Generate silent audio if no audio track is available.

To always generate silent audio, `removeAudio(true)` can be used in conjunction.

PiperOrigin-RevId: 502814315

demos/transformer/src/main/java/androidx/media3/demo/transformer/ConfigurationActivity.java

@@ -61,7 +61,7 @@ public final class ConfigurationActivity extends AppCompatActivity {
   public static final String SHOULD_REMOVE_AUDIO = "should_remove_audio";
   public static final String SHOULD_REMOVE_VIDEO = "should_remove_video";
   public static final String SHOULD_FLATTEN_FOR_SLOW_MOTION = "should_flatten_for_slow_motion";
-  public static final String FORCE_SILENT_AUDIO = "force_silent_audio";
+  public static final String GENERATE_SILENT_AUDIO = "generate_silent_audio";
   public static final String AUDIO_MIME_TYPE = "audio_mime_type";
   public static final String VIDEO_MIME_TYPE = "video_mime_type";
   public static final String RESOLUTION_HEIGHT = "resolution_height";
@@ -210,7 +210,7 @@ public final class ConfigurationActivity extends AppCompatActivity {
   private @MonotonicNonNull CheckBox removeAudioCheckbox;
   private @MonotonicNonNull CheckBox removeVideoCheckbox;
   private @MonotonicNonNull CheckBox flattenForSlowMotionCheckbox;
-  private @MonotonicNonNull CheckBox forceSilentAudioCheckbox;
+  private @MonotonicNonNull CheckBox generateSilentAudioCheckbox;
   private @MonotonicNonNull Spinner audioMimeSpinner;
   private @MonotonicNonNull Spinner videoMimeSpinner;
   private @MonotonicNonNull Spinner resolutionHeightSpinner;
@@ -271,7 +271,7 @@ public final class ConfigurationActivity extends AppCompatActivity {
 
     flattenForSlowMotionCheckbox = findViewById(R.id.flatten_for_slow_motion_checkbox);
 
-    forceSilentAudioCheckbox = findViewById(R.id.force_silent_audio_checkbox);
+    generateSilentAudioCheckbox = findViewById(R.id.generate_silent_audio_checkbox);
 
     ArrayAdapter<String> audioMimeAdapter =
         new ArrayAdapter<>(/* context= */ this, R.layout.spinner_item);
@@ -382,7 +382,7 @@ public final class ConfigurationActivity extends AppCompatActivity {
     "removeAudioCheckbox",
     "removeVideoCheckbox",
     "flattenForSlowMotionCheckbox",
-    "forceSilentAudioCheckbox",
+    "generateSilentAudioCheckbox",
     "audioMimeSpinner",
     "videoMimeSpinner",
     "resolutionHeightSpinner",
@@ -402,7 +402,7 @@ public final class ConfigurationActivity extends AppCompatActivity {
     bundle.putBoolean(SHOULD_REMOVE_AUDIO, removeAudioCheckbox.isChecked());
     bundle.putBoolean(SHOULD_REMOVE_VIDEO, removeVideoCheckbox.isChecked());
     bundle.putBoolean(SHOULD_FLATTEN_FOR_SLOW_MOTION, flattenForSlowMotionCheckbox.isChecked());
-    bundle.putBoolean(FORCE_SILENT_AUDIO, forceSilentAudioCheckbox.isChecked());
+    bundle.putBoolean(GENERATE_SILENT_AUDIO, generateSilentAudioCheckbox.isChecked());
     String selectedAudioMimeType = String.valueOf(audioMimeSpinner.getSelectedItem());
     if (!SAME_AS_INPUT_OPTION.equals(selectedAudioMimeType)) {
       bundle.putString(AUDIO_MIME_TYPE, selectedAudioMimeType);
@@ -738,7 +738,7 @@ public final class ConfigurationActivity extends AppCompatActivity {
 
   @RequiresNonNull({
     "removeVideoCheckbox",
-    "forceSilentAudioCheckbox",
+    "generateSilentAudioCheckbox",
     "audioMimeSpinner",
     "videoMimeSpinner",
     "resolutionHeightSpinner",
@@ -760,7 +760,7 @@ public final class ConfigurationActivity extends AppCompatActivity {
 
   @RequiresNonNull({
     "removeAudioCheckbox",
-    "forceSilentAudioCheckbox",
+    "generateSilentAudioCheckbox",
     "audioMimeSpinner",
     "videoMimeSpinner",
     "resolutionHeightSpinner",
@@ -781,7 +781,7 @@ public final class ConfigurationActivity extends AppCompatActivity {
   }
 
   @RequiresNonNull({
-    "forceSilentAudioCheckbox",
+    "generateSilentAudioCheckbox",
     "audioMimeSpinner",
     "videoMimeSpinner",
     "resolutionHeightSpinner",
@@ -793,7 +793,7 @@ public final class ConfigurationActivity extends AppCompatActivity {
     "selectVideoEffectsButton"
   })
   private void enableTrackSpecificOptions(boolean isAudioEnabled, boolean isVideoEnabled) {
-    forceSilentAudioCheckbox.setEnabled(isVideoEnabled);
+    generateSilentAudioCheckbox.setEnabled(isVideoEnabled);
     audioMimeSpinner.setEnabled(isAudioEnabled);
     videoMimeSpinner.setEnabled(isVideoEnabled);
     resolutionHeightSpinner.setEnabled(isVideoEnabled);

demos/transformer/src/main/java/androidx/media3/demo/transformer/TransformerActivity.java

@@ -309,6 +309,8 @@ public final class TransformerActivity extends AppCompatActivity {
       transformerBuilder
           .setRemoveAudio(bundle.getBoolean(ConfigurationActivity.SHOULD_REMOVE_AUDIO))
           .setRemoveVideo(bundle.getBoolean(ConfigurationActivity.SHOULD_REMOVE_VIDEO))
+          .experimentalSetGenerateSilentAudio(
+              bundle.getBoolean(ConfigurationActivity.GENERATE_SILENT_AUDIO))
           .setEncoderFactory(
               new DefaultEncoderFactory.Builder(this.getApplicationContext())
                   .setEnableFallback(bundle.getBoolean(ConfigurationActivity.ENABLE_FALLBACK))
@@ -322,9 +324,6 @@ public final class TransformerActivity extends AppCompatActivity {
       if (bundle.getBoolean(ConfigurationActivity.ENABLE_DEBUG_PREVIEW)) {
         transformerBuilder.setDebugViewProvider(new DemoDebugViewProvider());
       }
-
-      transformerBuilder.experimentalSetForceSilentAudio(
-          bundle.getBoolean(ConfigurationActivity.FORCE_SILENT_AUDIO));
     }
 
     return transformerBuilder

demos/transformer/src/main/res/layout/configuration_activity.xml

@@ -116,9 +116,9 @@
         android:layout_weight="1">
         <TextView
           android:layout_gravity="center_vertical"
-          android:text="@string/force_silent_audio" />
+          android:text="@string/generate_silent_audio" />
         <CheckBox
-          android:id="@+id/force_silent_audio_checkbox"
+          android:id="@+id/generate_silent_audio_checkbox"
           android:layout_gravity="end" />
       </TableRow>
       <TableRow

demos/transformer/src/main/res/values/strings.xml

@@ -66,7 +66,7 @@
   <string name="permission_denied">Permission Denied</string>
   <string name="hide_input_video">Hide input video</string>
   <string name="show_input_video">Show input video</string>
-  <string name="force_silent_audio">Force silent audio</string>
+  <string name="generate_silent_audio">Generate silent audio</string>
   <string name="overlay_alpha">Alpha</string>
   <string name="overlay_uri">Uri</string>
   <string name="bitmap_overlay_settings">Specify bitmap overlay settings</string>

libraries/test_data/src/test/assets/transformerdumps/mp4/sample_18byte_nclx_colr.mp4.silentaudio.dump

@@ -0,0 +1,466 @@
+format 0:
+  id = 1
+  sampleMimeType = video/avc
+  codecs = avc1.64001F
+  maxInputSize = 36722
+  width = 1080
+  height = 720
+  frameRate = 29.970028
+  colorInfo:
+    colorSpace = 1
+    colorRange = 2
+    colorTransfer = 3
+    hdrStaticInfo = length 0, hash 0
+  initializationData:
+    data = length 29, hash 4746B5D9
+    data = length 10, hash 7A0D0F2B
+format 1:
+  sampleMimeType = audio/mp4a-latm
+  channelCount = 2
+  sampleRate = 44100
+  pcmEncoding = 2
+sample:
+  trackIndex = 1
+  dataHashCode = 1742602241
+  size = 4096
+  isKeyFrame = true
+  presentationTimeUs = 0
+sample:
+  trackIndex = 1
+  dataHashCode = 1742602241
+  size = 4096
+  isKeyFrame = true
+  presentationTimeUs = 23220
+sample:
+  trackIndex = 1
+  dataHashCode = 1742602241
+  size = 4096
+  isKeyFrame = true
+  presentationTimeUs = 46440
+sample:
+  trackIndex = 1
+  dataHashCode = 1742602241
+  size = 4096
+  isKeyFrame = true
+  presentationTimeUs = 69660
+sample:
+  trackIndex = 1
+  dataHashCode = 1742602241
+  size = 4096
+  isKeyFrame = true
+  presentationTimeUs = 92880
+sample:
+  trackIndex = 1
+  dataHashCode = 1742602241
+  size = 4096
+  isKeyFrame = true
+  presentationTimeUs = 116100
+sample:
+  trackIndex = 1
+  dataHashCode = 1742602241
+  size = 4096
+  isKeyFrame = true
+  presentationTimeUs = 139320
+sample:
+  trackIndex = 1
+  dataHashCode = 1742602241
+  size = 4096
+  isKeyFrame = true
+  presentationTimeUs = 162540
+sample:
+  trackIndex = 1
+  dataHashCode = 1742602241
+  size = 4096
+  isKeyFrame = true
+  presentationTimeUs = 185760
+sample:
+  trackIndex = 1
+  dataHashCode = 1742602241
+  size = 4096
+  isKeyFrame = true
+  presentationTimeUs = 208980
+sample:
+  trackIndex = 1
+  dataHashCode = 1742602241
+  size = 4096
+  isKeyFrame = true
+  presentationTimeUs = 232200
+sample:
+  trackIndex = 1
+  dataHashCode = 1742602241
+  size = 4096
+  isKeyFrame = true
+  presentationTimeUs = 255420
+sample:
+  trackIndex = 1
+  dataHashCode = 1742602241
+  size = 4096
+  isKeyFrame = true
+  presentationTimeUs = 278640
+sample:
+  trackIndex = 1
+  dataHashCode = 1742602241
+  size = 4096
+  isKeyFrame = true
+  presentationTimeUs = 301860
+sample:
+  trackIndex = 1
+  dataHashCode = 1742602241
+  size = 4096
+  isKeyFrame = true
+  presentationTimeUs = 325080
+sample:
+  trackIndex = 1
+  dataHashCode = 1742602241
+  size = 4096
+  isKeyFrame = true
+  presentationTimeUs = 348300
+sample:
+  trackIndex = 1
+  dataHashCode = 1742602241
+  size = 4096
+  isKeyFrame = true
+  presentationTimeUs = 371520
+sample:
+  trackIndex = 1
+  dataHashCode = 1742602241
+  size = 4096
+  isKeyFrame = true
+  presentationTimeUs = 394740
+sample:
+  trackIndex = 1
+  dataHashCode = 1742602241
+  size = 4096
+  isKeyFrame = true
+  presentationTimeUs = 417960
+sample:
+  trackIndex = 1
+  dataHashCode = 1742602241
+  size = 4096
+  isKeyFrame = true
+  presentationTimeUs = 441180
+sample:
+  trackIndex = 1
+  dataHashCode = 1742602241
+  size = 4096
+  isKeyFrame = true
+  presentationTimeUs = 464400
+sample:
+  trackIndex = 1
+  dataHashCode = 1742602241
+  size = 4096
+  isKeyFrame = true
+  presentationTimeUs = 487620
+sample:
+  trackIndex = 0
+  dataHashCode = -770308242
+  size = 36692
+  isKeyFrame = true
+  presentationTimeUs = 0
+sample:
+  trackIndex = 0
+  dataHashCode = -732087136
+  size = 5312
+  isKeyFrame = false
+  presentationTimeUs = 66733
+sample:
+  trackIndex = 0
+  dataHashCode = 468156717
+  size = 599
+  isKeyFrame = false
+  presentationTimeUs = 33366
+sample:
+  trackIndex = 0
+  dataHashCode = 1150349584
+  size = 7735
+  isKeyFrame = false
+  presentationTimeUs = 200200
+sample:
+  trackIndex = 0
+  dataHashCode = 1443582006
+  size = 987
+  isKeyFrame = false
+  presentationTimeUs = 133466
+sample:
+  trackIndex = 0
+  dataHashCode = -310585145
+  size = 673
+  isKeyFrame = false
+  presentationTimeUs = 100100
+sample:
+  trackIndex = 0
+  dataHashCode = 807460688
+  size = 523
+  isKeyFrame = false
+  presentationTimeUs = 166833
+sample:
+  trackIndex = 0
+  dataHashCode = 1936487090
+  size = 6061
+  isKeyFrame = false
+  presentationTimeUs = 333666
+sample:
+  trackIndex = 0
+  dataHashCode = -32297181
+  size = 992
+  isKeyFrame = false
+  presentationTimeUs = 266933
+sample:
+  trackIndex = 0
+  dataHashCode = 1529616406
+  size = 623
+  isKeyFrame = false
+  presentationTimeUs = 233566
+sample:
+  trackIndex = 1
+  dataHashCode = 1742602241
+  size = 4096
+  isKeyFrame = true
+  presentationTimeUs = 510840
+sample:
+  trackIndex = 1
+  dataHashCode = 1742602241
+  size = 4096
+  isKeyFrame = true
+  presentationTimeUs = 534059
+sample:
+  trackIndex = 1
+  dataHashCode = 1742602241
+  size = 4096
+  isKeyFrame = true
+  presentationTimeUs = 557279
+sample:
+  trackIndex = 1
+  dataHashCode = 1742602241
+  size = 4096
+  isKeyFrame = true
+  presentationTimeUs = 580499
+sample:
+  trackIndex = 1
+  dataHashCode = 1742602241
+  size = 4096
+  isKeyFrame = true
+  presentationTimeUs = 603719
+sample:
+  trackIndex = 1
+  dataHashCode = 1742602241
+  size = 4096
+  isKeyFrame = true
+  presentationTimeUs = 626939
+sample:
+  trackIndex = 1
+  dataHashCode = 1742602241
+  size = 4096
+  isKeyFrame = true
+  presentationTimeUs = 650159
+sample:
+  trackIndex = 1
+  dataHashCode = 1742602241
+  size = 4096
+  isKeyFrame = true
+  presentationTimeUs = 673379
+sample:
+  trackIndex = 1
+  dataHashCode = 1742602241
+  size = 4096
+  isKeyFrame = true
+  presentationTimeUs = 696599
+sample:
+  trackIndex = 1
+  dataHashCode = 1742602241
+  size = 4096
+  isKeyFrame = true
+  presentationTimeUs = 719819
+sample:
+  trackIndex = 1
+  dataHashCode = 1742602241
+  size = 4096
+  isKeyFrame = true
+  presentationTimeUs = 743039
+sample:
+  trackIndex = 1
+  dataHashCode = 1742602241
+  size = 4096
+  isKeyFrame = true
+  presentationTimeUs = 766259
+sample:
+  trackIndex = 1
+  dataHashCode = 1742602241
+  size = 4096
+  isKeyFrame = true
+  presentationTimeUs = 789479
+sample:
+  trackIndex = 1
+  dataHashCode = 1742602241
+  size = 4096
+  isKeyFrame = true
+  presentationTimeUs = 812699
+sample:
+  trackIndex = 0
+  dataHashCode = 1949198785
+  size = 421
+  isKeyFrame = false
+  presentationTimeUs = 300300
+sample:
+  trackIndex = 0
+  dataHashCode = -147880287
+  size = 4899
+  isKeyFrame = false
+  presentationTimeUs = 433766
+sample:
+  trackIndex = 0
+  dataHashCode = 1369083472
+  size = 568
+  isKeyFrame = false
+  presentationTimeUs = 400400
+sample:
+  trackIndex = 0
+  dataHashCode = 965782073
+  size = 620
+  isKeyFrame = false
+  presentationTimeUs = 367033
+sample:
+  trackIndex = 0
+  dataHashCode = -261176150
+  size = 5450
+  isKeyFrame = false
+  presentationTimeUs = 567233
+sample:
+  trackIndex = 0
+  dataHashCode = -1830836678
+  size = 1051
+  isKeyFrame = false
+  presentationTimeUs = 500500
+sample:
+  trackIndex = 0
+  dataHashCode = 1767407540
+  size = 874
+  isKeyFrame = false
+  presentationTimeUs = 467133
+sample:
+  trackIndex = 0
+  dataHashCode = 918440283
+  size = 781
+  isKeyFrame = false
+  presentationTimeUs = 533866
+sample:
+  trackIndex = 0
+  dataHashCode = -1408463661
+  size = 4725
+  isKeyFrame = false
+  presentationTimeUs = 700700
+sample:
+  trackIndex = 0
+  dataHashCode = 1569455924
+  size = 1022
+  isKeyFrame = false
+  presentationTimeUs = 633966
+sample:
+  trackIndex = 1
+  dataHashCode = 1742602241
+  size = 4096
+  isKeyFrame = true
+  presentationTimeUs = 835919
+sample:
+  trackIndex = 1
+  dataHashCode = 1742602241
+  size = 4096
+  isKeyFrame = true
+  presentationTimeUs = 859139
+sample:
+  trackIndex = 1
+  dataHashCode = 1742602241
+  size = 4096
+  isKeyFrame = true
+  presentationTimeUs = 882359
+sample:
+  trackIndex = 1
+  dataHashCode = 1742602241
+  size = 4096
+  isKeyFrame = true
+  presentationTimeUs = 905579
+sample:
+  trackIndex = 1
+  dataHashCode = 1742602241
+  size = 4096
+  isKeyFrame = true
+  presentationTimeUs = 928799
+sample:
+  trackIndex = 1
+  dataHashCode = 1742602241
+  size = 4096
+  isKeyFrame = true
+  presentationTimeUs = 952019
+sample:
+  trackIndex = 1
+  dataHashCode = 1742602241
+  size = 4096
+  isKeyFrame = true
+  presentationTimeUs = 975239
+sample:
+  trackIndex = 1
+  dataHashCode = 226248705
+  size = 448
+  isKeyFrame = true
+  presentationTimeUs = 998459
+sample:
+  trackIndex = 0
+  dataHashCode = -1723778407
+  size = 790
+  isKeyFrame = false
+  presentationTimeUs = 600600
+sample:
+  trackIndex = 0
+  dataHashCode = 1578275472
+  size = 610
+  isKeyFrame = false
+  presentationTimeUs = 667333
+sample:
+  trackIndex = 0
+  dataHashCode = 1989768395
+  size = 2751
+  isKeyFrame = false
+  presentationTimeUs = 834166
+sample:
+  trackIndex = 0
+  dataHashCode = -1215674502
+  size = 745
+  isKeyFrame = false
+  presentationTimeUs = 767433
+sample:
+  trackIndex = 0
+  dataHashCode = -814473606
+  size = 621
+  isKeyFrame = false
+  presentationTimeUs = 734066
+sample:
+  trackIndex = 0
+  dataHashCode = 498370894
+  size = 505
+  isKeyFrame = false
+  presentationTimeUs = 800800
+sample:
+  trackIndex = 0
+  dataHashCode = -1051506468
+  size = 1268
+  isKeyFrame = false
+  presentationTimeUs = 967633
+sample:
+  trackIndex = 0
+  dataHashCode = -1025604144
+  size = 880
+  isKeyFrame = false
+  presentationTimeUs = 900900
+sample:
+  trackIndex = 0
+  dataHashCode = -913586520
+  size = 530
+  isKeyFrame = false
+  presentationTimeUs = 867533
+sample:
+  trackIndex = 0
+  dataHashCode = 1340459242
+  size = 568
+  isKeyFrame = false
+  presentationTimeUs = 934266
+released = true

libraries/transformer/src/main/java/androidx/media3/transformer/AudioSamplePipeline.java

@@ -64,15 +64,15 @@ import org.checkerframework.dataflow.qual.Pure;
       long streamOffsetUs,
       TransformationRequest transformationRequest,
       ImmutableList<AudioProcessor> audioProcessors,
-      long forceSilentAudioDurationUs,
+      long generateSilentAudioDurationUs,
       Codec.EncoderFactory encoderFactory,
       MuxerWrapper muxerWrapper,
       FallbackListener fallbackListener)
       throws TransformationException {
     super(inputFormat, streamStartPositionUs, muxerWrapper);
 
-    if (forceSilentAudioDurationUs != C.TIME_UNSET) {
+    if (generateSilentAudioDurationUs != C.TIME_UNSET) {
-      silentAudioGenerator = new SilentAudioGenerator(inputFormat, forceSilentAudioDurationUs);
+      silentAudioGenerator = new SilentAudioGenerator(inputFormat, generateSilentAudioDurationUs);
     } else {
       silentAudioGenerator = null;
     }

libraries/transformer/src/main/java/androidx/media3/transformer/Transformer.java

@@ -88,7 +88,7 @@ public final class Transformer {
     private ImmutableList<Effect> videoEffects;
     private boolean removeAudio;
     private boolean removeVideo;
-    private boolean forceSilentAudio;
+    private boolean generateSilentAudio;
     private ListenerSet<Transformer.Listener> listeners;
     @Nullable private AssetLoader.Factory assetLoaderFactory;
     private FrameProcessor.Factory frameProcessorFactory;
@@ -125,7 +125,7 @@ public final class Transformer {
       this.videoEffects = transformer.videoEffects;
       this.removeAudio = transformer.removeAudio;
       this.removeVideo = transformer.removeVideo;
-      this.forceSilentAudio = transformer.forceSilentAudio;
+      this.generateSilentAudio = transformer.generateSilentAudio;
       this.listeners = transformer.listeners;
       this.assetLoaderFactory = transformer.assetLoaderFactory;
       this.frameProcessorFactory = transformer.frameProcessorFactory;
@@ -400,29 +400,32 @@ public final class Transformer {
     }
 
     /**
-     * Sets whether to force silent audio for the output file, ignoring any existing audio.
+     * Sets whether to generate silent audio for the output file, if there is no audio available.
      *
      * <p>This method is experimental and may be removed or changed without warning.
      *
+     * <p>To replace existing audio with silence, call {@link #setRemoveAudio(boolean)} as well.
+     *
      * <p>Audio properties/format:
      *
      * <ul>
      *   <li>Duration will match duration of the input media.
      *   <li>Sample mime type will match {@link TransformationRequest#audioMimeType}, or {@link
      *       MimeTypes#AUDIO_AAC} if {@code null}.
-     *   <li>Sample rate will be 44100hz. This can be modified by passing a {@link
+     *   <li>Sample rate will be {@code 44100} hz. This can be modified by passing a {@link
      *       SonicAudioProcessor} to {@link #setAudioProcessors(List)}, using {@link
      *       SonicAudioProcessor#setOutputSampleRateHz(int)}.
-     *   <li>Channel count will be 2. This can be modified by implementing a custom {@link
+     *   <li>Channel count will be {@code 2}. This can be modified by implementing a custom {@link
      *       AudioProcessor} and passing it to {@link #setAudioProcessors(List)}.
      * </ul>
      *
-     * @param forceSilentAudio Whether to output silent audio for the output file.
+     * @param generateSilentAudio Whether to generate silent audio for the output file if there is
+     *     no audio track.
      * @return This builder.
      */
     @CanIgnoreReturnValue
-    public Builder experimentalSetForceSilentAudio(boolean forceSilentAudio) {
+    public Builder experimentalSetGenerateSilentAudio(boolean generateSilentAudio) {
-      this.forceSilentAudio = forceSilentAudio;
+      this.generateSilentAudio = generateSilentAudio;
       return this;
     }
 
@@ -459,7 +462,7 @@ public final class Transformer {
           videoEffects,
           removeAudio,
           removeVideo,
-          forceSilentAudio,
+          generateSilentAudio,
           listeners,
           assetLoaderFactory,
           frameProcessorFactory,
@@ -579,7 +582,7 @@ public final class Transformer {
   private final ImmutableList<Effect> videoEffects;
   private final boolean removeAudio;
   private final boolean removeVideo;
-  private final boolean forceSilentAudio;
+  private final boolean generateSilentAudio;
   private final ListenerSet<Transformer.Listener> listeners;
   private final AssetLoader.Factory assetLoaderFactory;
   private final FrameProcessor.Factory frameProcessorFactory;
@@ -598,7 +601,7 @@ public final class Transformer {
       ImmutableList<Effect> videoEffects,
       boolean removeAudio,
       boolean removeVideo,
-      boolean forceSilentAudio,
+      boolean generateSilentAudio,
       ListenerSet<Listener> listeners,
       AssetLoader.Factory assetLoaderFactory,
       FrameProcessor.Factory frameProcessorFactory,
@@ -607,10 +610,6 @@ public final class Transformer {
       Looper looper,
       DebugViewProvider debugViewProvider,
       Clock clock) {
-    if (forceSilentAudio) {
-      removeAudio = true;
-    }
-    checkState(!removeVideo || !forceSilentAudio, "Silent only audio track needs a video track.");
     checkState(!removeAudio || !removeVideo, "Audio and video cannot both be removed.");
     this.context = context;
     this.transformationRequest = transformationRequest;
@@ -618,7 +617,7 @@ public final class Transformer {
     this.videoEffects = videoEffects;
     this.removeAudio = removeAudio;
     this.removeVideo = removeVideo;
-    this.forceSilentAudio = forceSilentAudio;
+    this.generateSilentAudio = generateSilentAudio;
     this.listeners = listeners;
     this.assetLoaderFactory = assetLoaderFactory;
     this.frameProcessorFactory = frameProcessorFactory;
@@ -761,7 +760,7 @@ public final class Transformer {
             videoEffects,
             removeAudio,
             removeVideo,
-            forceSilentAudio,
+            generateSilentAudio,
             assetLoaderFactory,
             frameProcessorFactory,
             encoderFactory,

libraries/transformer/src/main/java/androidx/media3/transformer/TransformerInternal.java

@@ -93,7 +93,6 @@ import org.checkerframework.checker.nullness.qual.MonotonicNonNull;
   private final TransformationRequest transformationRequest;
   private final ImmutableList<AudioProcessor> audioProcessors;
   private final ImmutableList<Effect> videoEffects;
-  private final boolean forceSilentAudio;
   private final FrameProcessor.Factory frameProcessorFactory;
   private final CapturingEncoderFactory encoderFactory;
   private final Listener listener;
@@ -108,6 +107,7 @@ import org.checkerframework.checker.nullness.qual.MonotonicNonNull;
   private final ConditionVariable transformerConditionVariable;
   private final TransformationResult.Builder transformationResultBuilder;
 
+  private boolean generateSilentAudio;
   private boolean isDrainingPipelines;
   private @Transformer.ProgressState int progressState;
   private @MonotonicNonNull RuntimeException cancelException;
@@ -124,7 +124,7 @@ import org.checkerframework.checker.nullness.qual.MonotonicNonNull;
       ImmutableList<Effect> videoEffects,
       boolean removeAudio,
       boolean removeVideo,
-      boolean forceSilentAudio,
+      boolean generateSilentAudio,
       AssetLoader.Factory assetLoaderFactory,
       FrameProcessor.Factory frameProcessorFactory,
       Codec.EncoderFactory encoderFactory,
@@ -138,7 +138,7 @@ import org.checkerframework.checker.nullness.qual.MonotonicNonNull;
     this.transformationRequest = transformationRequest;
     this.audioProcessors = audioProcessors;
     this.videoEffects = videoEffects;
-    this.forceSilentAudio = forceSilentAudio;
+    this.generateSilentAudio = generateSilentAudio;
     this.frameProcessorFactory = frameProcessorFactory;
     this.encoderFactory = new CapturingEncoderFactory(encoderFactory);
     this.listener = listener;
@@ -373,9 +373,6 @@ import org.checkerframework.checker.nullness.qual.MonotonicNonNull;
         return;
       }
       this.trackCount.set(trackCount);
-      if (forceSilentAudio) {
-        this.trackCount.incrementAndGet();
-      }
     }
 
     @Override
@@ -386,6 +383,14 @@ import org.checkerframework.checker.nullness.qual.MonotonicNonNull;
         long streamOffsetUs)
         throws TransformationException {
       if (!trackAdded) {
+        if (generateSilentAudio) {
+          if (this.trackCount.get() == 1 && MimeTypes.isVideo(format.sampleMimeType)) {
+            this.trackCount.incrementAndGet();
+          } else {
+            generateSilentAudio = false;
+          }
+        }
+
         // Call setTrackCount() methods here so that they are called from the same thread as the
         // MuxerWrapper and FallbackListener methods called when building the sample pipelines.
         muxerWrapper.setTrackCount(trackCount.get());
@@ -397,7 +402,7 @@ import org.checkerframework.checker.nullness.qual.MonotonicNonNull;
           getSamplePipeline(format, supportedOutputTypes, streamStartPositionUs, streamOffsetUs);
       internalHandler.obtainMessage(MSG_REGISTER_SAMPLE_PIPELINE, samplePipeline).sendToTarget();
 
-      if (forceSilentAudio) {
+      if (generateSilentAudio) {
         Format silentAudioFormat =
             new Format.Builder()
                 .setSampleMimeType(MimeTypes.AUDIO_AAC)
@@ -482,7 +487,7 @@ import org.checkerframework.checker.nullness.qual.MonotonicNonNull;
             streamOffsetUs,
             transformationRequest,
             audioProcessors,
-            forceSilentAudio ? durationUs : C.TIME_UNSET,
+            generateSilentAudio ? durationUs : C.TIME_UNSET,
             encoderFactory,
             muxerWrapper,
             fallbackListener);
@@ -528,7 +533,7 @@ import org.checkerframework.checker.nullness.qual.MonotonicNonNull;
       if (!audioProcessors.isEmpty()) {
         return true;
       }
-      if (forceSilentAudio) {
+      if (generateSilentAudio) {
         return true;
       }
 

libraries/transformer/src/test/java/androidx/media3/transformer/TransformerEndToEndTest.java

@@ -271,10 +271,40 @@ public final class TransformerEndToEndTest {
   }
 
   @Test
-  public void startTransformation_silentAudio_completesSuccessfully() throws Exception {
+  public void startTransformation_silentAudioOnAudioOnly_isIgnored() throws Exception {
     Transformer transformer =
         createTransformerBuilder(/* enableFallback= */ false)
-            .experimentalSetForceSilentAudio(true)
+            .experimentalSetGenerateSilentAudio(true)
+            .build();
+    MediaItem mediaItem = MediaItem.fromUri(ASSET_URI_PREFIX + FILE_AUDIO_UNSUPPORTED_BY_ENCODER);
+
+    transformer.startTransformation(mediaItem, outputPath);
+    TransformerTestRunner.runLooper(transformer);
+
+    DumpFileAsserts.assertOutput(
+        context, testMuxer, getDumpFileName(FILE_AUDIO_UNSUPPORTED_BY_ENCODER));
+  }
+
+  @Test
+  public void startTransformation_silentAudioOnAudioVideo_isIgnored() throws Exception {
+    Transformer transformer =
+        createTransformerBuilder(/* enableFallback= */ false)
+            .experimentalSetGenerateSilentAudio(true)
+            .build();
+    MediaItem mediaItem = MediaItem.fromUri(ASSET_URI_PREFIX + FILE_AUDIO_VIDEO);
+
+    transformer.startTransformation(mediaItem, outputPath);
+    TransformerTestRunner.runLooper(transformer);
+
+    DumpFileAsserts.assertOutput(context, testMuxer, getDumpFileName(FILE_AUDIO_VIDEO));
+  }
+
+  @Test
+  public void startTransformation_silentAudioRemoveAudio_completesSuccessfully() throws Exception {
+    Transformer transformer =
+        createTransformerBuilder(/* enableFallback= */ false)
+            .experimentalSetGenerateSilentAudio(true)
+            .setRemoveAudio(true)
             .build();
     MediaItem mediaItem = MediaItem.fromUri(ASSET_URI_PREFIX + FILE_AUDIO_VIDEO);
 
@@ -285,6 +315,36 @@ public final class TransformerEndToEndTest {
         context, testMuxer, getDumpFileName(FILE_AUDIO_VIDEO + ".silentaudio"));
   }
 
+  @Test
+  public void startTransformation_silentAudioRemoveVideo_isIgnored() throws Exception {
+    Transformer transformer =
+        createTransformerBuilder(/* enableFallback= */ false)
+            .experimentalSetGenerateSilentAudio(true)
+            .setRemoveVideo(true)
+            .build();
+    MediaItem mediaItem = MediaItem.fromUri(ASSET_URI_PREFIX + FILE_AUDIO_VIDEO);
+
+    transformer.startTransformation(mediaItem, outputPath);
+    TransformerTestRunner.runLooper(transformer);
+    DumpFileAsserts.assertOutput(
+        context, testMuxer, getDumpFileName(FILE_AUDIO_VIDEO + ".novideo"));
+  }
+
+  @Test
+  public void startTransformation_silentAudioOnVideoOnly_completesSuccessfully() throws Exception {
+    Transformer transformer =
+        createTransformerBuilder(/* enableFallback= */ false)
+            .experimentalSetGenerateSilentAudio(true)
+            .build();
+    MediaItem mediaItem = MediaItem.fromUri(ASSET_URI_PREFIX + FILE_VIDEO_ONLY);
+
+    transformer.startTransformation(mediaItem, outputPath);
+    TransformerTestRunner.runLooper(transformer);
+
+    DumpFileAsserts.assertOutput(
+        context, testMuxer, getDumpFileName(FILE_VIDEO_ONLY + ".silentaudio"));
+  }
+
   @Test
   public void startTransformation_adjustSampleRate_completesSuccessfully() throws Exception {
     SonicAudioProcessor sonicAudioProcessor = new SonicAudioProcessor();