/audio/transcriptionsTranscribe audio to text
Transcribes audio into text in the input language. Send the request as multipart/form-data with the audio file in the file field. Supply file and model at minimum — language and response_format are optional. Returns a transcription object or a stream of transcript events.
- IdempotentThe SDK sends
Idempotency-Key, so a retried request is only applied once.
Transcription request. file and model are required; all other fields are optional.
file is required and must be an audio file in one of the supported formats: flac, mp3, mp4, mpeg, mpga, m4a, ogg, wav, or webm. model is required and must be one of: whisper-1, gpt-transcribe, gpt-4o-transcribe, gpt-4o-mini-transcribe, gpt-4o-mini-transcribe-2025-12-15, or gpt-4o-transcribe-diarize. temperature must be between 0 and 1. response_format must be one of: json, text, srt, verbose_json, vtt, or diarized_json. A 429 is returned when the request rate limit is exceeded.