```
## Set up tracing
Import `LangSmithGoogleADKLivePlugin` and register it on your `Runner`. It runs alongside your own `run_live` loop, so your loop only handles audio playback, barge-ins, and UI updates:
```python theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
from google.adk.agents.run_config import RunConfig, StreamingMode
from google.adk.runners import Runner
from google.genai import types as genai_types
from langsmith.integrations.google_adk_live import LangSmithGoogleADKLivePlugin
plugin = LangSmithGoogleADKLivePlugin(project_name="gemini-live-voice")
runner = Runner(
app_name="voice-app",
agent=root_agent,
session_service=session_service,
plugins=[plugin],
)
run_config = RunConfig(
response_modalities=["AUDIO"],
streaming_mode=StreamingMode.BIDI,
input_audio_transcription=genai_types.AudioTranscriptionConfig(),
output_audio_transcription=genai_types.AudioTranscriptionConfig(),
)
async for event in runner.run_live(
user_id=user_id,
session_id=adk_session.id,
live_request_queue=queue,
run_config=run_config,
):
... # play audio, handle barge-in, update the UI
```
Transcription is opt-in. To show transcripts, set both `input_audio_transcription` and `output_audio_transcription` on the `RunConfig`.
On a graceful end (the live request queue closing), ADK sends its `after_run` callback and the plugin finalizes the trace for you.
On a cancelled run, such as a console app that stops `run_live` on Ctrl-C, ADK may not send that callback, so call `plugin.finalize(session_id=adk_session.id)` during teardown. Otherwise, you lose the trace and the audio attachment built at finalize. The call is idempotent, so it does nothing if ADK's callback already ran.
## Group a conversation into a thread
Each conversation is captured as its own trace with its own thread ID. To supply your own ID, for example to group the conversation with related interactions in a LangSmith [thread](/langsmith/threads), pass a `thread_id_provider` to the plugin:
```python theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
plugin = LangSmithGoogleADKLivePlugin(
project_name="gemini-live-voice",
thread_id_provider=lambda: thread_id,
)
```
A single plugin instance is shared across every `run_live` call and it resolves the thread ID once, at the start of each conversation. The default keeps concurrent conversations separate. If you pass a `thread_id_provider` on a server handling concurrent conversations, it returns the ID for the current conversation rather than a fixed value; for example, by reading a `ContextVar` set at the start of each run.
## Record the conversation audio
If you feed your microphone and playback audio to the plugin, it attaches a single stereo recording (user left, agent right) to the trace:
```python theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
plugin.record_user_audio(mic_chunk) # user mic PCM16
plugin.record_agent_audio(played_chunk) # agent PCM16 as played
```
To accurately reflect what is heard, record the user's microphone capture before resampling it for ADK and tap the speaker for the agent's audio. Feed both channels at the same sample rate (the plugin's `sample_rate` is 24 kHz by default). For the underlying attachment API, see [Upload files with traces](/langsmith/upload-files-with-traces).
## Next steps
Core conventions for tracing voice agents.
Attach the conversation audio recording to your trace.
***
[Connect these docs](/use-these-docs) to Claude, VSCode, and more via MCP for real-time answers.
[Edit this page on GitHub](https://github.com/langchain-ai/docs/edit/main/src/langsmith/trace-gemini-live.mdx) or [file an issue](https://github.com/langchain-ai/docs/issues/new/choose).