Skip to main content
Transcribe audio files to text using ASR (Automatic Speech Recognition) models.

Endpoint

Basic Usage

Parameters

Supported audio formats: MP3, WAV, M4A, FLAC, OGG, WebM.

Response

The shape follows response_format. Every format is billed the same: per minute of audio. Omitted or verbose_json:
srt:
The OpenAI SDKs return text, srt and vtt as a string. The Runcrate SDKs always return an object; for those three formats its text field holds the transcript or the subtitle file.