diff --git a/changelog/4447.fixed.md b/changelog/4447.fixed.md new file mode 100644 index 000000000..bec01f4f8 --- /dev/null +++ b/changelog/4447.fixed.md @@ -0,0 +1 @@ +- Extended the `cancel_on_interruption=False` regression fix to `GrokRealtimeLLMService` and `AzureRealtimeLLMService` (the latter picks up the fix transitively by inheriting from `OpenAIRealtimeLLMService`). Same shape as the original fix for `AWSNovaSonicLLMService` and `OpenAIRealtimeLLMService`: each service now detects async-tool messages in the LLM context and routes the final result to its formal tool-result channel. Streamed intermediate results (`FunctionCallResultProperties(is_final=False)`) are not supported on these realtime services. `UltravoxRealtimeLLMService` now logs a one-time warning when async-tool messages appear in the context, since Ultravox freezes the conversation during tool execution and so the "keep talking while the tool runs" intent of `cancel_on_interruption=False` is structurally not achievable there. `GeminiLiveLLMService` and `InworldRealtimeLLMService` are excluded for now: Gemini Live's async-tool path needs deeper investigation, and Inworld appears to have a pre-existing problem with even simple tool calling on its Realtime API. diff --git a/examples/realtime/realtime-azure-async-tool.py b/examples/realtime/realtime-azure-async-tool.py new file mode 100644 index 000000000..46e561e0a --- /dev/null +++ b/examples/realtime/realtime-azure-async-tool.py @@ -0,0 +1,195 @@ +# +# Copyright (c) 2024-2026, Daily +# +# SPDX-License-Identifier: BSD 2-Clause License +# + +"""Example: async function call with the Azure Realtime LLM service. + +The ``get_current_weather`` tool is registered with +``cancel_on_interruption=False`` and simulates a slow API call (10s sleep). +While the call is in flight the conversation continues; the result arrives +later via the async-tool mechanism and is forwarded to Azure Realtime as a +``function_call_output`` so the model can integrate it naturally into its +next turn. +""" + +import asyncio +import os +import random +from datetime import datetime + +from dotenv import load_dotenv +from loguru import logger + +from pipecat.adapters.schemas.function_schema import FunctionSchema +from pipecat.adapters.schemas.tools_schema import ToolsSchema +from pipecat.audio.vad.silero import SileroVADAnalyzer +from pipecat.frames.frames import LLMRunFrame +from pipecat.pipeline.pipeline import Pipeline +from pipecat.pipeline.runner import PipelineRunner +from pipecat.pipeline.task import PipelineParams, PipelineTask +from pipecat.processors.aggregators.llm_context import LLMContext +from pipecat.processors.aggregators.llm_response_universal import ( + LLMContextAggregatorPair, + LLMUserAggregatorParams, +) +from pipecat.runner.types import RunnerArguments +from pipecat.runner.utils import create_transport +from pipecat.services.azure.realtime.llm import AzureRealtimeLLMService +from pipecat.services.llm_service import FunctionCallParams +from pipecat.services.openai.realtime.events import ( + AudioConfiguration, + AudioInput, + InputAudioTranscription, + SessionProperties, +) +from pipecat.transports.base_transport import BaseTransport, TransportParams +from pipecat.transports.daily.transport import DailyParams +from pipecat.transports.websocket.fastapi import FastAPIWebsocketParams + +load_dotenv(override=True) + + +async def fetch_weather_from_api(params: FunctionCallParams): + # Simulate a long-running API call so we can demonstrate that the + # conversation continues while the tool is in flight. + await asyncio.sleep(10) + temperature = ( + random.randint(60, 85) + if params.arguments["format"] == "fahrenheit" + else random.randint(15, 30) + ) + await params.result_callback( + { + "conditions": "nice", + "temperature": temperature, + "location": params.arguments["location"], + "format": params.arguments["format"], + "timestamp": datetime.now().strftime("%Y%m%d_%H%M%S"), + } + ) + + +weather_function = FunctionSchema( + name="get_current_weather", + description="Get the current weather", + properties={ + "location": { + "type": "string", + "description": "The city and state, e.g. San Francisco, CA", + }, + "format": { + "type": "string", + "enum": ["celsius", "fahrenheit"], + "description": "The temperature unit to use. Infer this from the users location.", + }, + }, + required=["location", "format"], +) + +tools = ToolsSchema(standard_tools=[weather_function]) + + +system_instruction = ( + "You are a friendly assistant. The user and you will engage in a spoken " + "dialog exchanging the transcripts of a natural real-time conversation. " + "Keep your responses short, generally two or three sentences for chatty " + "scenarios. When the user asks for the weather, call get_current_weather. " + "While you wait for the result, keep chatting with the user. When the " + "result arrives, share it with the user naturally." +) + + +transport_params = { + "daily": lambda: DailyParams( + audio_in_enabled=True, + audio_out_enabled=True, + ), + "twilio": lambda: FastAPIWebsocketParams( + audio_in_enabled=True, + audio_out_enabled=True, + ), + "webrtc": lambda: TransportParams( + audio_in_enabled=True, + audio_out_enabled=True, + ), +} + + +async def run_bot(transport: BaseTransport, runner_args: RunnerArguments): + logger.info(f"Starting bot") + + llm = AzureRealtimeLLMService( + api_key=os.environ["AZURE_REALTIME_API_KEY"], + base_url=os.environ["AZURE_REALTIME_BASE_URL"], + settings=AzureRealtimeLLMService.Settings( + system_instruction=system_instruction, + session_properties=SessionProperties( + audio=AudioConfiguration( + input=AudioInput( + transcription=InputAudioTranscription(model="whisper-1"), + ) + ), + ), + ), + ) + + llm.register_function( + "get_current_weather", + fetch_weather_from_api, + cancel_on_interruption=False, + ) + + context = LLMContext(tools=tools) + user_aggregator, assistant_aggregator = LLMContextAggregatorPair( + context, + user_params=LLMUserAggregatorParams(vad_analyzer=SileroVADAnalyzer()), + ) + + pipeline = Pipeline( + [ + transport.input(), + user_aggregator, + llm, + transport.output(), + assistant_aggregator, + ] + ) + + task = PipelineTask( + pipeline, + params=PipelineParams( + enable_metrics=True, + enable_usage_metrics=True, + ), + idle_timeout_secs=runner_args.pipeline_idle_timeout_secs, + ) + + @transport.event_handler("on_client_connected") + async def on_client_connected(transport, client): + logger.info(f"Client connected") + context.add_message( + {"role": "developer", "content": "Please introduce yourself to the user."} + ) + await task.queue_frames([LLMRunFrame()]) + + @transport.event_handler("on_client_disconnected") + async def on_client_disconnected(transport, client): + logger.info(f"Client disconnected") + await task.cancel() + + runner = PipelineRunner(handle_sigint=runner_args.handle_sigint) + await runner.run(task) + + +async def bot(runner_args: RunnerArguments): + """Main bot entry point compatible with Pipecat Cloud.""" + transport = await create_transport(runner_args, transport_params) + await run_bot(transport, runner_args) + + +if __name__ == "__main__": + from pipecat.runner.run import main + + main() diff --git a/examples/realtime/realtime-grok-async-tool.py b/examples/realtime/realtime-grok-async-tool.py new file mode 100644 index 000000000..c54668fbb --- /dev/null +++ b/examples/realtime/realtime-grok-async-tool.py @@ -0,0 +1,179 @@ +# +# Copyright (c) 2024-2026, Daily +# +# SPDX-License-Identifier: BSD 2-Clause License +# + +"""Example: async function call with the Grok Realtime LLM service. + +The ``get_current_weather`` tool is registered with +``cancel_on_interruption=False`` and simulates a slow API call (10s sleep). +While the call is in flight the conversation continues; the result arrives +later via the async-tool mechanism and is forwarded to Grok Realtime as a +``function_call_output`` so the model can integrate it naturally into its +next turn. +""" + +import asyncio +import os +import random +from datetime import datetime + +from dotenv import load_dotenv +from loguru import logger + +from pipecat.adapters.schemas.function_schema import FunctionSchema +from pipecat.adapters.schemas.tools_schema import ToolsSchema +from pipecat.frames.frames import LLMRunFrame +from pipecat.pipeline.pipeline import Pipeline +from pipecat.pipeline.runner import PipelineRunner +from pipecat.pipeline.task import PipelineParams, PipelineTask +from pipecat.processors.aggregators.llm_context import LLMContext +from pipecat.processors.aggregators.llm_response_universal import LLMContextAggregatorPair +from pipecat.runner.types import RunnerArguments +from pipecat.runner.utils import create_transport +from pipecat.services.llm_service import FunctionCallParams +from pipecat.services.xai.realtime.events import SessionProperties +from pipecat.services.xai.realtime.llm import GrokRealtimeLLMService +from pipecat.transports.base_transport import BaseTransport, TransportParams +from pipecat.transports.daily.transport import DailyParams +from pipecat.transports.websocket.fastapi import FastAPIWebsocketParams + +load_dotenv(override=True) + + +async def fetch_weather_from_api(params: FunctionCallParams): + # Simulate a long-running API call so we can demonstrate that the + # conversation continues while the tool is in flight. + await asyncio.sleep(10) + temperature = ( + random.randint(60, 85) + if params.arguments["format"] == "fahrenheit" + else random.randint(15, 30) + ) + await params.result_callback( + { + "conditions": "nice", + "temperature": temperature, + "location": params.arguments["location"], + "format": params.arguments["format"], + "timestamp": datetime.now().strftime("%Y%m%d_%H%M%S"), + } + ) + + +weather_function = FunctionSchema( + name="get_current_weather", + description="Get the current weather", + properties={ + "location": { + "type": "string", + "description": "The city and state, e.g. San Francisco, CA", + }, + "format": { + "type": "string", + "enum": ["celsius", "fahrenheit"], + "description": "The temperature unit to use. Infer this from the users location.", + }, + }, + required=["location", "format"], +) + +tools = ToolsSchema(standard_tools=[weather_function]) + + +system_instruction = ( + "You are a friendly assistant. The user and you will engage in a spoken " + "dialog exchanging the transcripts of a natural real-time conversation. " + "Keep your responses short, generally two or three sentences for chatty " + "scenarios. When the user asks for the weather, call get_current_weather. " + "While you wait for the result, keep chatting with the user. When the " + "result arrives, share it with the user naturally." +) + + +# Note: Grok has built-in server-side VAD, so we don't need local VAD. +transport_params = { + "daily": lambda: DailyParams( + audio_in_enabled=True, + audio_out_enabled=True, + ), + "twilio": lambda: FastAPIWebsocketParams( + audio_in_enabled=True, + audio_out_enabled=True, + ), + "webrtc": lambda: TransportParams( + audio_in_enabled=True, + audio_out_enabled=True, + ), +} + + +async def run_bot(transport: BaseTransport, runner_args: RunnerArguments): + logger.info(f"Starting bot") + + llm = GrokRealtimeLLMService( + api_key=os.environ["XAI_API_KEY"], + settings=GrokRealtimeLLMService.Settings( + system_instruction=system_instruction, + session_properties=SessionProperties( + voice="Ara", + ), + ), + ) + + llm.register_function( + "get_current_weather", + fetch_weather_from_api, + cancel_on_interruption=False, + ) + + context = LLMContext(tools=tools) + user_aggregator, assistant_aggregator = LLMContextAggregatorPair(context) + + pipeline = Pipeline( + [ + transport.input(), + user_aggregator, + llm, + transport.output(), + assistant_aggregator, + ] + ) + + task = PipelineTask( + pipeline, + params=PipelineParams( + enable_metrics=True, + enable_usage_metrics=True, + ), + idle_timeout_secs=runner_args.pipeline_idle_timeout_secs, + ) + + @transport.event_handler("on_client_connected") + async def on_client_connected(transport, client): + logger.info(f"Client connected") + context.add_message( + {"role": "developer", "content": "Please introduce yourself to the user."} + ) + await task.queue_frames([LLMRunFrame()]) + + @transport.event_handler("on_client_disconnected") + async def on_client_disconnected(transport, client): + logger.info(f"Client disconnected") + await task.cancel() + + runner = PipelineRunner(handle_sigint=runner_args.handle_sigint) + await runner.run(task) + + +async def bot(runner_args: RunnerArguments): + """Main bot entry point compatible with Pipecat Cloud.""" + transport = await create_transport(runner_args, transport_params) + await run_bot(transport, runner_args) + + +if __name__ == "__main__": + from pipecat.runner.run import main + + main() diff --git a/src/pipecat/services/ultravox/llm.py b/src/pipecat/services/ultravox/llm.py index 967dc7886..d11fa6a01 100644 --- a/src/pipecat/services/ultravox/llm.py +++ b/src/pipecat/services/ultravox/llm.py @@ -45,7 +45,8 @@ from pipecat.frames.frames import ( UserAudioRawFrame, VADUserStoppedSpeakingFrame, ) -from pipecat.processors.aggregators.llm_context import LLMContext +from pipecat.processors.aggregators import async_tool_messages +from pipecat.processors.aggregators.llm_context import LLMContext, LLMSpecificMessage from pipecat.processors.frame_processor import FrameDirection from pipecat.services.llm_service import FunctionCallFromLLM, LLMService from pipecat.services.settings import NOT_GIVEN, LLMSettings, _NotGiven, assert_given @@ -218,6 +219,7 @@ class UltravoxRealtimeLLMService(LLMService): self._disconnecting = False self._bot_responding: Literal[None, "text", "voice"] = None self._last_user_id: str | None = None + self._async_tool_warning_logged: bool = False self._sample_rate = 48000 self._resampler = create_stream_resampler() @@ -413,6 +415,33 @@ class UltravoxRealtimeLLMService(LLMService): await self.push_frame(frame, direction) async def _handle_context(self, context: LLMContext): + # If the user registered a function with cancel_on_interruption=False, + # the aggregator emits async-tool-style messages into the context. We + # don't (currently) honor those on Ultravox: the Ultravox API freezes + # the conversation during tool execution + # (https://docs.ultravox.ai/tools/async-tools#custom-tool-timeouts), + # so the "keep talking while the tool runs" intent of the flag is + # structurally not achievable here. Surface a one-time warning so + # users see they're not getting what they expect. + if not self._async_tool_warning_logged: + for message in context.get_messages(): + if isinstance(message, LLMSpecificMessage): + continue + if async_tool_messages.parse_message(message) is not None: + logger.error( + f"{self}: cancel_on_interruption=False is not supported by " + f"Ultravox: the conversation freezes during tool execution, so " + f"the 'keep talking while the tool runs' intent of the flag " + f"would not be achievable anyway. Use " + f"cancel_on_interruption=True (the default) or a non-realtime " + f"LLM service if your tool needs the async semantics." + ) + await self.push_error( + error_msg="cancel_on_interruption=False is not supported by Ultravox.", + ) + self._async_tool_warning_logged = True + break + # Ultravox handles all context server-side, so the only context we may # need to handle here is new function call results. for message in reversed(context.messages): diff --git a/src/pipecat/services/xai/realtime/llm.py b/src/pipecat/services/xai/realtime/llm.py index 79ea421ed..8b8e0d94a 100644 --- a/src/pipecat/services/xai/realtime/llm.py +++ b/src/pipecat/services/xai/realtime/llm.py @@ -47,7 +47,8 @@ from pipecat.frames.frames import ( UserStoppedSpeakingFrame, ) from pipecat.metrics.metrics import LLMTokenUsage -from pipecat.processors.aggregators.llm_context import LLMContext +from pipecat.processors.aggregators import async_tool_messages +from pipecat.processors.aggregators.llm_context import LLMContext, LLMSpecificMessage from pipecat.processors.frame_processor import FrameDirection from pipecat.services.llm_service import FunctionCallFromLLM, LLMService from pipecat.services.settings import ( @@ -913,6 +914,43 @@ class GrokRealtimeLLMService(LLMService[GrokRealtimeLLMAdapter]): sent_new_result = False for message in self._context.get_messages(): + # LLMSpecificMessages are opaque provider-specific payloads, not + # standard tool-result messages — skip them. + if isinstance(message, LLMSpecificMessage): + continue + + # Async-tool messages live alongside regular tool messages in the + # context; detect and route them before the regular logic so we + # don't try to send the async-tool envelope JSON as a tool result. + async_payload = async_tool_messages.parse_message(message) + if async_payload is not None: + if async_payload.tool_call_id in self._completed_tool_calls: + continue + if async_payload.kind == "started": + # The provider already issued the tool call and natively + # awaits a result; nothing to send for the started marker. + continue + if async_payload.kind == "intermediate": + logger.error( + f"{self}: Grok Realtime does not support streamed async " + f"tool results; dropping intermediate result for " + f"tool_call_id={async_payload.tool_call_id}. Use a " + f"non-realtime LLM service if your tool needs to " + f"stream intermediate results." + ) + await self.push_error( + error_msg="Grok Realtime does not support streamed async tool results.", + ) + continue + # kind == "final": deliver via the formal tool-result channel + # — same path as a synchronous tool result, just delayed. + if send_new_results: + sent_new_result = True + await self._send_tool_result(async_payload.tool_call_id, async_payload.result) + self._completed_tool_calls.add(async_payload.tool_call_id) + continue + + # Look for newly-completed "regular" (as opposed to async-tool) results if message.get("role") and message.get("content") != "IN_PROGRESS": tool_call_id = message.get("tool_call_id") if tool_call_id and tool_call_id not in self._completed_tool_calls: @@ -939,6 +977,7 @@ class GrokRealtimeLLMService(LLMService[GrokRealtimeLLMAdapter]): async def _send_tool_result(self, tool_call_id: str, result: str): """Send a tool call result to Grok.""" + logger.debug(f"Sending tool result to Grok Realtime for tool_call_id={tool_call_id}") item = events.ConversationItem( type="function_call_output", call_id=tool_call_id,