CallSpeechEvent

Fields

• idstringrequired
• speakerenum (string)required

  Who was speaking during this segment.

  One ofuseragent

• started_speaking_at_msnumberrequired

  Milliseconds since session start when speech began.

• stopped_speaking_at_msnumberrequirednullable

  Milliseconds since session start when speech ended. Null when the session ended before the turn closed.

• transcriptstringrequirednullable

  Recognized text for this segment, when available.

• was_cutbooleanrequirednullable

  Agent-only: true when the agent's audio was interrupted (real or auto-resumed). Null on user segments.

• is_real_turnbooleanrequired

  False when the segment was flagged as noise/blip rather than a meaningful turn.

• stt_transcription_delay_msnumberrequirednullable

  User-only: time from end-of-speech to final transcript, in ms.

• llm_ttft_msnumberrequirednullable

  Agent-only: LLM time-to-first-token for the reply, in ms.

• tts_ttfb_msnumberrequirednullable

  Agent-only: TTS time-to-first-audio-byte for the reply, in ms.

• llm_winner_modelstringrequirednullable

  Agent-only: the model that produced the reply — the winning leg when the LLM is a race, which is also the model llm_ttft_ms describes.

• llm_system_promptstringrequirednullable

  Agent-only: despite the name, the whole prompt the LLM was given for this turn — system instructions plus conversation history — as JSON.