Type Alias DictationResponse

DictationResponse: {
    audio_duration_ms: number;
    confidence: number;
    final_text: string;
    llm_error?: string | null;
    llm_response?: string | null;
    request_time_ms?: number;
    session_id: string;
    sync_time_ms?: number;
    text: string;
    words: DictationWord[];
}

The result of a dictation request.

Type declaration

  • audio_duration_ms: number

    The total audio duration in milliseconds.

  • confidence: number

    The overall transcript confidence, in the range 0-1.

  • final_text: string

    The text to show the user: llm_response when the LLM pass produced one, text otherwise. Derived by the SDK from those two fields, so reading it is safe whether or not an llm_instruction was sent.

  • Optionalllm_error?: string | null

    Why the LLM pass failed, when it did. null otherwise.

  • Optionalllm_response?: string | null

    The transcript rewritten by the LLM pass llm_instruction asked for. null when no instruction was sent or the pass failed.

  • Optionalrequest_time_ms?: number

    The end-to-end server-side request time in milliseconds: auth, multipart parse, decode, inference, the LLM pass, and serialization.

  • session_id: string

    The server-generated UUID for this request. Record it to correlate a request with support.

  • Optionalsync_time_ms?: number

    The time in milliseconds spent transcribing, excluding the LLM pass.

  • text: string

    The raw transcript text, before any LLM pass.

  • words: DictationWord[]

    Per-word text and confidence.

Settings

Member Visibility