Audio

Speech to text

POST /v1/audio/transcriptions
{
  "model": "automatic",
  "file": "<base64 audio>"
}

Text to speech

POST /v1/audio/speech
{
  "model": "automatic",
  "input": "Hello world",
  "voice": "default",
  "response_format": "mp3"
}

Speech-to-speech and realtime voice: see Platform API.