OpenAI launches realtime voice APIs

- OpenAI released three realtime voice models—GPT‑Realtime‑2, GPT‑Realtime‑Translate and GPT‑Realtime‑Whisper—adding live reasoning and streaming interfaces to the API. - eWeek reports the flagship realtime voice model expanded its context window from 32K to 128K tokens, enabling far longer conversational state for voice sessions. - The new realtime tier also comes with tiered pricing, changing latency, session and cost tradeoffs for any system built around live voice agents. (ghacks.net) (eweek.com)