Nodes/OpenKatana/OpenKatana · Transcribe audio
ComfyUI Node

OpenKatana · Transcribe audio

Transcribe speech locally with word timings (faster-whisper; a Whisper model already in the local cache, see get_capabilities transcription; never downloaded). Writes a words file for make_captions; times are seconds of the source file. Machine output: check words, names and timings against the audio before captioning. (project_id: str, source: str, model: str | None = None, language: str | None = None, start: float = 0, duration: float | None = None) -> dict

By rk0neki·Created about a month ago·Updated 16 days ago· 82
OpenKatana · Transcribe audio
    • result_json
    ◄project_idmy-film►
    ◄arguments{}►
    ◄after—►
    CategoryOpenKatana/Tools

    Inputs (3)

    NameTypeDefaultDescription
    project_idSTRINGmy-film—
    argumentsSTRING{}—
    afteroptSTRING—

    Outputs (1)

    NameTypeDescription
    result_jsonSTRING—