFileParaformer.java

java-api-examples/run-non-streaming-decode-file-transducer.sh

@@ -0,0 +1,52 @@
+#!/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
+
+if [ ! -f ./sherpa-onnx-zipformer-gigaspeech-2023-12-12/tokens.txt ]; then
+  curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/sherpa-onnx-zipformer-gigaspeech-2023-12-12.tar.bz2
+
+  tar xvf sherpa-onnx-zipformer-gigaspeech-2023-12-12.tar.bz2
+  rm sherpa-onnx-zipformer-gigaspeech-2023-12-12.tar.bz2
+fi
+
+java \
+  -Djava.library.path=$PWD/../build/lib \
+  -cp ../sherpa-onnx/java-api/build/sherpa-onnx.jar \
+  NonStreamingDecodeFileTransducer.java