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

frames: StartInterruptionFrame is deprecated, use InterruptionFrame

Browse Source
This commit is contained in:
Aleix Conchillo Flaqué
2025-09-08 19:07:25 -07:00
parent e27b23694d
commit 9d9f10ae0e
42 changed files with 135 additions and 108 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
CHANGELOG.md
View File

@@ -29,6 +29,8 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
### Deprecated
- `StartInterruptionFrame` is now deprected, use `InterruptionFrame` instead.
- Deprecate `VisionImageFrameAggregator` because `VisionImageRawFrame` has been
removed. See the `12*` examples for the new recommended replacement pattern.

4
examples/foundational/07c-interruptible-deepgram-vad.py
View File

@@ -12,8 +12,8 @@ from dotenv import load_dotenv
from loguru import logger
from pipecat.frames.frames import (
InterruptionFrame,
LLMRunFrame,
StartInterruptionFrame,
UserStartedSpeakingFrame,
UserStoppedSpeakingFrame,
)
@@ -97,7 +97,7 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
@stt.event_handler("on_speech_started")
async def on_speech_started(stt, *args, **kwargs):
await task.queue_frames([StartInterruptionFrame(), UserStartedSpeakingFrame()])
await task.queue_frames([InterruptionFrame(), UserStartedSpeakingFrame()])
@stt.event_handler("on_utterance_end")
async def on_utterance_end(stt, *args, **kwargs):

6
examples/foundational/07s-interruptible-google-audio-in.py
View File

@@ -16,10 +16,10 @@ from pipecat.audio.vad.silero import SileroVADAnalyzer
from pipecat.frames.frames import (
Frame,
InputAudioRawFrame,
InterruptionFrame,
LLMFullResponseEndFrame,
LLMFullResponseStartFrame,
LLMRunFrame,
StartInterruptionFrame,
TextFrame,
TranscriptionFrame,
UserStartedSpeakingFrame,
@@ -181,9 +181,7 @@ class TranscriptionContextFixup(FrameProcessor):
if isinstance(frame, MagicDemoTranscriptionFrame):
self._transcript = frame.text
elif isinstance(frame, LLMFullResponseEndFrame) or isinstance(
frame, StartInterruptionFrame
):
elif isinstance(frame, LLMFullResponseEndFrame) or isinstance(frame, InterruptionFrame):
self.swap_user_audio()
self.add_transcript_back_to_inference_output()
self._transcript = ""

6
examples/foundational/22b-natural-conversation-proposal.py
View File

@@ -18,9 +18,9 @@ from pipecat.frames.frames import (
Frame,
FunctionCallInProgressFrame,
FunctionCallResultFrame,
InterruptionFrame,
LLMRunFrame,
StartFrame,
StartInterruptionFrame,
SystemFrame,
TextFrame,
TranscriptionFrame,
@@ -144,7 +144,7 @@ class OutputGate(FrameProcessor):
await self._start()
if isinstance(frame, (EndFrame, CancelFrame)):
await self._stop()
if isinstance(frame, StartInterruptionFrame):
if isinstance(frame, InterruptionFrame):
self._frames_buffer = []
self.close_gate()
await self.push_frame(frame, direction)
@@ -232,7 +232,7 @@ class TurnDetectionLLM(Pipeline):
async def pass_only_llm_trigger_frames(frame):
return (
isinstance(frame, OpenAILLMContextFrame)
or isinstance(frame, StartInterruptionFrame)
or isinstance(frame, InterruptionFrame)
or isinstance(frame, FunctionCallInProgressFrame)
or isinstance(frame, FunctionCallResultFrame)
)

6
examples/foundational/22c-natural-conversation-mixed-llms.py
View File

@@ -18,9 +18,9 @@ from pipecat.frames.frames import (
Frame,
FunctionCallInProgressFrame,
FunctionCallResultFrame,
InterruptionFrame,
LLMRunFrame,
StartFrame,
StartInterruptionFrame,
SystemFrame,
TextFrame,
TranscriptionFrame,
@@ -347,7 +347,7 @@ class OutputGate(FrameProcessor):
await self._start()
if isinstance(frame, (EndFrame, CancelFrame)):
await self._stop()
if isinstance(frame, StartInterruptionFrame):
if isinstance(frame, InterruptionFrame):
self._frames_buffer = []
self.close_gate()
await self.push_frame(frame, direction)
@@ -426,7 +426,7 @@ class TurnDetectionLLM(Pipeline):
async def pass_only_llm_trigger_frames(frame):
return (
isinstance(frame, OpenAILLMContextFrame)
or isinstance(frame, StartInterruptionFrame)
or isinstance(frame, InterruptionFrame)
or isinstance(frame, FunctionCallInProgressFrame)
or isinstance(frame, FunctionCallResultFrame)
)

4
examples/foundational/22d-natural-conversation-gemini-audio.py
View File

@@ -20,10 +20,10 @@ from pipecat.frames.frames import (
FunctionCallInProgressFrame,
FunctionCallResultFrame,
InputAudioRawFrame,
InterruptionFrame,
LLMFullResponseStartFrame,
LLMRunFrame,
StartFrame,
StartInterruptionFrame,
SystemFrame,
TextFrame,
TranscriptionFrame,
@@ -570,7 +570,7 @@ class OutputGate(FrameProcessor):
await self._start()
if isinstance(frame, (EndFrame, CancelFrame)):
await self._stop()
if isinstance(frame, StartInterruptionFrame):
if isinstance(frame, InterruptionFrame):
self._frames_buffer = []
self.close_gate()
await self.push_frame(frame, direction)

4
examples/foundational/30-observer.py
View File

@@ -15,8 +15,8 @@ from pipecat.frames.frames import (
BotStartedSpeakingFrame,
BotStoppedSpeakingFrame,
EndFrame,
InterruptionFrame,
LLMRunFrame,
StartInterruptionFrame,
TTSTextFrame,
UserStartedSpeakingFrame,
)
@@ -69,7 +69,7 @@ class CustomObserver(BaseObserver):
# Create direction arrow
arrow = "→" if direction == FrameDirection.DOWNSTREAM else "←"
if isinstance(frame, StartInterruptionFrame) and isinstance(src, BaseOutputTransport):
if isinstance(frame, InterruptionFrame) and isinstance(src, BaseOutputTransport):
logger.info(f"⚡ INTERRUPTION START: {src} {arrow} {dst} at {time_sec:.2f}s")
elif isinstance(frame, BotStartedSpeakingFrame):
logger.info(f"🤖 BOT START SPEAKING: {src} {arrow} {dst} at {time_sec:.2f}s")

30
src/pipecat/frames/frames.py
View File

@@ -820,7 +820,7 @@ class FrameProcessorResumeUrgentFrame(SystemFrame):
@dataclass
class StartInterruptionFrame(SystemFrame):
class InterruptionFrame(SystemFrame):
"""Frame indicating user started speaking (interruption detected).
Emitted by the BaseInputTransport to indicate that a user has started
@@ -832,6 +832,34 @@ class StartInterruptionFrame(SystemFrame):
pass
@dataclass
class StartInterruptionFrame(InterruptionFrame):
"""Frame indicating user started speaking (interruption detected).
.. deprecated:: 0.0.85
This frame is deprecated and will be removed in a future version.
Instead, use `InterruptionFrame`.
Emitted by the BaseInputTransport to indicate that a user has started
speaking (i.e. is interrupting). This is similar to
UserStartedSpeakingFrame except that it should be pushed concurrently
with other frames (so the order is not guaranteed).
"""
def __post_init__(self):
super().__post_init__()
import warnings
with warnings.catch_warnings():
warnings.simplefilter("always")
warnings.warn(
"StartInterruptionFrame is deprecated and will be removed in a future version. "
"Instead, use InterruptionFrame.",
DeprecationWarning,
stacklevel=2,
)
@dataclass
class UserStartedSpeakingFrame(SystemFrame):
"""Frame indicating user has started speaking.

8
src/pipecat/processors/aggregators/llm_response.py
View File

@@ -36,6 +36,7 @@ from pipecat.frames.frames import (
FunctionCallsStartedFrame,
InputAudioRawFrame,
InterimTranscriptionFrame,
InterruptionFrame,
LLMFullResponseEndFrame,
LLMFullResponseStartFrame,
LLMMessagesAppendFrame,
@@ -48,7 +49,6 @@ from pipecat.frames.frames import (
OpenAILLMContextAssistantTimestampFrame,
SpeechControlParamsFrame,
StartFrame,
StartInterruptionFrame,
TextFrame,
TranscriptionFrame,
UserImageRawFrame,
@@ -138,7 +138,7 @@ class LLMFullResponseAggregator(FrameProcessor):
"""
await super().process_frame(frame, direction)
if isinstance(frame, StartInterruptionFrame):
if isinstance(frame, InterruptionFrame):
await self._call_event_handler("on_completion", self._aggregation, False)
self._aggregation = ""
self._started = False
@@ -838,7 +838,7 @@ class LLMAssistantContextAggregator(LLMContextResponseAggregator):
"""
await super().process_frame(frame, direction)
if isinstance(frame, StartInterruptionFrame):
if isinstance(frame, InterruptionFrame):
await self._handle_interruptions(frame)
await self.push_frame(frame, direction)
elif isinstance(frame, LLMFullResponseStartFrame):
@@ -904,7 +904,7 @@ class LLMAssistantContextAggregator(LLMContextResponseAggregator):
if frame.run_llm:
await self.push_context_frame(FrameDirection.UPSTREAM)
async def _handle_interruptions(self, frame: StartInterruptionFrame):
async def _handle_interruptions(self, frame: InterruptionFrame):
await self.push_aggregation()
self._started = 0
await self.reset()

6
src/pipecat/processors/aggregators/llm_response_universal.py
View File

@@ -37,6 +37,7 @@ from pipecat.frames.frames import (
FunctionCallsStartedFrame,
InputAudioRawFrame,
InterimTranscriptionFrame,
InterruptionFrame,
LLMContextAssistantTimestampFrame,
LLMContextFrame,
LLMFullResponseEndFrame,
@@ -48,7 +49,6 @@ from pipecat.frames.frames import (
LLMSetToolsFrame,
SpeechControlParamsFrame,
StartFrame,
StartInterruptionFrame,
TextFrame,
TranscriptionFrame,
UserImageRawFrame,
@@ -579,7 +579,7 @@ class LLMAssistantAggregator(LLMContextAggregator):
"""
await super().process_frame(frame, direction)
if isinstance(frame, StartInterruptionFrame):
if isinstance(frame, InterruptionFrame):
await self._handle_interruptions(frame)
await self.push_frame(frame, direction)
elif isinstance(frame, LLMFullResponseStartFrame):
@@ -645,7 +645,7 @@ class LLMAssistantAggregator(LLMContextAggregator):
if frame.run_llm:
await self.push_context_frame(FrameDirection.UPSTREAM)
async def _handle_interruptions(self, frame: StartInterruptionFrame):
async def _handle_interruptions(self, frame: InterruptionFrame):
await self._push_aggregation()
self._started = 0
await self.reset()

4
src/pipecat/processors/filters/stt_mute_filter.py
View File

@@ -25,8 +25,8 @@ from pipecat.frames.frames import (
FunctionCallResultFrame,
InputAudioRawFrame,
InterimTranscriptionFrame,
InterruptionFrame,
StartFrame,
StartInterruptionFrame,
STTMuteFrame,
TranscriptionFrame,
UserStartedSpeakingFrame,
@@ -204,7 +204,7 @@ class STTMuteFilter(FrameProcessor):
if isinstance(
frame,
(
StartInterruptionFrame,
InterruptionFrame,
VADUserStartedSpeakingFrame,
VADUserStoppedSpeakingFrame,
UserStartedSpeakingFrame,

4
src/pipecat/processors/frame_processor.py
View File

@@ -28,8 +28,8 @@ from pipecat.frames.frames import (
FrameProcessorPauseUrgentFrame,
FrameProcessorResumeFrame,
FrameProcessorResumeUrgentFrame,
InterruptionFrame,
StartFrame,
StartInterruptionFrame,
SystemFrame,
)
from pipecat.metrics.metrics import LLMTokenUsage, MetricsData
@@ -588,7 +588,7 @@ class FrameProcessor(BaseObject):
if isinstance(frame, StartFrame):
await self.__start(frame)
elif isinstance(frame, StartInterruptionFrame):
elif isinstance(frame, InterruptionFrame):
await self._start_interruption()
await self.stop_all_metrics()
elif isinstance(frame, CancelFrame):

4
src/pipecat/processors/transcript_processor.py
View File

@@ -19,7 +19,7 @@ from pipecat.frames.frames import (
CancelFrame,
EndFrame,
Frame,
StartInterruptionFrame,
InterruptionFrame,
TranscriptionFrame,
TranscriptionMessage,
TranscriptionUpdateFrame,
@@ -195,7 +195,7 @@ class AssistantTranscriptProcessor(BaseTranscriptProcessor):
"""
await super().process_frame(frame, direction)
if isinstance(frame, (StartInterruptionFrame, CancelFrame)):
if isinstance(frame, (InterruptionFrame, CancelFrame)):
# Push frame first otherwise our emitted transcription update frame
# might get cleaned up.
await self.push_frame(frame, direction)

4
src/pipecat/serializers/exotel.py
View File

@@ -20,8 +20,8 @@ from pipecat.frames.frames import (
Frame,
InputAudioRawFrame,
InputDTMFFrame,
InterruptionFrame,
StartFrame,
StartInterruptionFrame,
TransportMessageFrame,
TransportMessageUrgentFrame,
)
@@ -98,7 +98,7 @@ class ExotelFrameSerializer(FrameSerializer):
Returns:
Serialized data as string or bytes, or None if the frame isn't handled.
"""
if isinstance(frame, StartInterruptionFrame):
if isinstance(frame, InterruptionFrame):
answer = {"event": "clear", "streamSid": self._stream_sid}
return json.dumps(answer)
elif isinstance(frame, AudioRawFrame):

4
src/pipecat/serializers/plivo.py
View File

@@ -22,8 +22,8 @@ from pipecat.frames.frames import (
Frame,
InputAudioRawFrame,
InputDTMFFrame,
InterruptionFrame,
StartFrame,
StartInterruptionFrame,
TransportMessageFrame,
TransportMessageUrgentFrame,
)
@@ -122,7 +122,7 @@ class PlivoFrameSerializer(FrameSerializer):
self._hangup_attempted = True
await self._hang_up_call()
return None
elif isinstance(frame, StartInterruptionFrame):
elif isinstance(frame, InterruptionFrame):
answer = {"event": "clearAudio", "streamId": self._stream_id}
return json.dumps(answer)
elif isinstance(frame, AudioRawFrame):

4
src/pipecat/serializers/telnyx.py
View File

@@ -29,8 +29,8 @@ from pipecat.frames.frames import (
Frame,
InputAudioRawFrame,
InputDTMFFrame,
InterruptionFrame,
StartFrame,
StartInterruptionFrame,
)
from pipecat.serializers.base_serializer import FrameSerializer, FrameSerializerType
@@ -137,7 +137,7 @@ class TelnyxFrameSerializer(FrameSerializer):
self._hangup_attempted = True
await self._hang_up_call()
return None
elif isinstance(frame, StartInterruptionFrame):
elif isinstance(frame, InterruptionFrame):
answer = {"event": "clear"}
return json.dumps(answer)
elif isinstance(frame, AudioRawFrame):

4
src/pipecat/serializers/twilio.py
View File

@@ -22,8 +22,8 @@ from pipecat.frames.frames import (
Frame,
InputAudioRawFrame,
InputDTMFFrame,
InterruptionFrame,
StartFrame,
StartInterruptionFrame,
TransportMessageFrame,
TransportMessageUrgentFrame,
)
@@ -122,7 +122,7 @@ class TwilioFrameSerializer(FrameSerializer):
self._hangup_attempted = True
await self._hang_up_call()
return None
elif isinstance(frame, StartInterruptionFrame):
elif isinstance(frame, InterruptionFrame):
answer = {"event": "clear", "streamSid": self._stream_sid}
return json.dumps(answer)
elif isinstance(frame, AudioRawFrame):

4
src/pipecat/services/asyncai/tts.py
View File

@@ -20,8 +20,8 @@ from pipecat.frames.frames import (
EndFrame,
ErrorFrame,
Frame,
InterruptionFrame,
StartFrame,
StartInterruptionFrame,
TTSAudioRawFrame,
TTSStartedFrame,
TTSStoppedFrame,
@@ -275,7 +275,7 @@ class AsyncAITTSService(InterruptibleTTSService):
direction: The direction to push the frame.
"""
await super().push_frame(frame, direction)
if isinstance(frame, (TTSStoppedFrame, StartInterruptionFrame)):
if isinstance(frame, (TTSStoppedFrame, InterruptionFrame)):
self._started = False
async def _receive_messages(self):

4
src/pipecat/services/aws_nova_sonic/context.py
View File

@@ -21,13 +21,13 @@ from pipecat.frames.frames import (
DataFrame,
Frame,
FunctionCallResultFrame,
InterruptionFrame,
LLMFullResponseEndFrame,
LLMFullResponseStartFrame,
LLMMessagesAppendFrame,
LLMMessagesUpdateFrame,
LLMSetToolChoiceFrame,
LLMSetToolsFrame,
StartInterruptionFrame,
TextFrame,
UserImageRawFrame,
)
@@ -306,7 +306,7 @@ class AWSNovaSonicAssistantContextAggregator(OpenAIAssistantContextAggregator):
if isinstance(
frame,
(
StartInterruptionFrame,
InterruptionFrame,
LLMFullResponseStartFrame,
LLMFullResponseEndFrame,
TextFrame,

4
src/pipecat/services/cartesia/tts.py
View File

@@ -20,8 +20,8 @@ from pipecat.frames.frames import (
EndFrame,
ErrorFrame,
Frame,
InterruptionFrame,
StartFrame,
StartInterruptionFrame,
TTSAudioRawFrame,
TTSStartedFrame,
TTSStoppedFrame,
@@ -371,7 +371,7 @@ class CartesiaTTSService(AudioContextWordTTSService):
return self._websocket
raise Exception("Websocket not connected")
async def _handle_interruption(self, frame: StartInterruptionFrame, direction: FrameDirection):
async def _handle_interruption(self, frame: InterruptionFrame, direction: FrameDirection):
await super()._handle_interruption(frame, direction)
await self.stop_all_metrics()
if self._context_id:

8
src/pipecat/services/elevenlabs/tts.py
View File

@@ -25,9 +25,9 @@ from pipecat.frames.frames import (
EndFrame,
ErrorFrame,
Frame,
InterruptionFrame,
LLMFullResponseEndFrame,
StartFrame,
StartInterruptionFrame,
TTSAudioRawFrame,
TTSStartedFrame,
TTSStoppedFrame,
@@ -460,7 +460,7 @@ class ElevenLabsTTSService(AudioContextWordTTSService):
direction: The direction to push the frame.
"""
await super().push_frame(frame, direction)
if isinstance(frame, (TTSStoppedFrame, StartInterruptionFrame)):
if isinstance(frame, (TTSStoppedFrame, InterruptionFrame)):
self._started = False
if isinstance(frame, TTSStoppedFrame):
await self.add_word_timestamps([("Reset", 0)])
@@ -549,7 +549,7 @@ class ElevenLabsTTSService(AudioContextWordTTSService):
return self._websocket
raise Exception("Websocket not connected")
async def _handle_interruption(self, frame: StartInterruptionFrame, direction: FrameDirection):
async def _handle_interruption(self, frame: InterruptionFrame, direction: FrameDirection):
"""Handle interruption by closing the current context."""
await super()._handle_interruption(frame, direction)
@@ -856,7 +856,7 @@ class ElevenLabsHttpTTSService(WordTTSService):
direction: The direction to push the frame.
"""
await super().push_frame(frame, direction)
if isinstance(frame, (StartInterruptionFrame, TTSStoppedFrame)):
if isinstance(frame, (InterruptionFrame, TTSStoppedFrame)):
# Reset timing on interruption or stop
self._reset_state()

4
src/pipecat/services/fish/tts.py
View File

@@ -21,8 +21,8 @@ from pipecat.frames.frames import (
EndFrame,
ErrorFrame,
Frame,
InterruptionFrame,
StartFrame,
StartInterruptionFrame,
TTSAudioRawFrame,
TTSStartedFrame,
TTSStoppedFrame,
@@ -259,7 +259,7 @@ class FishAudioTTSService(InterruptibleTTSService):
return self._websocket
raise Exception("Websocket not connected")
async def _handle_interruption(self, frame: StartInterruptionFrame, direction: FrameDirection):
async def _handle_interruption(self, frame: InterruptionFrame, direction: FrameDirection):
await super()._handle_interruption(frame, direction)
await self.stop_all_metrics()
self._request_id = None

4
src/pipecat/services/gemini_multimodal_live/gemini.py
View File

@@ -33,6 +33,7 @@ from pipecat.frames.frames import (
InputAudioRawFrame,
InputImageRawFrame,
InputTextRawFrame,
InterruptionFrame,
LLMContextFrame,
LLMFullResponseEndFrame,
LLMFullResponseStartFrame,
@@ -41,7 +42,6 @@ from pipecat.frames.frames import (
LLMTextFrame,
LLMUpdateSettingsFrame,
StartFrame,
StartInterruptionFrame,
TranscriptionFrame,
TTSAudioRawFrame,
TTSStartedFrame,
@@ -752,7 +752,7 @@ class GeminiMultimodalLiveLLMService(LLMService):
elif isinstance(frame, InputImageRawFrame):
await self._send_user_video(frame)
await self.push_frame(frame, direction)
elif isinstance(frame, StartInterruptionFrame):
elif isinstance(frame, InterruptionFrame):
await self._handle_interruption()
await self.push_frame(frame, direction)
elif isinstance(frame, UserStartedSpeakingFrame):

6
src/pipecat/services/llm_service.py
View File

@@ -36,12 +36,12 @@ from pipecat.frames.frames import (
FunctionCallResultFrame,
FunctionCallResultProperties,
FunctionCallsStartedFrame,
InterruptionFrame,
LLMConfigureOutputFrame,
LLMFullResponseEndFrame,
LLMFullResponseStartFrame,
LLMTextFrame,
StartFrame,
StartInterruptionFrame,
UserImageRequestFrame,
)
from pipecat.processors.aggregators.llm_context import LLMContext
@@ -269,7 +269,7 @@ class LLMService(AIService):
"""
await super().process_frame(frame, direction)
if isinstance(frame, StartInterruptionFrame):
if isinstance(frame, InterruptionFrame):
await self._handle_interruptions(frame)
elif isinstance(frame, LLMConfigureOutputFrame):
self._skip_tts = frame.skip_tts
@@ -286,7 +286,7 @@ class LLMService(AIService):
await super().push_frame(frame, direction)
async def _handle_interruptions(self, _: StartInterruptionFrame):
async def _handle_interruptions(self, _: InterruptionFrame):
for function_name, entry in self._functions.items():
if entry.cancel_on_interruption:
await self._cancel_function_call(function_name)

4
src/pipecat/services/lmnt/tts.py
View File

@@ -16,8 +16,8 @@ from pipecat.frames.frames import (
EndFrame,
ErrorFrame,
Frame,
InterruptionFrame,
StartFrame,
StartInterruptionFrame,
TTSAudioRawFrame,
TTSStartedFrame,
TTSStoppedFrame,
@@ -180,7 +180,7 @@ class LmntTTSService(InterruptibleTTSService):
direction: The direction to push the frame.
"""
await super().push_frame(frame, direction)
if isinstance(frame, (TTSStoppedFrame, StartInterruptionFrame)):
if isinstance(frame, (TTSStoppedFrame, InterruptionFrame)):
self._started = False
async def _connect(self):

4
src/pipecat/services/neuphonic/tts.py
View File

@@ -25,9 +25,9 @@ from pipecat.frames.frames import (
EndFrame,
ErrorFrame,
Frame,
InterruptionFrame,
LLMFullResponseEndFrame,
StartFrame,
StartInterruptionFrame,
TTSAudioRawFrame,
TTSSpeakFrame,
TTSStartedFrame,
@@ -224,7 +224,7 @@ class NeuphonicTTSService(InterruptibleTTSService):
direction: The direction to push the frame.
"""
await super().push_frame(frame, direction)
if isinstance(frame, (TTSStoppedFrame, StartInterruptionFrame)):
if isinstance(frame, (TTSStoppedFrame, InterruptionFrame)):
self._started = False
async def process_frame(self, frame: Frame, direction: FrameDirection):

6
src/pipecat/services/openai_realtime/openai.py
View File

@@ -23,6 +23,7 @@ from pipecat.frames.frames import (
Frame,
InputAudioRawFrame,
InterimTranscriptionFrame,
InterruptionFrame,
LLMContextFrame,
LLMFullResponseEndFrame,
LLMFullResponseStartFrame,
@@ -31,7 +32,6 @@ from pipecat.frames.frames import (
LLMTextFrame,
LLMUpdateSettingsFrame,
StartFrame,
StartInterruptionFrame,
TranscriptionFrame,
TTSAudioRawFrame,
TTSStartedFrame,
@@ -366,7 +366,7 @@ class OpenAIRealtimeLLMService(LLMService):
elif isinstance(frame, InputAudioRawFrame):
if not self._audio_input_paused:
await self._send_user_audio(frame)
elif isinstance(frame, StartInterruptionFrame):
elif isinstance(frame, InterruptionFrame):
await self._handle_interruption()
elif isinstance(frame, UserStartedSpeakingFrame):
await self._handle_user_started_speaking(frame)
@@ -717,7 +717,7 @@ class OpenAIRealtimeLLMService(LLMService):
async def _handle_evt_speech_started(self, evt):
await self._truncate_current_audio_response()
await self._start_interruption() # cancels this processor task
await self.push_frame(StartInterruptionFrame()) # cancels downstream tasks
await self.push_frame(InterruptionFrame()) # cancels downstream tasks
await self.push_frame(UserStartedSpeakingFrame())
async def _handle_evt_speech_stopped(self, evt):

6
src/pipecat/services/openai_realtime_beta/openai.py
View File

@@ -24,6 +24,7 @@ from pipecat.frames.frames import (
Frame,
InputAudioRawFrame,
InterimTranscriptionFrame,
InterruptionFrame,
LLMContextFrame,
LLMFullResponseEndFrame,
LLMFullResponseStartFrame,
@@ -32,7 +33,6 @@ from pipecat.frames.frames import (
LLMTextFrame,
LLMUpdateSettingsFrame,
StartFrame,
StartInterruptionFrame,
TranscriptionFrame,
TTSAudioRawFrame,
TTSStartedFrame,
@@ -364,7 +364,7 @@ class OpenAIRealtimeBetaLLMService(LLMService):
elif isinstance(frame, InputAudioRawFrame):
if not self._audio_input_paused:
await self._send_user_audio(frame)
elif isinstance(frame, StartInterruptionFrame):
elif isinstance(frame, InterruptionFrame):
await self._handle_interruption()
elif isinstance(frame, UserStartedSpeakingFrame):
await self._handle_user_started_speaking(frame)
@@ -659,7 +659,7 @@ class OpenAIRealtimeBetaLLMService(LLMService):
async def _handle_evt_speech_started(self, evt):
await self._truncate_current_audio_response()
await self._start_interruption() # cancels this processor task
await self.push_frame(StartInterruptionFrame()) # cancels downstream tasks
await self.push_frame(InterruptionFrame()) # cancels downstream tasks
await self.push_frame(UserStartedSpeakingFrame())
async def _handle_evt_speech_stopped(self, evt):

4
src/pipecat/services/playht/tts.py
View File

@@ -25,8 +25,8 @@ from pipecat.frames.frames import (
EndFrame,
ErrorFrame,
Frame,
InterruptionFrame,
StartFrame,
StartInterruptionFrame,
TTSAudioRawFrame,
TTSStartedFrame,
TTSStoppedFrame,
@@ -312,7 +312,7 @@ class PlayHTTTSService(InterruptibleTTSService):
return self._websocket
raise Exception("Websocket not connected")
async def _handle_interruption(self, frame: StartInterruptionFrame, direction: FrameDirection):
async def _handle_interruption(self, frame: InterruptionFrame, direction: FrameDirection):
"""Handle interruption by stopping metrics and clearing request ID."""
await super()._handle_interruption(frame, direction)
await self.stop_all_metrics()

7
src/pipecat/services/rime/tts.py
View File

@@ -24,15 +24,14 @@ from pipecat.frames.frames import (
EndFrame,
ErrorFrame,
Frame,
InterruptionFrame,
StartFrame,
StartInterruptionFrame,
TTSAudioRawFrame,
TTSStartedFrame,
TTSStoppedFrame,
)
from pipecat.processors.frame_processor import FrameDirection
from pipecat.services.tts_service import AudioContextWordTTSService, TTSService
from pipecat.transcriptions import language
from pipecat.transcriptions.language import Language
from pipecat.utils.text.base_text_aggregator import BaseTextAggregator
from pipecat.utils.text.skip_tags_aggregator import SkipTagsAggregator
@@ -280,7 +279,7 @@ class RimeTTSService(AudioContextWordTTSService):
return self._websocket
raise Exception("Websocket not connected")
async def _handle_interruption(self, frame: StartInterruptionFrame, direction: FrameDirection):
async def _handle_interruption(self, frame: InterruptionFrame, direction: FrameDirection):
"""Handle interruption by clearing current context."""
await super()._handle_interruption(frame, direction)
await self.stop_all_metrics()
@@ -375,7 +374,7 @@ class RimeTTSService(AudioContextWordTTSService):
direction: The direction to push the frame.
"""
await super().push_frame(frame, direction)
if isinstance(frame, (TTSStoppedFrame, StartInterruptionFrame)):
if isinstance(frame, (TTSStoppedFrame, InterruptionFrame)):
if isinstance(frame, TTSStoppedFrame):
await self.add_word_timestamps([("Reset", 0)])

4
src/pipecat/services/sarvam/tts.py
View File

@@ -20,9 +20,9 @@ from pipecat.frames.frames import (
EndFrame,
ErrorFrame,
Frame,
InterruptionFrame,
LLMFullResponseEndFrame,
StartFrame,
StartInterruptionFrame,
TTSAudioRawFrame,
TTSStartedFrame,
TTSStoppedFrame,
@@ -455,7 +455,7 @@ class SarvamTTSService(InterruptibleTTSService):
direction: The direction to push the frame.
"""
await super().push_frame(frame, direction)
if isinstance(frame, (TTSStoppedFrame, StartInterruptionFrame)):
if isinstance(frame, (TTSStoppedFrame, InterruptionFrame)):
self._started = False
async def process_frame(self, frame: Frame, direction: FrameDirection):

4
src/pipecat/services/simli/video.py
View File

@@ -15,8 +15,8 @@ from pipecat.frames.frames import (
CancelFrame,
EndFrame,
Frame,
InterruptionFrame,
OutputImageRawFrame,
StartInterruptionFrame,
TTSAudioRawFrame,
TTSStoppedFrame,
UserStartedSpeakingFrame,
@@ -179,7 +179,7 @@ class SimliVideoService(FrameProcessor):
return
elif isinstance(frame, (EndFrame, CancelFrame)):
await self._stop()
elif isinstance(frame, (StartInterruptionFrame, UserStartedSpeakingFrame)):
elif isinstance(frame, (InterruptionFrame, UserStartedSpeakingFrame)):
if not self._previously_interrupted:
await self._simli_client.clearBuffer()
self._previously_interrupted = self._is_trinity_avatar

4
src/pipecat/services/tavus/video.py
View File

@@ -23,12 +23,12 @@ from pipecat.frames.frames import (
CancelFrame,
EndFrame,
Frame,
InterruptionFrame,
OutputAudioRawFrame,
OutputImageRawFrame,
OutputTransportReadyFrame,
SpeechOutputAudioRawFrame,
StartFrame,
StartInterruptionFrame,
TTSAudioRawFrame,
TTSStartedFrame,
)
@@ -222,7 +222,7 @@ class TavusVideoService(AIService):
"""
await super().process_frame(frame, direction)
if isinstance(frame, StartInterruptionFrame):
if isinstance(frame, InterruptionFrame):
await self._handle_interruptions()
await self.push_frame(frame, direction)
elif isinstance(frame, TTSAudioRawFrame):

18
src/pipecat/services/tts_service.py
View File

@@ -20,10 +20,10 @@ from pipecat.frames.frames import (
ErrorFrame,
Frame,
InterimTranscriptionFrame,
InterruptionFrame,
LLMFullResponseEndFrame,
LLMFullResponseStartFrame,
StartFrame,
StartInterruptionFrame,
TextFrame,
TranscriptionFrame,
TTSAudioRawFrame,
@@ -309,7 +309,7 @@ class TTSService(AIService):
and not isinstance(frame, TranscriptionFrame)
):
await self._process_text_frame(frame)
elif isinstance(frame, StartInterruptionFrame):
elif isinstance(frame, InterruptionFrame):
await self._handle_interruption(frame, direction)
await self.push_frame(frame, direction)
elif isinstance(frame, (LLMFullResponseEndFrame, EndFrame)):
@@ -367,14 +367,14 @@ class TTSService(AIService):
await super().push_frame(frame, direction)
if self._push_stop_frames and (
isinstance(frame, StartInterruptionFrame)
isinstance(frame, InterruptionFrame)
or isinstance(frame, TTSStartedFrame)
or isinstance(frame, TTSAudioRawFrame)
or isinstance(frame, TTSStoppedFrame)
):
await self._stop_frame_queue.put(frame)
async def _handle_interruption(self, frame: StartInterruptionFrame, direction: FrameDirection):
async def _handle_interruption(self, frame: InterruptionFrame, direction: FrameDirection):
self._processing_text = False
await self._text_aggregator.handle_interruption()
for filter in self._text_filters:
@@ -438,7 +438,7 @@ class TTSService(AIService):
)
if isinstance(frame, TTSStartedFrame):
has_started = True
elif isinstance(frame, (TTSStoppedFrame, StartInterruptionFrame)):
elif isinstance(frame, (TTSStoppedFrame, InterruptionFrame)):
has_started = False
except asyncio.TimeoutError:
if has_started:
@@ -523,7 +523,7 @@ class WordTTSService(TTSService):
elif isinstance(frame, (LLMFullResponseEndFrame, EndFrame)):
await self.flush_audio()
async def _handle_interruption(self, frame: StartInterruptionFrame, direction: FrameDirection):
async def _handle_interruption(self, frame: InterruptionFrame, direction: FrameDirection):
await super()._handle_interruption(frame, direction)
self._llm_response_started = False
self.reset_word_timestamps()
@@ -613,7 +613,7 @@ class InterruptibleTTSService(WebsocketTTSService):
# user interrupts we need to reconnect.
self._bot_speaking = False
async def _handle_interruption(self, frame: StartInterruptionFrame, direction: FrameDirection):
async def _handle_interruption(self, frame: InterruptionFrame, direction: FrameDirection):
await super()._handle_interruption(frame, direction)
if self._bot_speaking:
await self._disconnect()
@@ -685,7 +685,7 @@ class InterruptibleWordTTSService(WebsocketWordTTSService):
# user interrupts we need to reconnect.
self._bot_speaking = False
async def _handle_interruption(self, frame: StartInterruptionFrame, direction: FrameDirection):
async def _handle_interruption(self, frame: InterruptionFrame, direction: FrameDirection):
await super()._handle_interruption(frame, direction)
if self._bot_speaking:
await self._disconnect()
@@ -813,7 +813,7 @@ class AudioContextWordTTSService(WebsocketWordTTSService):
await super().cancel(frame)
await self._stop_audio_context_task()
async def _handle_interruption(self, frame: StartInterruptionFrame, direction: FrameDirection):
async def _handle_interruption(self, frame: InterruptionFrame, direction: FrameDirection):
await super()._handle_interruption(frame, direction)
await self._stop_audio_context_task()
self._create_audio_context_task()

6
src/pipecat/transports/base_input.py
View File

@@ -33,10 +33,10 @@ from pipecat.frames.frames import (
Frame,
InputAudioRawFrame,
InputImageRawFrame,
InterruptionFrame,
MetricsFrame,
SpeechControlParamsFrame,
StartFrame,
StartInterruptionFrame,
StopFrame,
SystemFrame,
UserSpeakingFrame,
@@ -340,7 +340,7 @@ class BaseInputTransport(FrameProcessor):
logger.debug("Bot interruption")
if self.interruptions_allowed:
await self._start_interruption()
await self.push_frame(StartInterruptionFrame())
await self.push_frame(InterruptionFrame())
async def _handle_user_interruption(self, vad_state: VADState, emulated: bool = False):
"""Handle user interruption events based on speaking state."""
@@ -366,7 +366,7 @@ class BaseInputTransport(FrameProcessor):
# Push an out-of-band frame (i.e. not using the ordered push
# frame task) to stop everything, specially at the output
# transport.
await self.push_frame(StartInterruptionFrame())
await self.push_frame(InterruptionFrame())
elif self.interruption_strategies and self._bot_speaking:
logger.debug(
"User started speaking while bot is speaking with interruption config - "

8
src/pipecat/transports/base_output.py
View File

@@ -30,6 +30,7 @@ from pipecat.frames.frames import (
EndFrame,
Frame,
InputTransportMessageUrgentFrame,
InterruptionFrame,
MixerControlFrame,
OutputAudioRawFrame,
OutputDTMFFrame,
@@ -39,7 +40,6 @@ from pipecat.frames.frames import (
SpeechOutputAudioRawFrame,
SpriteFrame,
StartFrame,
StartInterruptionFrame,
SystemFrame,
TransportMessageFrame,
TransportMessageUrgentFrame,
@@ -299,7 +299,7 @@ class BaseOutputTransport(FrameProcessor):
elif isinstance(frame, CancelFrame):
await self.cancel(frame)
await self.push_frame(frame, direction)
elif isinstance(frame, StartInterruptionFrame):
elif isinstance(frame, InterruptionFrame):
await self.push_frame(frame, direction)
await self._handle_frame(frame)
elif isinstance(frame, TransportMessageUrgentFrame) and not isinstance(
@@ -340,7 +340,7 @@ class BaseOutputTransport(FrameProcessor):
sender = self._media_senders[frame.transport_destination]
if isinstance(frame, StartInterruptionFrame):
if isinstance(frame, InterruptionFrame):
await sender.handle_interruptions(frame)
elif isinstance(frame, OutputAudioRawFrame):
await sender.handle_audio_frame(frame)
@@ -491,7 +491,7 @@ class BaseOutputTransport(FrameProcessor):
await self._cancel_clock_task()
await self._cancel_video_task()
async def handle_interruptions(self, _: StartInterruptionFrame):
async def handle_interruptions(self, _: InterruptionFrame):
"""Handle interruption events by restarting tasks and clearing buffers.
Args:

4
src/pipecat/transports/tavus/transport.py
View File

@@ -25,9 +25,9 @@ from pipecat.frames.frames import (
EndFrame,
Frame,
InputAudioRawFrame,
InterruptionFrame,
OutputAudioRawFrame,
StartFrame,
StartInterruptionFrame,
TransportMessageFrame,
TransportMessageUrgentFrame,
)
@@ -618,7 +618,7 @@ class TavusOutputTransport(BaseOutputTransport):
direction: The direction of frame flow in the pipeline.
"""
await super().process_frame(frame, direction)
if isinstance(frame, StartInterruptionFrame):
if isinstance(frame, InterruptionFrame):
await self._handle_interruptions()
async def _handle_interruptions(self):

4
src/pipecat/transports/websocket/fastapi.py
View File

@@ -26,9 +26,9 @@ from pipecat.frames.frames import (
EndFrame,
Frame,
InputAudioRawFrame,
InterruptionFrame,
OutputAudioRawFrame,
StartFrame,
StartInterruptionFrame,
TransportMessageFrame,
TransportMessageUrgentFrame,
)
@@ -398,7 +398,7 @@ class FastAPIWebsocketOutputTransport(BaseOutputTransport):
"""
await super().process_frame(frame, direction)
if isinstance(frame, StartInterruptionFrame):
if isinstance(frame, InterruptionFrame):
await self._write_frame(frame)
self._next_send_time = 0

4
src/pipecat/transports/websocket/server.py
View File

@@ -25,9 +25,9 @@ from pipecat.frames.frames import (
EndFrame,
Frame,
InputAudioRawFrame,
InterruptionFrame,
OutputAudioRawFrame,
StartFrame,
StartInterruptionFrame,
TransportMessageFrame,
TransportMessageUrgentFrame,
)
@@ -334,7 +334,7 @@ class WebsocketServerOutputTransport(BaseOutputTransport):
"""
await super().process_frame(frame, direction)
if isinstance(frame, StartInterruptionFrame):
if isinstance(frame, InterruptionFrame):
await self._write_frame(frame)
self._next_send_time = 0

6
tests/test_context_aggregators.py
View File

@@ -17,11 +17,11 @@ from pipecat.frames.frames import (
FunctionCallResultFrame,
FunctionCallResultProperties,
InterimTranscriptionFrame,
InterruptionFrame,
LLMFullResponseEndFrame,
LLMFullResponseStartFrame,
OpenAILLMContextAssistantTimestampFrame,
SpeechControlParamsFrame,
StartInterruptionFrame,
TextFrame,
TranscriptionFrame,
UserStartedSpeakingFrame,
@@ -618,7 +618,7 @@ class BaseTestAssistantContextAggreagator:
TextFrame(text="Pipecat."),
LLMFullResponseEndFrame(),
SleepFrame(AGGREGATION_SLEEP),
StartInterruptionFrame(),
InterruptionFrame(),
LLMFullResponseStartFrame(),
TextFrame(text="How are "),
TextFrame(text="you?"),
@@ -626,7 +626,7 @@ class BaseTestAssistantContextAggreagator:
]
expected_down_frames = [
*self.EXPECTED_CONTEXT_FRAMES,
StartInterruptionFrame,
InterruptionFrame,
*self.EXPECTED_CONTEXT_FRAMES,
]
await run_test(

6
tests/test_llm_response.py
View File

@@ -7,10 +7,10 @@
import unittest
from pipecat.frames.frames import (
InterruptionFrame,
LLMFullResponseEndFrame,
LLMFullResponseStartFrame,
LLMTextFrame,
StartInterruptionFrame,
)
from pipecat.processors.aggregators.llm_response import LLMFullResponseAggregator
from pipecat.tests.utils import SleepFrame, run_test
@@ -113,7 +113,7 @@ class TestLLMFullResponseAggregator(unittest.IsolatedAsyncioTestCase):
LLMFullResponseStartFrame(),
LLMTextFrame("Hello "),
SleepFrame(),
StartInterruptionFrame(),
InterruptionFrame(),
LLMFullResponseStartFrame(),
LLMTextFrame("Hello "),
LLMTextFrame("there!"),
@@ -122,7 +122,7 @@ class TestLLMFullResponseAggregator(unittest.IsolatedAsyncioTestCase):
expected_down_frames = [
LLMFullResponseStartFrame,
LLMTextFrame,
StartInterruptionFrame,
InterruptionFrame,
LLMFullResponseStartFrame,
LLMTextFrame,
LLMTextFrame,

6
tests/test_transcript_processor.py
View File

@@ -14,7 +14,7 @@ from pipecat.frames.frames import (
BotStartedSpeakingFrame,
BotStoppedSpeakingFrame,
CancelFrame,
StartInterruptionFrame,
InterruptionFrame,
TranscriptionFrame,
TranscriptionMessage,
TranscriptionUpdateFrame,
@@ -238,7 +238,7 @@ class TestUserTranscriptProcessor(unittest.IsolatedAsyncioTestCase):
TTSTextFrame(text="Hello"),
TTSTextFrame(text="world!"),
SleepFrame(),
StartInterruptionFrame(), # User interrupts here
InterruptionFrame(), # User interrupts here
SleepFrame(),
BotStartedSpeakingFrame(),
TTSTextFrame(text="New"),
@@ -252,7 +252,7 @@ class TestUserTranscriptProcessor(unittest.IsolatedAsyncioTestCase):
BotStartedSpeakingFrame,
TTSTextFrame, # "Hello"
TTSTextFrame, # "world!"
StartInterruptionFrame,
InterruptionFrame,
TranscriptionUpdateFrame, # First message (emitted due to interruption)
BotStartedSpeakingFrame,
TTSTextFrame, # "New"