To print the

STTMetrics

with your logger in LiveKit, you can use one of the following approaches: 1. Using the Built-in

log_metrics

Helper FunctionThe simplest way is to use the built-in

log_metrics

utility function, which automatically formats and logs `STTMetrics`:

from livekit.agents import metrics, MetricsCollectedEvent

@session.on("metrics_collected")
def _on_metrics_collected(ev: MetricsCollectedEvent):
    metrics.log_metrics(ev.metrics)

This will log the message "STT metrics" and include fields like

audio_duration

. You can also pass a custom logger:

metrics.log_metrics(ev.metrics, logger=your_custom_logger)

2. Setting Up Custom STT Metrics Event HandlersFor more control, you can set up event handlers directly on the STT component:

import logging
from livekit.agents.metrics import STTMetrics

logger = logging.getLogger("stt-metrics")
logger.setLevel(<http://logging.INFO|logging.INFO>)

class STTMetricsAgent(Agent):
    def __init__(self):
        super().__init__(
            stt=deepgram.STT(),
            # ... other components
        )
        
        def stt_wrapper(metrics: STTMetrics):
            asyncio.create_task(self.on_stt_metrics_collected(metrics))
            
        self.stt.on("metrics_collected", stt_wrapper)

    async def on_stt_metrics_collected(self, metrics: STTMetrics):
        <http://logger.info|logger.info>(
            f"STT Metrics - Duration: {metrics.duration:.4f}s, "
            f"Audio Duration: {metrics.audio_duration:.4f}s, "
            f"Streamed: {metrics.streamed}, "
            f"Speech ID: {metrics.speech_id}"
        )

Available Fields in `STTMetrics`When logging

STTMetrics

, you can access the following fields: • `type`: The metrics type • `label`: The metrics label • `request_id`: The request identifier • `timestamp`: When the metrics were collected • `duration`: Processing duration in seconds • `speech_id`: The speech identifier • `error`: Any error that occurred • `streamed`: Whether STT is in streaming mode • `audio_duration`: Duration of the audio input in seconds The built-in

log_metrics

function is ideal for standardized logging, while custom event handlers allow for more tailored logging formats. Sources: livekit-agents/livekit/agents/metrics/utils.py | Capturing metrics | LiveKit Docs | metrics/metrics_llm.py