Thanks — I understand you want a Python function to subscribe to audio tracks between remote participants so participants can hear each other (bidirectional audio). LiveKit automatically handles audio forwarding and subscriptions by default, but if you need an explicit function to verify or enforce subscriptions (for example, if auto-subscribe is disabled or you want to re-subscribe), here’s a concise, practical Python example you can use. Notes before the code: • LiveKit auto-subscribes to remote audio by default, so typically you don’t need manual subscription. • Use this function when you need to confirm subscriptions or re-enable them when auto-subscribe is turned off. Example: ensure all remote audio tracks are subscribed (async)

from livekit import rtc

async def ensure_audio_subscription(room: rtc.Room):
    """
    Ensure all remote participants' audio tracks are subscribed.
    Typically not necessary when auto-subscribe is enabled, but useful
    if auto-subscribe is disabled or you want to re-subscribe.
    """
    for participant in room.remote_participants.values():
        for pub in participant.track_publications.values():
            # Only handle published audio tracks
            if pub.track and pub.track.kind == rtc.TrackKind.KIND_AUDIO:
                # If currently not subscribed, request subscription
                if not pub.is_subscribed:
                    try:
                        await pub.set_subscribed(True)
                        print(f"Subscribed to audio track {pub.sid} from {participant.identity}")
                    except Exception as e:
                        print(f"Failed to subscribe to {pub.sid} from {participant.identity}: {e}")

Usage example (hook on participant connected):

@server.room.on("participant_connected")
async def on_participant_connected(participant: rtc.RemoteParticipant):
    # room is available via server.room in this example
    await ensure_audio_subscription(server.room)

If you want to start the client with explicit auto-subscribe settings when connecting:

# When constructing/connecting the client, enable audio auto-subscribe
room = await rtc.connect(url, token, auto_subscribe=rtc.AutoSubscribe.AUDIO_AND_VIDEO)

If this doesn’t solve your case, a few quick clarifying questions to help further: • Are you using auto-subscribe (default) or do you need manual subscription control? • Are you working with the server-side SDK (Agents/framework) or client SDKs running in browsers/native apps? • Any errors or specific behavior you’re seeing when participants join (no audio, one-way audio, track published but not heard)? I can provide adjusted code samples for: • Manual subscribe/unsubscribe flows, • Handling per-participant mute/unmute, • Client-side (browser/native) subscription examples if needed. Which of the above would you like next? Sources: Client SDKs | livekit/livekit_composite | DeepWiki | Capturing the First Few Seconds of a Call for Processing | livekit/livekit_composite | DeepWiki