Add CTC HLG decoding for JNI (#810)

java-api-examples/NonStreamingDecodeFileNemo.java

@@ -0,0 +1,49 @@
+// Copyright 2024 Xiaomi Corporation
+
+// This file shows how to use an offline NeMo CTC model, i.e., non-streaming NeMo CTC model,,
+// to decode files.
+import com.k2fsa.sherpa.onnx.*;
+
+public class NonStreamingDecodeFileNemo {
+  public static void main(String[] args) {
+    // please refer to
+    // https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/sherpa-onnx-nemo-ctc-en-citrinet-512.tar.bz2
+    // to download model files
+    String model = "./sherpa-onnx-nemo-ctc-en-citrinet-512/model.int8.onnx";
+    String tokens = "./sherpa-onnx-nemo-ctc-en-citrinet-512/tokens.txt";
+
+    String waveFilename = "./sherpa-onnx-nemo-ctc-en-citrinet-512/test_wavs/1.wav";
+
+    WaveReader reader = new WaveReader(waveFilename);
+
+    OfflineNemoEncDecCtcModelConfig nemo =
+        OfflineNemoEncDecCtcModelConfig.builder().setModel(model).build();
+
+    OfflineModelConfig modelConfig =
+        OfflineModelConfig.builder()
+            .setNemo(nemo)
+            .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

@@ -8,6 +8,7 @@ This directory contains examples for the JAVA API of sherpa-onnx.
 
 ```
 ./run-streaming-decode-file-ctc.sh
+./run-streaming-decode-file-ctc-hlg.sh
 ./run-streaming-decode-file-paraformer.sh
 ./run-streaming-decode-file-transducer.sh
 ```
@@ -18,4 +19,5 @@ This directory contains examples for the JAVA API of sherpa-onnx.
 ./run-non-streaming-decode-file-paraformer.sh
 ./run-non-streaming-decode-file-transducer.sh
 ./run-non-streaming-decode-file-whisper.sh
+./run-non-streaming-decode-file-nemo.sh
 ```

java-api-examples/StreamingDecodeFileCtcHLG.java

@@ -0,0 +1,58 @@
+// Copyright 2024 Xiaomi Corporation
+
+// This file shows how to use an online CTC model, i.e., streaming CTC model,
+// to decode files.
+import com.k2fsa.sherpa.onnx.*;
+
+public class StreamingDecodeFileCtcHLG {
+  public static void main(String[] args) {
+    // please refer to
+    // https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/sherpa-onnx-streaming-zipformer-ctc-small-2024-03-18.tar.bz2
+    // to download model files
+    String model =
+        "./sherpa-onnx-streaming-zipformer-ctc-small-2024-03-18/ctc-epoch-30-avg-3-chunk-16-left-128.int8.onnx";
+    String tokens = "./sherpa-onnx-streaming-zipformer-ctc-small-2024-03-18/tokens.txt";
+    String hlg = "./sherpa-onnx-streaming-zipformer-ctc-small-2024-03-18/HLG.fst";
+    String waveFilename = "./sherpa-onnx-streaming-zipformer-ctc-small-2024-03-18/test_wavs/8k.wav";
+
+    WaveReader reader = new WaveReader(waveFilename);
+
+    OnlineZipformer2CtcModelConfig ctc =
+        OnlineZipformer2CtcModelConfig.builder().setModel(model).build();
+
+    OnlineModelConfig modelConfig =
+        OnlineModelConfig.builder()
+            .setZipformer2Ctc(ctc)
+            .setTokens(tokens)
+            .setNumThreads(1)
+            .setDebug(true)
+            .build();
+
+    OnlineCtcFstDecoderConfig ctcFstDecoderConfig =
+        OnlineCtcFstDecoderConfig.builder().setGraph("hlg").build();
+
+    OnlineRecognizerConfig config =
+        OnlineRecognizerConfig.builder()
+            .setOnlineModelConfig(modelConfig)
+            .setCtcFstDecoderConfig(ctcFstDecoderConfig)
+            .build();
+
+    OnlineRecognizer recognizer = new OnlineRecognizer(config);
+    OnlineStream stream = recognizer.createStream();
+    stream.acceptWaveform(reader.getSamples(), reader.getSampleRate());
+
+    float[] tailPaddings = new float[(int) (0.3 * reader.getSampleRate())];
+    stream.acceptWaveform(tailPaddings, reader.getSampleRate());
+
+    while (recognizer.isReady(stream)) {
+      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/run-non-streaming-decode-file-nemo.sh

@@ -0,0 +1,51 @@
+#!/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-nemo-ctc-en-citrinet-512/tokens.txt ]; then
+  curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/sherpa-onnx-nemo-ctc-en-citrinet-512.tar.bz2
+  tar xvf sherpa-onnx-nemo-ctc-en-citrinet-512.tar.bz2
+  rm sherpa-onnx-nemo-ctc-en-citrinet-512.tar.bz2
+fi
+
+java \
+  -Djava.library.path=$PWD/../build/lib \
+  -cp ../sherpa-onnx/java-api/build/sherpa-onnx.jar \
+  NonStreamingDecodeFileNemo.java

java-api-examples/run-streaming-decode-file-ctc-hlg.sh

@@ -0,0 +1,36 @@
+#!/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 ./sherpa-onnx-streaming-zipformer-ctc-small-2024-03-18/tokens.txt ]; then
+  curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/sherpa-onnx-streaming-zipformer-ctc-small-2024-03-18.tar.bz2
+  tar xvf sherpa-onnx-streaming-zipformer-ctc-small-2024-03-18.tar.bz2
+  rm sherpa-onnx-streaming-zipformer-ctc-small-2024-03-18.tar.bz2
+fi
+
+java \
+  -Djava.library.path=$PWD/../build/lib \
+  -cp ../sherpa-onnx/java-api/build/sherpa-onnx.jar \
+  StreamingDecodeFileCtcHLG.java