services(cartesia): schedule TTSStoppedFrame after text

src/pipecat/services/ai_services.py

@@ -434,6 +434,9 @@ class WordTTSService(TTSService):
                 if word == "LLMFullResponseEndFrame" and timestamp == 0:
                     frame = LLMFullResponseEndFrame()
                     frame.pts = last_pts
+                elif word == "TTSStoppedFrame" and timestamp == 0:
+                    frame = TTSStoppedFrame()
+                    frame.pts = last_pts
                 else:
                     frame = TextFrame(word)
                     frame.pts = self._initial_word_timestamp + timestamp

src/pipecat/services/cartesia.py

@@ -227,12 +227,13 @@ class CartesiaTTSService(WordTTSService):
                     continue
                 if msg["type"] == "done":
                     await self.stop_ttfb_metrics()
-                    await self.push_frame(TTSStoppedFrame())
                     # Unset _context_id but not the _context_id_start_timestamp
                     # because we are likely still playing out audio and need the
                     # timestamp to set send context frames.
                     self._context_id = None
-                    await self.add_word_timestamps([("LLMFullResponseEndFrame", 0)])
+                    await self.add_word_timestamps(
+                        [("TTSStoppedFrame", 0), ("LLMFullResponseEndFrame", 0)]
+                    )
                 elif msg["type"] == "timestamps":
                     await self.add_word_timestamps(
                         list(zip(msg["word_timestamps"]["words"], msg["word_timestamps"]["start"]))