diff --git a/RELEASENOTES.md b/RELEASENOTES.md
index abfcc3a15b..f13f7a5361 100644
--- a/RELEASENOTES.md
+++ b/RELEASENOTES.md
@@ -69,6 +69,8 @@
     *   Add experimental support for parsing subtitles during extraction. You
         can enable this using
         `HlsMediaSource.Factory.experimentalParseSubtitlesDuringExtraction()`.
+        This works for standalone WebVTT subtitles and IMSC1 subtitles muxed
+        into MP4.
 *   DASH Extension:
     *   Extend experimental support for parsing subtitles during extraction to
         work with standalone text files (previously it only worked with
diff --git a/libraries/exoplayer_hls/src/main/java/androidx/media3/exoplayer/hls/DefaultHlsExtractorFactory.java b/libraries/exoplayer_hls/src/main/java/androidx/media3/exoplayer/hls/DefaultHlsExtractorFactory.java
index c607d15c45..fc88fd3021 100644
--- a/libraries/exoplayer_hls/src/main/java/androidx/media3/exoplayer/hls/DefaultHlsExtractorFactory.java
+++ b/libraries/exoplayer_hls/src/main/java/androidx/media3/exoplayer/hls/DefaultHlsExtractorFactory.java
@@ -201,7 +201,11 @@ public final class DefaultHlsExtractorFactory implements HlsExtractorFactory {
       case FileTypes.MP3:
         return new Mp3Extractor(/* flags= */ 0, /* forcedFirstSampleTimestampUs= */ 0);
       case FileTypes.MP4:
-        return createFragmentedMp4Extractor(timestampAdjuster, format, muxedCaptionFormats);
+        Extractor mp4Extractor =
+            createFragmentedMp4Extractor(timestampAdjuster, format, muxedCaptionFormats);
+        return subtitleParserFactory != null
+            ? new SubtitleTranscodingExtractor(mp4Extractor, subtitleParserFactory)
+            : mp4Extractor;
       case FileTypes.TS:
         return createTsExtractor(
             payloadReaderFactoryFlags,
diff --git a/libraries/exoplayer_hls/src/test/java/androidx/media3/exoplayer/hls/e2etest/HlsPlaybackTest.java b/libraries/exoplayer_hls/src/test/java/androidx/media3/exoplayer/hls/e2etest/HlsPlaybackTest.java
index 1ecb12e2ac..58f640c8b5 100644
--- a/libraries/exoplayer_hls/src/test/java/androidx/media3/exoplayer/hls/e2etest/HlsPlaybackTest.java
+++ b/libraries/exoplayer_hls/src/test/java/androidx/media3/exoplayer/hls/e2etest/HlsPlaybackTest.java
@@ -32,7 +32,6 @@ import androidx.media3.test.utils.robolectric.ShadowMediaCodecConfig;
 import androidx.media3.test.utils.robolectric.TestPlayerRunHelper;
 import androidx.test.core.app.ApplicationProvider;
 import androidx.test.ext.junit.runners.AndroidJUnit4;
-import org.junit.Ignore;
 import org.junit.Rule;
 import org.junit.Test;
 import org.junit.runner.RunWith;
@@ -72,14 +71,15 @@ public final class HlsPlaybackTest {
   }
 
   @Test
-  @Ignore(
-      "Disabled until subtitles are reliably asserted in robolectric tests [internal b/174661563].")
   public void ttmlInMp4() throws Exception {
     Context applicationContext = ApplicationProvider.getApplicationContext();
     CapturingRenderersFactory capturingRenderersFactory =
         new CapturingRenderersFactory(applicationContext);
     ExoPlayer player =
         new ExoPlayer.Builder(applicationContext, capturingRenderersFactory)
+            .setMediaSourceFactory(
+                new HlsMediaSource.Factory(new DefaultDataSource.Factory(applicationContext))
+                    .experimentalParseSubtitlesDuringExtraction(true))
             .setClock(new FakeClock(/* isAutoAdvancing= */ true))
             .build();
     player.setVideoSurface(new Surface(new SurfaceTexture(/* texName= */ 1)));
diff --git a/libraries/test_data/src/test/assets/playbackdumps/hls/ttml-in-mp4.dump b/libraries/test_data/src/test/assets/playbackdumps/hls/ttml-in-mp4.dump
index 55283c8d1f..98332c6cd4 100644
--- a/libraries/test_data/src/test/assets/playbackdumps/hls/ttml-in-mp4.dump
+++ b/libraries/test_data/src/test/assets/playbackdumps/hls/ttml-in-mp4.dump
@@ -222,22 +222,26 @@ TextOutput:
     presentationTimeUs = 0
     Cues = []
   Subtitle[1]:
+    presentationTimeUs = 20000
+    Cue[0]:
+      text = First Subtitle after 20ms of silence
+  Subtitle[2]:
     presentationTimeUs = 250000
     Cue[0]:
       text = First Subtitle after 20ms of silence
              Second Subtitle overlaps with first
-  Subtitle[2]:
+  Subtitle[3]:
     presentationTimeUs = 400000
     Cue[0]:
       text = First Subtitle after 20ms of silence
-  Subtitle[3]:
+  Subtitle[4]:
     presentationTimeUs = 500000
     Cues = []
-  Subtitle[4]:
+  Subtitle[5]:
     presentationTimeUs = 700000
     Cue[0]:
       text = Last subtitle
              And More
-  Subtitle[5]:
+  Subtitle[6]:
     presentationTimeUs = 950000
     Cues = []