tests: fix unit tests

src/pipecat/tests/utils.py

@@ -36,7 +36,7 @@ class SleepFrame(SystemFrame):
         sleep: Duration to sleep in seconds before processing the next frame.
     """
 
-    sleep: float = 0.1
+    sleep: float = 0.2
 
 
 class HeartbeatsObserver(BaseObserver):
@@ -100,7 +100,7 @@ class QueuedFrameProcessor(FrameProcessor):
             queue_direction: The direction of frames to capture (UPSTREAM or DOWNSTREAM).
             ignore_start: Whether to ignore StartFrames when capturing.
         """
-        super().__init__()
+        super().__init__(enable_direct_mode=True)
         self._queue = queue
         self._queue_direction = queue_direction
         self._ignore_start = ignore_start

tests/test_context_aggregators.py

@@ -102,8 +102,8 @@ class BaseTestUserContextAggregator:
         ]
         expected_down_frames = [
             UserStartedSpeakingFrame,
-            UserStoppedSpeakingFrame,
             *self.EXPECTED_CONTEXT_FRAMES,
+            UserStoppedSpeakingFrame,
         ]
         await run_test(
             aggregator,
@@ -127,8 +127,8 @@ class BaseTestUserContextAggregator:
         ]
         expected_down_frames = [
             UserStartedSpeakingFrame,
-            UserStoppedSpeakingFrame,
             *self.EXPECTED_CONTEXT_FRAMES,
+            UserStoppedSpeakingFrame,
         ]
         await run_test(
             aggregator,
@@ -158,8 +158,8 @@ class BaseTestUserContextAggregator:
             UserStartedSpeakingFrame,
             UserStoppedSpeakingFrame,
             UserStartedSpeakingFrame,
-            UserStoppedSpeakingFrame,
             *self.EXPECTED_CONTEXT_FRAMES,
+            UserStoppedSpeakingFrame,
         ]
         await run_test(
             aggregator,
@@ -298,8 +298,8 @@ class BaseTestUserContextAggregator:
         expected_down_frames = [
             SpeechControlParamsFrame,
             UserStartedSpeakingFrame,
-            UserStoppedSpeakingFrame,
             *self.EXPECTED_CONTEXT_FRAMES,
+            UserStoppedSpeakingFrame,
             *self.EXPECTED_CONTEXT_FRAMES,
         ]
         await run_test(

tests/test_transcript_processor.py

@@ -151,8 +151,8 @@ class TestUserTranscriptProcessor(unittest.IsolatedAsyncioTestCase):
             TTSTextFrame,
             TTSTextFrame,
             TTSTextFrame,
-            BotStoppedSpeakingFrame,
             TranscriptionUpdateFrame,
+            BotStoppedSpeakingFrame,
         ]
 
         # Run test
@@ -179,7 +179,7 @@ class TestUserTranscriptProcessor(unittest.IsolatedAsyncioTestCase):
         self.assertIsNotNone(message.timestamp)
 
         # All frames should be passed through in order, with update at end
-        downstream_update = cast(TranscriptionUpdateFrame, received_frames[-1])
+        downstream_update = cast(TranscriptionUpdateFrame, received_frames[-2])
         self.assertEqual(downstream_update.messages[0].content, "Hello world! How are you?")
 
     async def test_empty_text_handling(self):
@@ -257,8 +257,8 @@ class TestUserTranscriptProcessor(unittest.IsolatedAsyncioTestCase):
             BotStartedSpeakingFrame,
             TTSTextFrame,  # "New"
             TTSTextFrame,  # "response"
-            BotStoppedSpeakingFrame,
             TranscriptionUpdateFrame,  # Second message
+            BotStoppedSpeakingFrame,
         ]
 
         # Run test