wx44wx/pipecat
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

user started/stopped speaking frames and interruption frames

Browse Source
This commit is contained in:
Kwindla Hultman Kramer
2024-10-14 16:07:04 -07:00
parent d2ae82fb38
commit d4269acd67
1 changed files with 9 additions and 10 deletions
Download Patch File Download Diff File Expand all files Collapse all files

19
src/pipecat/services/openai_realtime_beta/llm_and_context.py
View File

@@ -248,7 +248,6 @@ class OpenAILLMServiceRealtimeBeta(LLMService):
session_properties: events.SessionProperties = events.SessionProperties(), session_properties: events.SessionProperties = events.SessionProperties(),
start_audio_paused: bool = False, start_audio_paused: bool = False,
send_transcription_frames: bool = True, send_transcription_frames: bool = True,
send_user_started_speaking_frames: bool = False,
**kwargs, **kwargs,
): ):
super().__init__(base_url=base_url, **kwargs) super().__init__(base_url=base_url, **kwargs)
@@ -258,8 +257,6 @@ class OpenAILLMServiceRealtimeBeta(LLMService):
self._session_properties: events.SessionProperties = session_properties self._session_properties: events.SessionProperties = session_properties
self._audio_input_paused = start_audio_paused self._audio_input_paused = start_audio_paused
self._send_transcription_frames = send_transcription_frames self._send_transcription_frames = send_transcription_frames
# todo: wire _send_user_started_speaking_frames up correctly
self._send_user_started_speaking_frames = send_user_started_speaking_frames
self._websocket = None self._websocket = None
self._receive_task = None self._receive_task = None
self._context = None self._context = None
@@ -336,6 +333,7 @@ class OpenAILLMServiceRealtimeBeta(LLMService):
) )
) )
# #
# frame processing # frame processing
# #
@@ -608,17 +606,18 @@ class OpenAILLMServiceRealtimeBeta(LLMService):
async def _handle_evt_speech_started(self, evt): async def _handle_evt_speech_started(self, evt):
await self._truncate_current_audio_response() await self._truncate_current_audio_response()
if self._send_user_started_speaking_frames: # todo: might need to guard sending these when we fully support using either openai
await self.push_frame(UserStartedSpeakingFrame()) # turn detection of Pipecat turn detection
await self.push_frame(StartInterruptionFrame()) await self._start_interruption() # cancels this processor task
logger.debug("User started speaking") await self.push_frame(StartInterruptionFrame()) # cancels downstream tasks
await self.push_frame(UserStartedSpeakingFrame())
async def _handle_evt_speech_stopped(self, evt): async def _handle_evt_speech_stopped(self, evt):
await self.start_ttfb_metrics() await self.start_ttfb_metrics()
await self.start_processing_metrics() await self.start_processing_metrics()
if self._send_user_started_speaking_frames: await self._stop_interruption()
await self.push_frame(UserStoppedSpeakingFrame()) await self.push_frame(StopInterruptionFrame())
await self.push_frame(StopInterruptionFrame()) await self.push_frame(UserStoppedSpeakingFrame())
async def _handle_evt_error(self, evt): async def _handle_evt_error(self, evt):
# Errors are fatal to this connection. Send an ErrorFrame. # Errors are fatal to this connection. Send an ErrorFrame.