OpenAI launched three new audio models in its API: GPT-Realtime-2 (a voice model with GPT-5-class reasoning that can handle harder requests and carry conversations naturally), GPT-Realtime-Translate (a live translation model that translates speech from 70+ input languages into 13 output languages), and GPT-Realtime-Whisper (a streaming speech-to-text model that transcribes speech live). GPT-Realtime-2 includes new features like preambles (“let me check that”), parallel tool calls with audible status updates, an increased context window from 32K to 128K tokens, and adjustable reasoning effort from minimal to xhigh. Early testers include Zillow, Deutsche Telekom, Priceline, Vimeo, Glean, and Intercom.
