Skip to main content

Documentation Index

Fetch the complete documentation index at: https://arize-ax.mintlify.dev/docs/llms.txt

Use this file to discover all available pages before exploring further.

​
Pipecat Voice & Multimodal Agent Tracing

Pipecat is an open-source Python framework for building real-time voice and multimodal conversational agents. It orchestrates the audio transport, speech-to-text, LLM, and text-to-speech stages of a conversation into a single streaming pipeline. Arize provides first-class support for instrumenting Pipecat pipelines using the openinference-instrumentation-pipecat package. Each pipeline run is captured as a session in Arize with LLM, STT, TTS, and tool spans tied to a conversation_id so you can review the full audio-in to audio-out path of any turn.

​
Quick Start: Pipecat Python Integration

​
Installation & Setup

Install the OpenInference Pipecat instrumentor along with Pipecat and the Arize OTel helper: 
pip install openinference-instrumentation-pipecat pipecat-ai arize-otel
openinference-instrumentation-pipecat >=1.0 requires pipecat-ai >=1.0 and Python >=3.11. Pipecat 1.0 renamed observers, removed LLMMessagesFrame, and dropped Python 3.10 support. If you’re still on pipecat-ai<1.0, pin the instrumentor:
pip install 'openinference-instrumentation-pipecat<=0.1.4' 'pipecat-ai<1.0'

​
Instrumentation Setup

Configure the PipecatInstrumentor and tracer to send traces to Arize: 
# Import open-telemetry dependencies
from arize.otel import register

# Setup OTel via our convenience function
tracer_provider = register(
    space_id = "your-space-id", # in app space settings page
    api_key = "your-api-key", # in app space settings page
    project_name = "your-project-name", # name this to whatever you would like
)

# Import openinference instrumentor to map Pipecat traces to a standard format
from openinference.instrumentation.pipecat import PipecatInstrumentor

# Turn on the instrumentor
PipecatInstrumentor().instrument(tracer_provider=tracer_provider)

​
Example: Basic Pipecat Pipeline Usage

Build your pipeline as usual. Pass a conversation_id to PipelineTask so each conversation is grouped into a session in Arize: 
from pipecat.frames.frames import LLMRunFrame
from pipecat.pipeline.pipeline import Pipeline
from pipecat.pipeline.runner import PipelineRunner
from pipecat.pipeline.task import PipelineTask

# Build your pipeline (transport, STT, LLM, TTS, ...)
pipeline = Pipeline([
    # your stages here
])

# Conversation id groups turns under one session in Arize
task = PipelineTask(
    pipeline,
    conversation_id=conversation_id,
)

@transport.event_handler("on_client_connected")
async def on_client_connected(transport, client):
    await task.queue_frames([LLMRunFrame()])

runner = PipelineRunner(handle_sigint=runner_args.handle_sigint)
await runner.run(task)
Start running your Pipecat application and monitor traces in Arize. For advanced examples, explore the openinference-instrumentation-pipecat examples.

​
What is covered by the Instrumentation

Arize provides comprehensive observability for Pipecat’s real-time voice and multimodal pipelines, automatically tracing:

​
Conversation & Session Tracking

  • Conversation Sessions: All turns sharing a conversation_id grouped into a single session
  • Turn Boundaries: Each user-to-assistant exchange captured as a parent span
  • End-to-End Latency: The full audio-in to audio-out path for every turn

​
Pipeline Stage Spans

  • LLM Calls: Prompts, responses, token counts, and model metadata from your LLM service
  • Speech-to-Text (STT): Input audio transcription with latency
  • Text-to-Speech (TTS): Output audio synthesis with latency
  • Tool / Function Calls: When the LLM service invokes tools, their inputs, outputs, and duration

​
Performance & Reliability Monitoring

  • Stage Latency: Per-stage timing to identify bottlenecks in the audio pipeline
  • Token Usage: Prompt, completion, and total tokens across the conversation
  • Errors: Failures inside any pipeline stage surfaced as span errors