Skip to main content
LiveKit Agents is a framework for building real-time voice and video AI applications. Braintrust traces LiveKit Agents applications using OpenTelemetry to capture voice interactions, agent sessions, and realtime model usage.

​
Setup

pip install "braintrust[otel]" livekit-agents livekit-plugins-openai opentelemetry-sdk
The @braintrust/otel package supports both OpenTelemetry v1 and v2. For the best experience, use the same OpenTelemetry versions that LiveKit Agents uses to avoid compatibility issues.
Configure your environment variables:
.env
BRAINTRUST_API_KEY=your-api-key
BRAINTRUST_PARENT=project_name:livekit-demo
OPENAI_API_KEY=your-openai-api-key

​
Trace with LiveKit Agents

Configure Braintrust’s span processor and set it as LiveKit’s tracer provider.
livekit_agent.py
from braintrust.otel import BraintrustSpanProcessor
from livekit import agents
from livekit.agents import Agent, AgentSession, RoomInputOptions
from livekit.agents.telemetry import set_tracer_provider
from livekit.plugins import noise_cancellation, openai
from opentelemetry.sdk.trace import TracerProvider

def setup_braintrust_telemetry():
    """Setup Braintrust OTEL telemetry for agent monitoring"""
    trace_provider = TracerProvider()
    trace_provider.add_span_processor(BraintrustSpanProcessor())
    set_tracer_provider(trace_provider)

class Assistant(Agent):
    def __init__(self) -> None:
        super().__init__(instructions="You are a helpful voice AI assistant.")

async def entrypoint(ctx: agents.JobContext):
    # Setup telemetry
    setup_braintrust_telemetry()

    # Create agent session with OpenAI realtime model
    session = AgentSession(llm=openai.realtime.RealtimeModel(voice="coral"))

    # Start session with assistant agent
    await session.start(
        room=ctx.room,
        agent=Assistant(),
        room_input_options=RoomInputOptions(
            noise_cancellation=noise_cancellation.BVC(),
        ),
    )

# Run script locally with `python livekit_agent.py console`
if __name__ == "__main__":
    agents.cli.run_app(agents.WorkerOptions(entrypoint_fnc=entrypoint))
The BraintrustSpanProcessor automatically captures all LiveKit agent spans, including:
  • Voice interactions and audio processing
  • OpenAI Realtime API calls
  • Agent session lifecycle events
  • Custom spans you create with tracer.start_as_current_span() (Python) or tracer.startActiveSpan() (TypeScript)
You can add attributes to spans using span.set_attribute() (Python) or span.setAttribute() (TypeScript) to enrich your traces with custom metadata, making it easier to debug and analyze your agent’s behavior in Braintrust.

​
Next steps

Learn more about building with LiveKit Agents and tracing with Braintrust: LiveKit Agents Braintrust Tracing