Add KWS examples for Java API (#930)

java-api-examples/KeywordSpotterFromFile.java

@@ -0,0 +1,65 @@
+// Copyright 2024 Xiaomi Corporation
+
+// This file shows how to use a keyword spotter model to spot keywords from
+// a file.
+
+import com.k2fsa.sherpa.onnx.*;
+
+public class KyewordSpotterFromFile {
+  public static void main(String[] args) {
+    // please download test files from https://github.com/k2-fsa/sherpa-onnx/releases/tag/kws-models
+    String encoder =
+        "./sherpa-onnx-kws-zipformer-wenetspeech-3.3M-2024-01-01/encoder-epoch-12-avg-2-chunk-16-left-64.onnx";
+    String decoder =
+        "./sherpa-onnx-kws-zipformer-wenetspeech-3.3M-2024-01-01/decoder-epoch-12-avg-2-chunk-16-left-64.onnx";
+    String joiner =
+        "./sherpa-onnx-kws-zipformer-wenetspeech-3.3M-2024-01-01/joiner-epoch-12-avg-2-chunk-16-left-64.onnx";
+    String tokens = "./sherpa-onnx-kws-zipformer-wenetspeech-3.3M-2024-01-01/tokens.txt";
+
+    String keywordsFile =
+        "./sherpa-onnx-kws-zipformer-wenetspeech-3.3M-2024-01-01/test_wavs/test_keywords.txt";
+
+    String waveFilename = "./sherpa-onnx-kws-zipformer-wenetspeech-3.3M-2024-01-01/test_wavs/3.wav";
+
+    OnlineTransducerModelConfig transducer =
+        OnlineTransducerModelConfig.builder()
+            .setEncoder(encoder)
+            .setDecoder(decoder)
+            .setJoiner(joiner)
+            .build();
+
+    OnlineModelConfig modelConfig =
+        OnlineModelConfig.builder()
+            .setTransducer(transducer)
+            .setTokens(tokens)
+            .setNumThreads(1)
+            .setDebug(true)
+            .build();
+
+    KeywordSpotterConfig config =
+        KeywordSpotterConfig.builder()
+            .setOnlineModelConfig(modelConfig)
+            .setKeywordsFile(keywordsFile)
+            .build();
+
+    KeywordSpotter kws = new KeywordSpotter(config);
+    OnlineStream stream = kws.createStream();
+
+    WaveReader reader = new WaveReader(waveFilename);
+
+    stream.acceptWaveform(reader.getSamples(), reader.getSampleRate());
+
+    float[] tailPaddings = new float[(int) (0.8 * reader.getSampleRate())];
+    stream.acceptWaveform(tailPaddings, reader.getSampleRate());
+    while (kws.isReady(stream)) {
+      kws.decode(stream);
+
+      String keyword = kws.getResult(stream).getKeyword();
+      if (!keyword.isEmpty()) {
+        System.out.printf("Detected keyword: %s\n", keyword);
+      }
+    }
+
+    kws.release();
+  }
+}

java-api-examples/README.md

@@ -68,3 +68,9 @@ The punctuation model supports both English and Chinese.
 ```bash
 ./run-vad-non-streaming-paraformer.sh
 ```
+
+## Keyword spotter
+
+```bash
+./run-kws-from-file.sh
+```

java-api-examples/VadNonStreamingParaformer.java

@@ -91,6 +91,7 @@ public class VadNonStreamingParaformer {
           stream.acceptWaveform(segment.getSamples(), 16000);
           recognizer.decode(stream);
           String text = recognizer.getResult(stream).getText();
+          stream.release();
 
           if (!text.isEmpty()) {
             System.out.printf("%.3f--%.3f: %s\n", startTime, startTime + duration, text);
@@ -100,5 +101,8 @@ public class VadNonStreamingParaformer {
         }
       }
     }
+
+    vad.release();
+    recognizer.release();
   }
 }

java-api-examples/VadRemoveSilence.java

@@ -75,5 +75,7 @@ public class VadRemoveSilence {
     String outFilename = "lei-jun-test-no-silence.wav";
     WaveWriter.write(outFilename, allSamples, 16000);
     System.out.printf("Saved to %s\n", outFilename);
+
+    vad.release();
   }
 }

java-api-examples/run-kws-from-file.sh

@@ -0,0 +1,37 @@
+#!/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-kws-zipformer-wenetspeech-3.3M-2024-01-01/tokens.txt ]; then
+  curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/kws-models/sherpa-onnx-kws-zipformer-wenetspeech-3.3M-2024-01-01.tar.bz2
+  tar xvf sherpa-onnx-kws-zipformer-wenetspeech-3.3M-2024-01-01.tar.bz2
+  rm sherpa-onnx-kws-zipformer-wenetspeech-3.3M-2024-01-01.tar.bz2
+fi
+
+java \
+  -Djava.library.path=$PWD/../build/lib \
+  -cp ../sherpa-onnx/java-api/build/sherpa-onnx.jar \
+  ./KeywordSpotterFromFile.java