The Audio API provides two speech to text endpoints, transcriptions andtranslations, based on our state-of-the-art open source large-v2 Whispermodel.