videosphere
    Preparing search index...
    • Creates a spoken-lag tracker for one listen language.

      Models the listener's playback queue rather than inspecting it: clips play strictly in order and back to back, so the start time of each clip is whichever comes later — the moment it arrives, or the moment the previous clip finishes. That reproduces what a listener hears without needing any telemetry from their device.

      Parameters

      • Optionaloptions: { maxSourceGapMs?: number }

        Optional override for the source-gap cutoff.

      Returns TtsTimingTracker

      A tracker that records clips and reports aggregates.