@squidcloud/client
    Preparing search index...

    Interface WhisperOptions

    Options for transcribing audio using the Whisper model.

    interface WhisperOptions {
        modelName: "whisper-1";
        prompt?: string;
        responseFormat?: "json" | "text" | "srt" | "verbose_json" | "vtt";
        temperature?: number;
    }

    Hierarchy (View Summary)

    Index

    Properties

    modelName prompt? responseFormat? temperature?

    Properties

    modelName: "whisper-1"

    Specifies the Whisper-1 model for audio transcription.

    prompt?: string

    An optional prompt to guide the transcription process.

    responseFormat?: "json" | "text" | "srt" | "verbose_json" | "vtt"

    The format of the transcription response; defaults to 'json'.

    temperature?: number

    The temperature for sampling during transcription; defaults to model-specific value.

    Settings

    Member Visibility

    On This Page

    Properties