Interface AudioTranscriptionInput

AudioTranscriptionInput represents the input for audio transcription.

interface AudioTranscriptionInput {
    diarization?: boolean;
    file: string | File;
    injection?: boolean;
    language?: string;
    model: string;
    pii?: string;
    prompt?: string;
    replaceMethod?: string;
    responseFormat?: string;
    temperature?: number;
    timestampGranularities?: TimestampGranularity | TimestampGranularity[];
    toxicity?: boolean;
}

Properties

diarization?: boolean

diarization represents whether to diarize the audio and return speaker turns.

file: string | File

file represents the audio file to transcribe (can be File object or path string).

injection?: boolean

injection represents whether to check the output for a prompt injection.

language?: string

language represents the language the audio is in.

model: string

model represents the transcription model to use.

pii?: string

pii represents whether to check the output for PII.

prompt?: string

prompt represents an optional text to guide the model's style.

replaceMethod?: string

replaceMethod represents the method to replace any found PII.

responseFormat?: string

responseFormat represents the format for the response object.

temperature?: number

temperature represents the temperature parameter for controlling randomness.

timestampGranularities?: TimestampGranularity | TimestampGranularity[]

timestampGranularities represents timestamp granularities to populate.

toxicity?: boolean

toxicity represents whether to check the output for toxicity.