Type alias EOTInferenceMetrics

EOTInferenceMetrics: {
    detectionDelay: number;
    metadata?: MetricsMetadata;
    numRequests: number;
    predictionDuration: number;
    timestamp: number;
    totalDuration: number;
    type: "eot_inference_metrics";
}

Per-prediction telemetry for the audio EOT (end-of-turn) detector. Emitted by transports on each cloud or local prediction so we can track detection latency and inference time per call.

Type declaration

  • detectionDelay: number

    Latest total time from audio-frame creation to prediction receive, in milliseconds.

  • Optional metadata?: MetricsMetadata
  • numRequests: number

    Number of prediction requests served (incremental).

  • predictionDuration: number

    Latest time taken by the model side, in milliseconds.

  • timestamp: number
  • totalDuration: number

    Latest RTT time taken to perform inference, in milliseconds.

  • type: "eot_inference_metrics"

Settings

Member Visibility

Theme