whisper.rn 0.3.3 → 0.3.5

package/android/src/oldarch/java/com/rnwhisper/RNWhisperModule.java

@@ -1,19 +1,12 @@
 package com.rnwhisper;
 
 import androidx.annotation.NonNull;
-import android.util.Log;
-import android.os.Build;
-import android.os.Handler;
-import android.os.AsyncTask;
-import android.media.AudioRecord;
 
 import com.facebook.react.bridge.Promise;
 import com.facebook.react.bridge.ReactApplicationContext;
 import com.facebook.react.bridge.ReactContextBaseJavaModule;
 import com.facebook.react.bridge.ReactMethod;
-import com.facebook.react.bridge.LifecycleEventListener;
 import com.facebook.react.bridge.ReadableMap;
-import com.facebook.react.bridge.WritableMap;
 import com.facebook.react.module.annotations.ReactModule;
 
 import java.util.HashMap;
@@ -23,17 +16,14 @@ import java.io.FileInputStream;
 import java.io.PushbackInputStream;
 
 @ReactModule(name = RNWhisperModule.NAME)
-public class RNWhisperModule extends ReactContextBaseJavaModule implements LifecycleEventListener {
+public class RNWhisperModule extends ReactContextBaseJavaModule {
   public static final String NAME = "RNWhisper";
 
-  private ReactApplicationContext reactContext;
-  private Downloader downloader;
+  private RNWhisper rnwhisper;
 
   public RNWhisperModule(ReactApplicationContext reactContext) {
     super(reactContext);
-    reactContext.addLifecycleEventListener(this);
-    this.reactContext = reactContext;
-    this.downloader = new Downloader(reactContext);
+    rnwhisper = new RNWhisper(reactContext);
   }
 
   @Override
@@ -42,234 +32,38 @@ public class RNWhisperModule extends ReactContextBaseJavaModule implements Lifec
     return NAME;
   }
 
-  private HashMap<Integer, WhisperContext> contexts = new HashMap<>();
-
-  private int getResourceIdentifier(String filePath) {
-    int identifier = reactContext.getResources().getIdentifier(
-      filePath,
-      "drawable",
-      reactContext.getPackageName()
-    );
-    if (identifier == 0) {
-      identifier = reactContext.getResources().getIdentifier(
-        filePath,
-        "raw",
-        reactContext.getPackageName()
-      );
-    }
-    return identifier;
+  @Override
+  public HashMap<String, Object> getConstants() {
+    return rnwhisper.getTypedExportedConstants();
   }
 
   @ReactMethod
   public void initContext(final ReadableMap options, final Promise promise) {
-    new AsyncTask<Void, Void, Integer>() {
-      private Exception exception;
-
-      @Override
-      protected Integer doInBackground(Void... voids) {
-        try {
-          String modelPath = options.getString("filePath");
-          boolean isBundleAsset = options.getBoolean("isBundleAsset");
-
-          String modelFilePath = modelPath;
-          if (!isBundleAsset && (modelPath.startsWith("http://") || modelPath.startsWith("https://"))) {
-            modelFilePath = downloader.downloadFile(modelPath);
-          }
-
-          long context;
-          int resId = getResourceIdentifier(modelFilePath);
-          if (resId > 0) {
-            context = WhisperContext.initContextWithInputStream(
-              new PushbackInputStream(reactContext.getResources().openRawResource(resId))
-            );
-          } else if (isBundleAsset) {
-            context = WhisperContext.initContextWithAsset(reactContext.getAssets(), modelFilePath);
-          } else {
-            context = WhisperContext.initContext(modelFilePath);
-          }
-          if (context == 0) {
-            throw new Exception("Failed to initialize context");
-          }
-          int id = Math.abs(new Random().nextInt());
-          WhisperContext whisperContext = new WhisperContext(id, reactContext, context);
-          contexts.put(id, whisperContext);
-          return id;
-        } catch (Exception e) {
-          exception = e;
-          return null;
-        }
-      }
-
-      @Override
-      protected void onPostExecute(Integer id) {
-        if (exception != null) {
-          promise.reject(exception);
-          return;
-        }
-        promise.resolve(id);
-      }
-    }.execute();
+    rnwhisper.initContext(options, promise);
   }
 
   @ReactMethod
-  public void transcribeFile(int id, int jobId, String filePath, ReadableMap options, Promise promise) {
-    final WhisperContext context = contexts.get(id);
-    if (context == null) {
-      promise.reject("Context not found");
-      return;
-    }
-    if (context.isCapturing()) {
-      promise.reject("The context is in realtime transcribe mode");
-      return;
-    }
-    if (context.isTranscribing()) {
-      promise.reject("Context is already transcribing");
-      return;
-    }
-    new AsyncTask<Void, Void, WritableMap>() {
-      private Exception exception;
-
-      @Override
-      protected WritableMap doInBackground(Void... voids) {
-        try {
-          String waveFilePath = filePath;
-
-          if (filePath.startsWith("http://") || filePath.startsWith("https://")) {
-            waveFilePath = downloader.downloadFile(filePath);
-          }
-
-          int resId = getResourceIdentifier(waveFilePath);
-          if (resId > 0) {
-            return context.transcribeInputStream(
-              (int) jobId,
-              reactContext.getResources().openRawResource(resId),
-              options
-            );
-          }
-
-          return context.transcribeInputStream(
-            (int) jobId,
-            new FileInputStream(new File(waveFilePath)),
-            options
-          );
-        } catch (Exception e) {
-          exception = e;
-          return null;
-        }
-      }
-
-      @Override
-      protected void onPostExecute(WritableMap data) {
-        if (exception != null) {
-          promise.reject(exception);
-          return;
-        }
-        promise.resolve(data);
-      }
-    }.execute();
+  public void transcribeFile(double id, double jobId, String filePath, ReadableMap options, Promise promise) {
+    rnwhisper.transcribeFile(id, jobId, filePath, options, promise);
   }
 
   @ReactMethod
-  public void startRealtimeTranscribe(int id, int jobId, ReadableMap options, Promise promise) {
-    final WhisperContext context = contexts.get(id);
-    if (context == null) {
-      promise.reject("Context not found");
-      return;
-    }
-    if (context.isCapturing()) {
-      promise.reject("Context is already in capturing");
-      return;
-    }
-    int state = context.startRealtimeTranscribe(jobId, options);
-    if (state == AudioRecord.STATE_INITIALIZED) {
-      promise.resolve(null);
-      return;
-    }
-    promise.reject("Failed to start realtime transcribe. State: " + state);
+  public void startRealtimeTranscribe(double id, double jobId, ReadableMap options, Promise promise) {
+    rnwhisper.startRealtimeTranscribe(id, jobId, options, promise);
   }
 
   @ReactMethod
-  public void abortTranscribe(int contextId, int jobId, Promise promise) {
-    WhisperContext context = contexts.get(contextId);
-    if (context == null) {
-      promise.reject("Context not found");
-      return;
-    }
-    context.stopTranscribe(jobId);
+  public void abortTranscribe(double contextId, double jobId, Promise promise) {
+    rnwhisper.abortTranscribe(contextId, jobId, promise);
   }
 
   @ReactMethod
-  public void releaseContext(int id, Promise promise) {
-    new AsyncTask<Void, Void, Void>() {
-      private Exception exception;
-
-      @Override
-      protected Void doInBackground(Void... voids) {
-        try {
-          WhisperContext context = contexts.get(id);
-          if (context == null) {
-            throw new Exception("Context " + id + " not found");
-          }
-          context.release();
-          contexts.remove(id);
-        } catch (Exception e) {
-          exception = e;
-        }
-        return null;
-      }
-
-      @Override
-      protected void onPostExecute(Void result) {
-        if (exception != null) {
-          promise.reject(exception);
-          return;
-        }
-        promise.resolve(null);
-      }
-    }.execute();
+  public void releaseContext(double id, Promise promise) {
+    rnwhisper.releaseContext(id, promise);
   }
 
   @ReactMethod
   public void releaseAllContexts(Promise promise) {
-    new AsyncTask<Void, Void, Void>() {
-      private Exception exception;
-
-      @Override
-      protected Void doInBackground(Void... voids) {
-        try {
-          onHostDestroy();
-        } catch (Exception e) {
-          exception = e;
-        }
-        return null;
-      }
-
-      @Override
-      protected void onPostExecute(Void result) {
-        if (exception != null) {
-          promise.reject(exception);
-          return;
-        }
-        promise.resolve(null);
-      }
-    }.execute();
-  }
-
-  @Override
-  public void onHostResume() {
-  }
-
-  @Override
-  public void onHostPause() {
-  }
-
-  @Override
-  public void onHostDestroy() {
-    WhisperContext.abortAllTranscribe();
-    for (WhisperContext context : contexts.values()) {
-      context.release();
-    }
-    contexts.clear();
-    downloader.clearCache();
+    rnwhisper.releaseAllContexts(promise);
   }
 }

package/cpp/README.md

@@ -0,0 +1,4 @@
+# Note
+
+- Only `rn-whisper.h` / `rn-whisper.cpp` are the specific files for this project, others are sync from [whisper.cpp](https://github.com/ggerganov/whisper.cpp).
+- We can update the native source by using the [bootstrap](../scripts/bootstrap.sh) script.

package/cpp/coreml/whisper-encoder.mm

@@ -53,9 +53,11 @@ void whisper_coreml_encode(
                                            error: nil
     ];
 
-    whisper_encoder_implOutput * outCoreML = [(__bridge id) ctx->data predictionFromLogmel_data:inMultiArray error:nil];
+    @autoreleasepool {
+        whisper_encoder_implOutput * outCoreML = [(__bridge id) ctx->data predictionFromLogmel_data:inMultiArray error:nil];
 
-    memcpy(out, outCoreML.output.dataPointer, outCoreML.output.count * sizeof(float));
+        memcpy(out, outCoreML.output.dataPointer, outCoreML.output.count * sizeof(float));
+    }
 }
 
 #if __cplusplus

package/cpp/ggml.c

@@ -292,7 +292,7 @@ typedef double wsp_ggml_float;
 #if defined(_MSC_VER) || defined(__MINGW32__)
 #include <intrin.h>
 #else
-#if defined(__AVX__) || defined(__AVX2__) || defined(__AVX512F__) || defined(__SSSE3__)
+#if defined(__AVX__) || defined(__AVX2__) || defined(__AVX512F__) || defined(__SSSE3__) || defined(__SSE3__)
 #include <immintrin.h>
 #endif
 #endif
@@ -18721,6 +18721,14 @@ int wsp_ggml_cpu_has_sse3(void) {
 #endif
 }
 
+int wsp_ggml_cpu_has_ssse3(void) {
+#if defined(__SSSE3__)
+    return 1;
+#else
+    return 0;
+#endif
+}
+
 int wsp_ggml_cpu_has_vsx(void) {
 #if defined(__POWER9_VECTOR__)
     return 1;

package/cpp/ggml.h

@@ -1508,6 +1508,7 @@ extern "C" {
     WSP_GGML_API int wsp_ggml_cpu_has_clblast    (void);
     WSP_GGML_API int wsp_ggml_cpu_has_gpublas    (void);
     WSP_GGML_API int wsp_ggml_cpu_has_sse3       (void);
+    WSP_GGML_API int wsp_ggml_cpu_has_ssse3      (void);
     WSP_GGML_API int wsp_ggml_cpu_has_vsx        (void);
 
     //